Hillier's Fundamentals of Motor Vehicle Technology: Powertrain Electronics (Book 2), 5th Edition

Hillier’sFundamentals ofMotor Vehicle Technology

Book 2PowertrainElectronics

[647] Chapter 00.qxp 25/8/06 15:35 Page ii

Hillier’sFundamentals ofMotor Vehicle Technology5th Edition

Book 2

PowertrainElectronics

V.A.W. Hillier, Peter Coombes & David Rogers

[647] Chapter 00.qxp 25/8/06 15:35 Page iii

Text © V. A. W. Hillier 1966, 1972, 1981, 1991, 2006, P. Coombes 2006, D.R. Rogers 2006

The rights of V. A. W. Hillier, P. Coombes and D.R. Rogers to be identified as authorsof this work has been asserted by them in accordance with the Copyright, Designsand Patents Act 1988.

All rights reserved. No part of this publication may be reproduced or transmitted inany form or by any means, electronic or mechanical, including photocopy,recording or any information storage and retrieval system, without permission inwriting from the publisher or under licence from the Copyright Licensing AgencyLimited, of 90 Tottenham Court Road, London W1T 4LP.

Any person who commits any unauthorised act in relation to this publication maybe liable to criminal prosecution and civil claims for damages.

First published in 1966 by:Hutchinson EducationSecond edition 1972Third edition 1981 (ISBN 0 09 143161 1)Reprinted in 1990 (ISBN 0 7487 0317 9) by Stanley Thornes (Publishers) LtdFourth edition 1991

Fifth edition published in 2006 by:Nelson Thornes LtdDelta Place27 Bath RoadCHELTENHAMGL53 7THUnited Kingdom

06 07 08 09 10 / 10 9 8 7 6 5 4 3 2 1

A catalogue record for this book is available from the British Library

ISBN 0 7487 8099 8

Cover photograph: Aston Martin V12 Vanquish by David Kimber/Car and BikePhoto Library

Page make-up by GreenGate Publishing Services, Tonbridge, Kent

Printed and bound in Slovenia by Korotan – Ljubljana Ltd

[647] Chapter 00.qxp 25/8/06 16:41 Page iv

CONTENTS

List of abbreviations viAcknowledgements vii

1 INTRODUCTION TO POWERTRAIN ELECTRONICS

Application of electronics and computers 1‘Electronic systems’ or ‘computer

controlled systems’ 3Electronic control units (ECUs) 6Sensors: a means of providing information 11Examples of different types of sensor 13Obtaining information from analogue

and digital sensor signals 22Actuators: producing movement and

other functions 26Examples of different types of actuators 30ECU/actuator control signals 32

2 ENGINE MANAGEMENT – SPARK IGNITION

Emissions, reliability and durability 37Electronic ignition systems

(early generations) 42Computer controlled ignition systems 61Distributorless and direct ignition

systems 68Spark plugs 73

3 ENGINE MANAGEMENT – PETROL

Introduction to electronic petrol injection systems 77

Petrol injection system examples (multi-point injection) 97

Single-point (throttle body) petrol injection 112

Direct petrol injection 115Emissions and emission control

(petrol engines) 124Engine management (the conclusion) 148Engine system self-diagnosis (on-board

diagnostics) and EOBD 150

4 ENGINE MANAGEMENT – DIESEL INJECTION

Modern diesel fuel systems 163The rotary diesel injection pump 165Cold-start pre-heating systems 172Electronic control of diesel injection

(common rail systems) 174

5 TRANSMISSION

Purpose of the transmission system 186Transmission types 187History of electronic control 188Multiplexing 189Sensors and actuators used in

transmission systems 192Clutch electronic control 201Manual gearbox electronic control 204Torque converter electronic control 210Automatic gearbox transmission

management 212Continuously variable transmission

(CVT) 220Light hybrid powertrain technology

(starter–generator) 226Electronic differential and four-wheel

drive control 229Transmission diagnostics 233Transmission summary 235

Index 237

[647] Chapter 00.qxp 25/8/06 15:35 Page v

4WD four-wheel driveABD automatic brake differentialABS anti-lock braking systemAC alternating currentA/D analogue to digitalASR traction controlATF automatic transmission fluidCAN controller area networkCBW clutch-by-wireCD capacitor dischargeCI compression ignitionCO carbon monoxideCO2 carbon dioxideCPU central processing unitCSC cornering stability controlCTX constantly variable transaxle (Ford) CVT continuously variable transmissionDC direct currentDDC dynamic drift control

DRP dynamisches repelprogramm – German fordynamic control program

DSG direct-shift gearboxEBD electronic brake force distributionECU electronic control unitEDC electronic diesel controlEDL electronic differential lockEEC European Economic Community (now EU)EGR exhaust gas recirculationEOBD European on-board diagnosticsESP electronic stabilisation programmeEU European UnionEUDC European extra-urban driving cycleEVAP evaporative emissionsGT grand touringH2O waterHC hydrocarbon

HCCI homogeneous charge compression ignitionHEGO heated exhaust gas oxygen (Ford)HT high tensionIC internal combustionISG integrated starter–generatorLED light emitting diodeLOS limited operating strategyLSD limited slip differentialMAP manifold absolute pressureMIL malfunction indicator lampMTM mechatronics transmission moduleN2 nitrogenNO nitric oxideNO2 nitrogen dioxideNOx oxides of nitrogenNTC negative temperature coefficientO2 oxygenOBD on-board diagnosticsOHC overhead camPb leadPCU powertrain control unitppm parts per millionPTM Porsche traction managementPWM pulse width modulatedSAE Society of Automotive Engineers (USA)SUV sports utility vehicleRPM revolutions per minute (abbreviated to

rev/min when used with a number)TCS traction control systemTCU transmission control unitTDC top dead centreVBA variable bleed actuatorVE verteiler – German for distributor (VE is used

by Bosch for a type of diesel injection pump)WOT wide open throttle

LIST OF ABBREVIATIONS

[647] Chapter 00.qxp 25/8/06 15:35 Page vi

We should like to thank the following companies forpermission to make use of copyright and other material:

Audi AGBMW (UK) LtdRobert Bosch LtdButterworth-HeinemannHaldex Traction ABHaynes Publishing GroupJaguar Cars LtdLuK GmbH & CoPorsche Cars (GB) LtdSiemens VDO AutomotiveToyota (GB) LtdValeoVolkswagen (UK) Ltd

ACKNOWLEDGEMENTSEvery effort has been made to trace the copyrightholders but if any have been inadvertently overlookedthe publishers will be pleased to make the necessaryarrangement at the first opportunity.

Although many of the drawings are based oncommercial components, they are mainly intended toillustrate principles of motor vehicle technology. For thisreason, and because component design changes sorapidly, no drawing is claimed to be up to date.Students should refer to manufacturers’ publications forthe latest information.

[647] Chapter 00.qxp 25/8/06 15:35 Page vii

[647] Chapter 00.qxp 25/8/06 15:35 Page viii

INTRODUCTION TOPOWERTRAINELECTRONICSC

hapt

er 1

1.1 APPLICATION OF ELECTRONICS AND COMPUTERS

what is covered in this chapter . . .

Application of electronics and computers

‘Electronic systems’ or ‘computer controlled systems’

Electronic control units (ECUs)

Sensors: a means of providing information

Examples of different types of sensor

Obtaining information from analogue and digital sensor signals

Actuators: producing movement and other functions

Examples of different types of actuators

ECU/actuator control signals

1.1.1 The increased use of electronicand computer controlledsystems

Modern motor vehicles are fitted with a wide range ofelectronic and computer controlled systems. This bookdetails most of these systems and explains theiroperation, as well as giving guidance on maintenance,fault finding and diagnosis.

However, it is important to remember thatelectronic or computer control of a system is oftensimply a means of improving the operation or efficiencyof an existing mechanical system. Therefore manymechanical systems are also covered, especially wheretheir function and capability has been improvedthrough the application of electronics and computercontrol. See Hillier’s Fundamentals of Motor VehicleTechnology Book 1 for explanations of the basicmechanical systems that still form a fundamental partof motor vehicle technology.

There are of course many electronic systems that donot influence or control mechanical systems; these pureelectric/electronic systems are also covered.

There are many reasons for the increased use ofelectronic systems. Although vehicle systems differconsiderably in function and capability, they rely on thesame fundamental electrical and electronic principlesthat must be fully understood before a vehicle techniciancan work competently on a modern motor vehicle.

1.1.2 Why use electronics andcomputer control?

Most people who witnessed the cultural andtechnological changes that occurred during the last 30years of the twentieth century would probably regardthe electronics revolution as having had the greatestimpact on their working lives, significantly affecting therest of their lives as well. Although we are primarilyconcerned with the motor vehicle here, electronics havehad a substantial and fundamental impact on the waywe live and particularly on the way we work. Electronicsystems affect almost all aspects of our lives, with thedesign and production of consumer products beingparticularly affected. Domestic goods, entertainmentsystems and children’s toys have all changeddramatically because of electronics. While all of theabove examples are obvious and important, electronicshas also enabled computers to become everydaycommodities for professional and personal use.

Why have electronics had such an impact on ourlives and the things we buy and use? A simple answercould be that they are now much more affordable, butthis alone would not be a complete answer. Theapplication of electronics to so many products hasenabled dramatic improvements in the capability andfunction of almost all such products. A simpleexample is the process of writing a letter, whichprogressed from being hand written to being created

[647] Chapter 01 25/8/06 15:36 Page 1

on a mechanical typewriter. The mechanicaltypewriter was improved by the use of electronics, butthe introduction of the computer allowed businessesand then individuals to produce letters with muchgreater stylistic freedom. The computer allows theuser to correct errors, check spelling, change thelayout and achieve a more professional letter thanwas ever possible with any of the previous methods.This book has been produced using computers, withthe author typing the original text and producingsome of the illustrations on computer. The originaldocuments were then passed electronically (by e-mail)to the production company, which used computers tocreate the final style and prepare the book ready forprinting (the printer also uses computers andelectronics).

Apart from the quality improvements alreadymentioned computers have brought greatly increasedspeed; this book would have taken much longer to writeand produce without the benefit of electronics andcomputers. This is true of virtually everything thatmakes use of electronics. Speed and efficiency areimportant, but improvements in almost every way canbe achieved using electronics and computers.

So if we go back and again ask the question ‘Whyuse electronic control?’ we can perhaps now provide anumber of answers, including improvements in speed,in capability or function and in quality. The fact thatelectronics are now much more affordable andelectronic components considerably smaller than inthe past, facilitates wide use of electronics, resulting inall of those benefits so far discussed and many more.

1.1.3 Why use electronics andcomputer control on the motorvehicle?

Since the late 1960s motor vehicles have been fittedwith an increasing range of electronics and computercontrol. Cost and size reductions are obviouslyimportant because of the production volumes ofvehicles, space considerations and the need to keepdown the price paid by consumers (the people andcompanies that buy the vehicles).

Reducing emissions and improving safetyElectronics and electronic control (or computercontrol) have become increasingly necessary in motorvehicles. For example, without electronic control ofvehicle systems (primarily the engine managementand emission control systems), emissions from enginescould not have been reduced by so much. Legislationhas imposed tighter control on emissions; a balancehas been struck between what is wanted and what canbe achieved. The legislators seek continued reductionsin emissions and the vehicle manufacturers have beenable to achieve tremendous results, but withoutelectronics it would not have been possible to reduceemissions to anywhere close to the current low levels.

Safety is another area where electronics haveenabled improvements. The design of a motor vehicleis very dependent on computers that can analyse dataand then help to incorporate improved safety into thebasic vehicle structure. Safety systems such as anti-lockbrakes (ABS) and airbag systems could not function

2 Introduction to powertrain electronics Fundamentals of Motor Vehicle Technology: Book 2

Figure 1.1 Components used in a typical modern electronic computer controlled vehicle system (engine management system)

[647] Chapter 01 25/8/06 15:36 Page 2

anywhere like as efficiently or reliably without the useof electronics.

Consumer demandOne other important issue is consumer demand orexpectation. Not very long ago, only the most expensivevehicles had electronic or computer controlled luxuries.However, it is now expected that cheaper high volumevehicles will also have electronically controlled systems,including the ABS and airbag systems. In fact ABS is nowstandard on vehicles sold across Europe. Furtherexamples include: air conditioning with electroniccontrol (climate control), electric seat adjustment (oftenusing electronic control), sophisticated in-carentertainment systems (CD and DVD systems, etc.), aswell as driver aids such as satellite navigation ordynamic vehicle control systems. In fact, consumer

expectations for more and more electronically controlledvehicle systems is only matched by the desire of vehiclemanufacturers to sell more and more of these systems tothe consumer. When new or improved systems andfeatures are developed, the vehicle manufacturing andsales industries are only too willing to offer them toconsumers, who then develop an expectation.

Without electronics, almost all of these new safetysystems, the modern emission systems and othersystems would not be affordable, and would certainlynot be as functional or as efficient.

Electronic controls are now used for almost allvehicle systems

Emissions regulations are a key factor in theincreasing use of electronic and computer controlK

ey P

oin

ts

‘Electronic systems’ or ‘computer controlled systems’ 3

1.2 ‘ELECTRONIC SYSTEMS’ OR ‘COMPUTER CONTROLLED SYSTEMS’

Figure 1.2 Simple headlight circuit

Figure 1.3 Simple headlight circuit with a relay

1.2.1 Different levels ofsophistication andfunctionality

Electronic enhancement or computer controlAlthough different people will provide differentdefinitions of electronic systems and computer controlledsystems, it is possible for the purposes of this book toclearly separate the two types of system, as follows.

Electronic systemsAn electronic system uses electronics to improve thesafety, size, cost or efficiency of a system, but theelectronics do not necessarily control the system.

For example the evolution of motor vehicle lightingsystems shows how electronics can be used on a simplesystem. Figure 1.2 shows a headlight circuit that isswitched on by the driver when the light switch isturned to the appropriate position. When the switch isin the correct position, it allows electric current to flowfrom the battery directly to the light bulbs. Thedisadvantage of this type of circuit is that all of thecurrent passes through the light switch and through allof the wiring; the switch and wiring must therefore beof high quality and able to carry the relatively highcurrent (which creates heat).

Figure 1.3 shows the light circuit fitted with a relay.When the driver turns the light switch to the appropriateposition, it allows electric current to pass to the relay,which is then ‘energised’. However, to energise the relayrequires only a very low current; therefore, the switchand the wiring will be subjected to neither high currentnor heat, and can be produced more cheaply. When it isenergised, the relay contacts (or internal switch) areforced to close (owing to the magnetic field created bycurrent flowing through the relay winding), which thenallows a larger electric current to pass from the batterythrough to the light bulbs.If the relay is located close to the light bulbs, the wirecarrying the high current is relatively short, and becausethe longer length of wire between the switch and therelay carries only a low current, it can cost less than thewire required in Figure 1.2. As well as the reduced costof the wiring, the reduced current and heat passingthrough the light switch and much of the wiringprovides a safety benefit, allowing a less expensiveswitch to be used.

Figure 1.4 shows almost the same wiring circuit asFigure 1.3 but the relay has been replaced by anelectronic module. The electronic module performs thesame task as the relay but does not contain any moving

[647] Chapter 01 25/8/06 15:36 Page 3

parts: there are no contacts or internal switch. Themodule can consist of very few simple electroniccomponents (transistors and resistors, etc.), which areinexpensive and reliable.

Note, however, that the module does not control thelighting circuit (as is also the case with the relay); itsimply completes the lighting circuit in response toinput from the driver (when the light switch is turned tothe appropriate position).

Computer controlled systemsA computer controlled system could generally bedefined as a system in which some of the actions orfunctions are automated, as opposed to beingcontrolled by the driver or passenger. Using the simpleexample of the light circuit again, computer controlcould automatically switch on the lights when itbecame dark, such as at night or when the vehiclepasses into a tunnel.

For control to be automated, the computer wouldneed information from a sensor. A light sensor can beused to detect the amount of light and pass an electricalsignal (proportional to the amount of light) to thecomputer. The computer would then respond to theelectrical signal; i.e. if the signal had a specific value orwent above or below a certain value, the computerwould then switch on the lights.

It is possible that a simple version of an automatedlight system could use a sensor that is simply a switch,which provides either an on or off signal to thecomputer. When the light fades to a certain level, theswitch could close, thus completing the light circuit.Figure 1.5 shows a headlight circuit where a light sensorhas been included between the light switch (operatedby the driver) and the electronic module. This iseffectively the same circuit as shown in Figure 1.4, withthe addition of a simple light sensor switch. In thisexample, the sensor simply forms part of the circuitbetween the main switch and the electronic module;therefore if the light switch is in the on position, thelights will be switched on when the natural light fadesbelow the specified level. This type of system would notrepresent a fully computerised system.

However, Figure 1.6 shows a similar circuit wherethe electronic module is replaced by a moresophisticated computer module or electronic control

unit (usually referred to as an ECU). In this example,the light sensor is directly connected to the ECU andprovides a signal that varies with the amount of light,i.e. the voltage generated by the sensor could increaseor decrease as the light reduces. The computer wouldthen effectively make the decision as to when the lightswere switched on.

It is then in fact possible to increase the functionalityof the computer by adding more sensors. For example, arain sensor could be fitted to the vehicle to provideautomatic operation of the windscreen wipers. Thesignal from the rain sensor could then also be passed tothe light system ECU, thus allowing the ECU to switchon the lights when the rain sensor detected rain.

Although the above example is relatively simple, itshows that a modern computer controlled system uses acomputer or ECU to control actions and functions,depending on the information received. Many computercontrolled systems make use of a large number ofsensors passing information to the ECU, which may inturn be controlling more than one action or function.The above examples of headlight circuits represent ECUcontrolled functions, i.e. switching on a light bulb.However, when an ECU controls an action, it usuallydoes so by controlling what is referred to as an actuator.Electric motors and solenoids are typical actuators thatcan be controlled by an ECU; a number of examples willbe covered and explained within this book.


Figure 1.5 Headlight circuit with an electronic module and a lightsensor switch

Figure 1.6 Computer controlled headlight circuit with a lightsensor

Figure 1.4 Simple headlight circuit using an electronic module

[647] Chapter 01 25/8/06 15:36 Page 4

An ECU controlled systemAs shown above, an ECU receives information fromsensors, makes calculations and decisions, and thenoperates an actuator (or provides signals for electroniccomponents such as digital displays).

The essential point to remember is that an ECUcannot achieve its main objective, which is to operatean actuator or electronic component, unless theappropriate signals are received. This is true of all ECUcontrolled vehicle systems, and almost all othercomputers: some form of input signal is required beforea calculation and control process can take place. Even anormal PC (personal computer) used to write a letterrequires inputs from the keyboard and mouse before thewords are displayed on the monitor or before the lettercan be printed or e-mailed.

Figure 1.7 shows the basic principles of almost allECU controlled systems, whereby a sensor producessome form of electrical signal, which is passed to theECU. The ECU uses the information provided by thesignal to make the appropriate calculations, and thenpasses an electric control signal to an actuator or digitalcomponent such as the dashboard display.

Figure 1.8 shows a more complex arrangement foran ECU controlled system. This example would betypical of an early generation fuel injection systemwhere the ECU is controlling a number of actuators andwhere a number of sensors are used to provide therequired information.

Actuators that could be fitted to an enginemanagement system● Fuel injector solenoid (for fuel quantity control).● Idle speed stepper motor (for idle speed control).● Exhaust gas recirculation solenoid valve (part of an

emission control system).● Turbocharger wastegate solenoid valve (controlling

turbocharger boost pressure).● Ignition coil (in this instance, the ECU is in fact

controlling the ignition timing when it switches theignition coil on/off, although strictly speaking theignition coil is not an actuator).

Sensors that could be fitted to an enginemanagement system● Engine coolant temperature sensor.● Air temperature sensor (ambient).● Air temperature sensor (intake system).● MAP (manifold absolute pressure) sensor (an intake

manifold pressure/vacuum sensor for an indicationof engine load).

● Crankshaft position sensor (identifies the crankshaftposition for ignition and fuel injection timing, andalso indicates engine speed).

● Camshaft position sensor (providing additionalinformation for ignition and fuel injection timing).

● Throttle position sensor (indicates the amount ofthrottle opening and the rate at which the throttle isopened or closed).

● Boost pressure sensor (indicates the boost pressurein the intake manifold that has been created by theturbocharger).

● Lambda sensor 1 (indicates the oxygen content inthe exhaust gas passing into the catalytic converter,which enables the ECU to correct the fuel mixture).

● Lambda sensor 2 (indicates the oxygen content inthe exhaust gas leaving the catalytic converter,which helps the ECU assess if the catalytic converteris functioning efficiently).

The ECU controlled system shown in Figure 1.8 is infact typical of a modern engine management system,although this example does not show all of the sensorsand actuators that could be fitted. The example doeshowever illustrate a number of sensors and actuatorsthat can be controlled on a typical vehicle system that isfully computer controlled. The engine managementsystem is a good example of the absence of driver inputto the control of the system (apart from placing a footon the throttle to select the desired speed).

All complex systems can be considered as havinginputs, control and outputs

Sensors usually provide inputs, and actuators arecontrolled by ECU outputsK

ey P

oin

ts

‘Electronic systems’ or ‘computer controlled systems’ 5

Figure 1.7 ECU controlled circuit with a single sensor and singleactuator

Figure 1.8 ECU controlled circuit with multiple sensors andactuators

[647] Chapter 01 25/8/06 15:36 Page 5

1.3.2 Control

Having been designed with the capacity to make a pre-programmed decision, an ECU can then be used tocontrol other components. A simple example is the useof an ECU to switch on an electric heater when thetemperature gets cold. Information from a temperaturesensor would inform the ECU that the temperature wasfalling; it could then switch on an electrical circuit forthe heater.

With a simple version of this system, the ECU couldbe programmed to switch on the heater at apredetermined low temperature, and switch off theheater when the temperature has risen to apredetermined high temperature. Such a system wouldresult in the temperature rising and falling in cycles asthe heater was turned on and off. Note that thetemperature sensor could be a simple switch thatopened or closed at a predetermined temperature,providing an appropriate signal to the ECU.

A more sophisticated system could however bedesigned to maintain the temperature at a moreconstant level. If the ECU was designed so that it couldcontrol the electric current passing to the heater, thiswould enable the heater to provide low or high levels ofheat. The ECU program could include the assessment ofhow quickly or slowly the temperature was falling orrising, so that the ECU could switch on part or fullpower to the heater. If the temperature was fallingrapidly, the ECU could switch on full power to theheater. If the temperature was falling slowly, the ECUwould need only to switch on part power to the heater.In this more sophisticated system, the temperaturesensor would have to indicate the full range oftemperature values to the ECU, i.e. the signal from thesensor would have to change progressively with changein temperature; the ECU could consequently assess therate at which temperature was changing.

With the appropriate information from one or moresensors, the ECU can be programmed to provide theappropriate control over a component (such as theheater). The achievement of better or moresophisticated control of a component inevitably requiresmore sophisticated and complex programming of theECU. However, to achieve the required level ofsophisticated control usually requires a greater amountof more accurate information, i.e. a greater number ofsensors, each of which should provide more accurateinformation.

For example, compare an older fuel injection systemwith a modern engine management system. Because oftighter emission regulations and continuous efforts toimprove economy and performance, the modern enginemanagement system ECU must carry out many moretasks with greater levels of control than older systems.Figure 1.9 identifies some of the components in an early


1.3 ELECTRONIC CONTROL UNITS (ECUS)

See Hillier’s Fundamentals of Motor Vehicle TechnologyBook 3 for more detailed information about theelectronic components used in an ECU.

1.3.1 Decision making processThe electronic control unit is often referred to by manyother names, such as electronic control module, blackbox or simply the computer. However, the mostcommonly used name is the electronic control unit,which is generally abbreviated to ECU.

Although the ECU can provide a number of functionsand perform a number of tasks, it is primarily the ‘brain’of the system because it effectively makes decisions. Inreality, however, an ECU makes decisions based oninformation received (from sensors) and then performsa predetermined task (which has been programmed intothe ECU). Whereas a human brain is capable of ‘freethinking’, an ECU is very much restricted in its decisionmaking process because it can only make decisions thatit has been programmed to make.

To compare free thinking with programmed decisionmaking, imagine a car driver approaching a set of trafficlights when the green ‘go’ light is replaced by the amber‘caution or slow down’ light. The driver can make adecision either to slow down, or to accelerate and getacross the lights before the red ‘stop’ light isilluminated. This decision is based on an assessment ofthe conditions; different drivers will make differentdecisions, and in fact one driver could make differentdecisions on different occasions even if the conditionswere identical. To make a similar decision as to whetherto slow down or accelerate, an ECU would also assessconditions such as vehicle speed and distance to thetraffic lights, as well as road conditions (wet, icy, etc.).The ECU would then make the decision based on theprogramming. If the conditions (information) were thesame on every occasion, the ECU would always makethe same decision because the programming dictatesthe decision (not free thinking).

In reality, ECUs and computers in general areprogressively becoming more sophisticated, and theirprogramming is becoming increasingly complex. ECUscan adapt to changing conditions and can ‘learn’, whichallows alternative decisions to be made if the originaldecision does not have the desired effect. A human canmake a decision based on knowledge or information; ifthe first decision does not then produce the desiredresult, an alternative decision can be made because thehuman brain possesses the ability of free thinking.Modern ECUs do have a similar capability but it is aprogrammed one, designed by humans.

The decision making capability of an ECU istherefore dependent on the volume and accuracy ofinformation it receives, and the level of sophistication ofthe programming.

[647] Chapter 01 25/8/06 15:36 Page 6

type of computer controlled fuelling system, which hasrelatively few sensors and relatively few actuators, sothat the ECU has only a small number of tasks orcontrol functions to perform.

Figure 1.10 lists the components from a modernengine management system where the ECU has a muchlarger range of tasks to perform. The number of sensorsand actuators is therefore much greater than on earliersystems. On the modern system, the ECU controls amuch larger number of other components, and in facthas some control over other systems such as the airconditioning system (the engine management ECU caninfluence the operation of the air conditioningcompressor, so that the compressor, which is driven bythe engine, is switched off when full engine power isrequired).

Main casingAn ECU (Figure 1.11) is, amongst other things, acomputer. Readers who use PCs or laptops will knowthat they produce a considerable amount of heat. Inmany cases an electric fan is used to move cooling airaround the PC or laptop to remove some of the heat. Themore powerful the computer, the more heat it produces.An ECU is a powerful computer, and therefore producesheat that must be removed or dissipated. Although somevery early ECUs were located on the vehicle so that acooling fan could help remove some of the heat, ideallythey need to be located where they are unlikely to beexposed to moisture, as well as being isolated fromvibrations and kept away from engine heat. In general,therefore, although not always, ECUs are located withinthe passenger compartment. The ECU main casing isusually an alloy casting which, because it can be boltedto the vehicle bodywork, should help to dissipate heat.

MicroprocessorAs previously mentioned, a computer is regarded as thebrain of a controlled system; the ECU contains one ormore microprocessors which are the main decisionmaking components. As with a normal PC or laptop, themicroprocessor receives information to enable it tomake calculations (effectively the decisions). Themicroprocessor then provides an appropriate outputsignal, which is used to control an actuator or influenceanother system (usually by communicating withanother ECU). Figure 1.12 shows the essential functionswithin the ECU and the essential tasks of themicroprocessor.

If we refer back to the example of the ECUcontrolling a heater (section 1.3.2), the decisions as towhen to switch on the heater, and whether part or fullpower should be used for the heater, are calculated ordecided by the microprocessor.

Amplifier (output or driver stage)Microprocessors operate using very weak signals, i.e.low voltage and current, so would not be directly

Electronic control units (ECUs) 7

Figure 1.9 Earlier generation ECU controlled fuel injection system

Figure 1.10 Modern engine management system. The systemhas a large number of sensors and actuators and the ECUtherefore has a large number of tasks and control functions toperform including influence of other systems

Figure 1.11 Modern ECU and components

1.3.3 ECU components andconstruction

Hillier’s Fundamentals of Motor Vehicle Technology Book3 provides a detailed explanation of the componentsand operations of ECUs, but a brief explanation isrequired at this stage to enable the reader to appreciatethe complexity of the ECU.

[647] Chapter 01 25/8/06 15:36 Page 7

connected to the heater (section 1.3.2), which usesmuch higher voltages and currents. The same applies toan ECU that is controlling a vehicle system; mostvehicle systems operate on 12 volts with relatively highcurrents, which are much higher than the voltages usedwithin microprocessors. To overcome the problem, theoutput or control signal from the microprocessor willusually be passed to some form of amplifier. Theamplifier receives the control signal from themicroprocessor and then provides an amplified orstronger signal to the actuator.

The final part of the amplifier system is oftenreferred to as the output, power or driver stage. Thedriver stage amplifier often contains a power transistor,which may be seen mounted on the outside of the ECUcasing to help with heat dissipation. A simple powertransistor can be regarded as a switch that will switch ahigh power circuit on or off when an appropriate signalis received from a low power circuit. Therefore, if thetransistor is connected into the 12 volt circuit for anactuator, or, for this example, a light bulb, it will switchthe light bulb on or off when the appropriate lowvoltage signal is received from the microprocessor. Thesignal from the microprocessor could be a simple on oroff signal: the power transistor would then switch the12 volt circuit on or off.

Figure 1.13 shows a simple circuit where a light bulbis switched on or off using a power transistor. Note thatthe transistor is switching the earth or return part of the12 volt circuit. The transistor receives a signal from themicroprocessor and effectively emulates or copies thesignal onto the 12 volt circuit.

There are a number of ways in which a powertransistor can switch or affect a higher power circuit.Although a simple on or off function is commonly used,a transistor can emulate or copy a progressively

changing input signal. Therefore, if the signal passinginto the transistor progressively rises and falls instrength, the transistor can progressively increase anddecrease the current flow passing through the highpower circuit.

High speed switching of circuitsThe ECU on a modern vehicle system is often taskedwith switching a circuit on and off at very high speedand frequency, such as when an ignition coil or fuelinjector is switched on and off (which could occur asoften as 100 times a second on an engine operating athigh revolutions per minute). Therefore the decisionmaking process in the microprocessor would producean output signal that switches on and off at thisfrequency, and the power transistor would also switchon and off the 12 volt or power circuit at the samefrequency.

MemoryComputers, including ECUs, have a memory which isstored in a memory microchip. There are differenttypes of memory, but all of them essentially store adescription of the tasks that the ECU must perform.When the microprocessor is making calculations, itwill refer to the memory or ‘talk’ to the memory toestablish what task should be performed when certainitems of information are received. As an example, ifwe again refer to the computer controlled heatersystem covered in section 1.3.2, the informationreceived by the microprocessor could indicate a lowtemperature; the microprocessor would then refer tothe memory to find out what task to perform. Thememory would indicate that the task is to switch onthe heater.

The memory contains all of the necessary operatingdetails applicable to the system being controlled by theECU. For example, if the ECU is controlling a fuelinjection system, all the information about the fuellingrequirements are contained within the memory.Therefore, if the information passed to themicroprocessor includes engine speed, enginetemperature, throttle position, etc., the microprocessor


Figure 1.12 Signal processing in the ECU

Figure 1.13 Power transistor functioning as a switch in a lightcircuit

[647] Chapter 01 25/8/06 15:36 Page 8

refers to the memory to find out how long an injectorshould be switched on for (how long the injector shouldremain open so that the correct quantity of fuel can bedelivered). These operating details are placed or‘programmed’ into the memory either at the time ofECU manufacture or at a later time using dedicatedequipment (in both cases, this is referred to as thesoftware program). In many cases, it is possible toreprogram the memory using modified software, whichcan be useful if it is found that the original program hasa minor fault, such as causing a hesitation when thevehicle is under acceleration.

In the memory systems discussed so far, once thememory chip has been programmed with the operatingdetails, this program remains permanently in thememory chip. However, there are situations where thememory details change. A simple example is when amemory chip might receive information relating to thenumber of miles or kilometres that the vehicle hastravelled; this information could be used to calculatefuel consumption. However, when the driver resets thememory, the information is then erased, i.e. it is notpermanent. The memory chips that store this type ofinformation can lose it when the power is switched off,so it is often necessary to provide a back-up powersupply using a small battery (usually contained withinthe ECU) to prevent loss of data. Note that some ECUshave a permanent power supply from the vehiclebattery (even when the ignition is switched off). Inthese cases, the memory will be retained as long as thevehicle battery is not disconnected.

Analogue and digital signalsAn analogue signal can be regarded as a signal orindicator that continuously changes from one value toanother. A good example is a speedometer using aneedle to sweep around the gauge with changes inspeed: the visual display is an analogue type display,which shows progressive change.

A signal that relies on a change in voltage can alsobe analogue. An example is the change in voltage thatoccurs when a simple lighting dimmer control is alteredfrom the ‘dark’ to the ‘bright’ position. If a voltmeterwere connected to the output terminal of the dimmercontrol (which is usually a variable resistor), the voltagewould be seen to progressively increase and decreasewhen the control was altered.

A voltage signal produced by many sensors can bean analogue signal. An example is a throttle positionsensor, which uses a variable resistor in much the sameway as the light dimmer switch: when the throttle isopened or closed, the voltage progressively increases ordecreases (Figure 1.14).

Although earlier electronic systems relied onanalogue signals and in fact the electronics wereanalogue based, modern computers and electronicsystems are generally digital systems.

A digital signal provides a stepped or pulsed signal.A digital display can be used on a speedometer to

display speed in steps. These steps could be inincrements of 5 km/h or 5 mile/h. In such a case thedriver would only see the display change when thespeed increased by 5 km/h or 5 mile/h. Digitalelectronic signals are also structured in steps, whichgenerally consist of electrical pulses.

ECUs on modern vehicles operate using digitalelectronics. However, in basic terms, the digital processconsists of on and off pulses. In effect there are only twomain conditions that the ECU works with: the on andoff parts of the digital signal.

Signals that are either passing into, passing out of orpassing within an ECU should ideally also be digitalsignals. These on and off pulses can then be counted bythe ECU (counting either the on parts or the off parts ofthe signal). Alternatively the on and off pulses can beused as a reference by the ECU, which could result in theECU performing a predefined task. The ECU does in factexamine the digital signal in a number of ways, whichallows the ECU to extract different information from thesignal such as speed or frequency (Figure 1.15a).

In reality, when a digital signal is being used as aninformation signal passing into the ECU, it does notnecessarily have to be exactly on or off. An examplecould be a light switch in a 12 volt circuit, which wouldproduce an on signal of 12 volts and an off signal ofzero volts. However, the ECU could be programmed toaccept any voltage above 9 volts as being on, and anyvoltage below 3 volts as being off. Therefore, if thesignal voltage from a sensor progressively changesbetween zero volts and 12 volts (an analogue signal),the ECU could still respond to the same programmedvoltage thresholds of 9 volts as an upper limit and 3volts as a lower limit (Figure 1.15b). We should nottherefore always refer to a digital signal as being fullyon or off, but regard it as having upper or lowerthresholds, which can be monitored by the ECU asreference points.

Electronic control units (ECUs) 9

Figure 1.14 Analogue voltage signal produced by a throttleposition sensor

[647] Chapter 01 25/8/06 15:36 Page 9

Analogue to digital convertersBecause ECUs ideally require a digital signal, some formof conversion is necessary to change the analoguesignal from a sensor into a digital signal.

An example could be a temperature sensor, which isused as a means to switch on a cooling fan. The ECUcould switch on the cooling fan when the sensor signalvoltage reaches the 9 volt threshold, but the ECU wouldnot switch off the fan until the sensor voltage fell to the3 volt threshold (Figure 1.15b). The ECU wouldtherefore ideally require a modified signal that onlyidentified or ‘locked on’ to the 9 volt and the 3 voltthresholds. In effect, this modification process takesplace within the ECU: an analogue signal is passed tothe ECU, which contains a converter that converts theanalogue signal into a digital signal. Because manysensors produce analogue signals that need to beconverted to digital signals to enable the

microprocessor to function, a device known as ananalogue to digital converter (A/D converter) is used.

Figure 1.15c shows the principle of an A/Dconverter and an indication of a typical analogue signaland a digital signal. Refer to Hillier’s Fundamentals ofMotor Vehicle Technology Book 3 for more informationon analogue and digital signals as well as on A/Dconverters.

Note that an ECU can also contain converters thatchange digital signals into analogue signals. This mightbe necessary if the actuator operates using an analoguesignal. A simple example is a fuel gauge, which mayrequire an analogue signal to enable the gauge needleto indicate the fuel level. Although the microprocessoris accurately creating the applicable digital signal, itwould need to be converted to some form of analoguesignal to operate the gauge. In reality, more and moreactuators are using digital signals.


Figure 1.15 Analogue and digital signalsa Digital signal, where the pulses could be used to providespeed or frequency informationb Analogue signal where the ECU locks on to the 3 volt and9 volt thresholds reference pointsc Principle of analogue to digital converters

[647] Chapter 01 25/8/06 15:36 Page 10

The complete ECUA fully operational modern ECU will contain thosecomponents detailed above. Although many otherelectronic components are required to make an ECUoperate, those discussed so far are the main functionalcomponents.

In conclusion therefore, the ECU receivesinformation from sensors (the information might beeither digital or analogue). The digital informationpasses directly to the microprocessor, but the analogueinformation must be converted to a digital signal beforebeing passed to the microprocessor. The microprocessorthen assesses the information, refers to the programmedmemory to find out what tasks to perform, makes theappropriate calculations and passes an appropriatecontrol signal to the relevant actuator (or providessignals for an electronic component such as a digitaldisplay). Where the actuator is operated using highervoltages and currents (such as a fuel injector), the weakdigital signal from the microprocessor will need to beamplified using a power transistor or final stage.

The essential point to remember is that an ECUcannot achieve its main objective, which is to operatean actuator or electronic component, unless theappropriate signals are received. This is true of all ECUcontrolled vehicle systems and almost all other

computers: some form of input signal is required beforea calculation and control process can take place.

Note: Understanding of the ECU and an ECU controlledsystem enables a technician to perform diagnosticprocesses much more easily. If the function of eachsensor and each actuator is understood, a relativelyquick diagnosis can be carried out. Although specialisedtest equipment can be used, knowledge of the systemoperation greatly improves the ability to perform quickand accurate diagnosis.

Hillier’s Fundamentals of Motor Vehicle TechnologyBook 3 provides an in-depth examination of theoperation and construction of some sensors andactuators. In other chapters details of specific sensorsand actuators are dealt with in relation to specificsystems. However, the following two sections provide ageneral understanding of sensors and actuatorscommonly used on vehicle systems.

ECUs contain one or more microprocessors thatcarry out calculations and follow lists ofinstructions

ECUs contain A/D converters that act on sensorinputs, and D/A converters, as well as drivercircuits to control outputs

Key

Poin

ts

Sensors: a means of providing information 11

1.4 SENSORS: A MEANS OF PROVIDING INFORMATION

1.4.1 Sensor applicationsIt has previously been explained that an ECU controlledsystem requires information to enable the ECU to makethe appropriate calculations and decisions, which thenin turn enables it to control actuators or electronicdevices. The greater the amount of information that canbe supplied to the ECU, the greater the controlcapability and number of different control functions.

An ECU controlling an earlier generation ofelectronic fuel injection system may have required onlyfour or five sensors to provide the required informationto it. This is because the ECU would only have beenrequired to control the fuel injectors and therefore onlylimited amounts of information were necessary.However, later systems that also included control of theignition system, idle speed and emissions devices (thusforming an engine management system) would have asmany as 20 sensors, or more in some cases. As well ascontrolling more systems, modern ECUs require moreaccurate information from the sensors in order to meetstricter emissions legislation. Sensors have thereforebecome more sophisticated as well as increasing innumber.

Whatever a sensor might be required to measure, e.g.temperature or movement, it must be able to provide asignal to the ECU that can be interpreted by the ECU.

Although the different types of electrical signal arecovered later in this section, an example of change in theelectrical signal would occur when temperature changeswhich, for most temperature sensors, results in anincrease or decrease in the signal voltage passed fromthe sensor to the ECU.

Figure 1.16 indicates the more common examples ofparameters that sensors must detect or measure onmodern vehicle systems. Many other sensorapplications are not included in the chart, but it doesprovide a good indication of the types of informationand the types of applications for many sensors.

From Figure 1.16, it is possible to appreciate thatsensors perform a wide variety of measurement tasks.The parameters most commonly measured are:

● temperature (of fluids or exhaust gas)● movement (angular and linear), including

rotational sensing such as crankshaft speed● position (angular and linear), primarily for partial

rotation of components or partial linear movementbut also including exact angular position ofrotational sensors, e.g. the angle of rotation of acrankshaft at a given time

● pressure/vacuum● oxygen, using a specific type of sensor used to

measure the oxygen content in the exhaust gas.

[647] Chapter 01 25/8/06 15:36 Page 11

Mechanical and electronic sensing devicesAlthough some sensors use a combination ofmechanical and electrical components, which respondtogether to movement, position or pressure (andoccasionally temperature), wherever possible mostmodern sensors only use electronic/electricalcomponents. A typical example is a pressure sensor,which in the past used an aneroid capsule thatdeformed when the pressure changed (Figure 1.17).The deformation of the capsule caused a rod to move;the rod could be connected to a variable resistor whichaltered the voltage in the sensor’s electrical circuit.Later types of pressure/vacuum sensor use an electroniccomponent with no moving parts. Exposure to pressureor vacuum causes the resistance of the component tochange; this change in resistance then alters the voltagein the signal circuit (see Figure 1.17).


Figure 1.16 Sensors and sensor applications

Measurement task Common applications Additional applications

Engine coolant temperature Fuel/ignition/engine management/emission control Cooling fan, driver information displayAir flow (engine load sensing) Fuel/ignition/engine management/emission controlAir mass (engine load sensing) Fuel/ignition/engine management/emission controlAmbient air temperature Fuel/ignition/engine management/emission control Driver information display/air

conditioningIntake air temperature Fuel/ignition/engine management/emission controlEngine oil temperature Fuel/ignition/engine management/emission controlThrottle position Fuel/ignition/engine management/emission control Automatic transmission/anti-wheel

spin/other vehicle stability control/air conditioning

Engine speed Fuel/ignition/engine management/emission control Automatic transmission/anti-wheel spin/other vehicle stability control

Engine intake vacuum/pressure Fuel/ignition/engine management/emission control Automatic transmission(engine load sensing)Crankshaft angle position sensor Fuel/ignition/engine management/emission controlCamshaft angle position sensor Fuel/ignition/engine management/emission controlFuel pressure Fuel/ignition/engine management/emission controlFuel tank pressure Fuel/ignition/engine management/emission controlBoost pressure Turbocharger/supercharger Note: Information from other engine

management sensors will also be usedfor controlling turbo or superchargers

Oxygen (oxygen content of Fuel/ignition/engine management/emission controlexhaust gas)Exhaust gas temperature Fuel/ignition/engine management/emission controlPosition sensor for exhaust Fuel/ignition/engine management/emission controlgas recirculation valveWheel speed (vehicle speed) Anti-lock brakes/vehicle stability control Driver information (vehicle speed)/

automatic transmission/airbagBrake pedal position (on or off) Anti-lock brakes/vehicle stability controlAcceleration/deceleration sensing Anti-lock brakes/vehicle stability control Airbag/other safety systems(sideways movement as well as forward and backward movement)Steering angle Vehicle stability control Power steering

Figure 1.17 Two types of pressure sensora Capsule type pressure sensor using mechanical componentsb Electronic type pressure sensor

[647] Chapter 01 25/8/06 15:36 Page 12

Note: The explanations contained within this sectioncover a number of commonly used sensors withexamples of the types of signal they produce. Althoughother types of sensor are used in automotiveapplications, they will generally be adaptations of thosecovered below. However, other sensors are covered inapplicable sections within this book. For those readerswishing to have more detailed explanations of theelectrical and electronic background to these sensors,see Hillier’s Fundamentals of Motor Vehicle TechnologyBook 3, which provides advanced studies on electricaland electronic theory.

1.5.1 Temperature sensorsTemperature sensors (Figure 1.18a) are used in a widevariety of applications, especially in engine controlsystems, i.e. ignition, fuel and engine management.Additional applications include air conditioning systems,automatic transmissions and any system wheretemperature control or temperature measurement iscritical to the system operation.

Temperature sensors are manufactured using aresistance as the main component. The value of thisresistance changes with temperature. This type ofresistor is called a thermistor: the term is anamalgamation of therm (as in thermometer) andresistor. Because the sensor resistance forms part of anelectrical ‘series resistance’ circuit (other resistances arecontained within the ECU), when the temperature andtherefore the resistance changes, the voltage andcurrent in the circuit also change. The ECU, which of

course forms part of the circuit (Figure 1.18b) andsupplies the reference voltage, will now have a signalvoltage that changes with temperature.

As with almost all modern ECU controlled systems,a reference or starting voltage is applied to the sensorcircuit. This reference voltage originates at the ECU,which reduces the traditional 12 volt vehicle supply to astabilised or regulated voltage, typically around 5 volts.Note however that, because this circuit is used only toprovide a low power signal (and not to operate anactuator such as an electric motor), current flow in thecircuit is very low. The current flow passes from theECU, through the temperature sensor resistance andthen returns to the ECU. Because the circuit is a seriesresistance circuit, when the sensor resistance changesthe current in the circuit also changes, thus providingthe required temperature related signal.

There are generally two main types of resistancebased temperature sensors:

● With the first type, the resistance within the sensordecreases when the temperature increases. This typeis referred to as having a ‘negative temperaturecoefficient’ (NTC).

● With the second type, the resistance increases whenthe temperature increases. This type is referred to ashaving a positive temperature coefficient (PTC).

Temperature sensor analogue signalWith very few exceptions, temperature sensors producean analogue signal. The exceptions are sensors using aswitch, or contacts which close or open at specifiedtemperatures. In these cases the signal will be either onor off.

Examples of different types of sensor 13

1.5 EXAMPLES OF DIFFERENT TYPES OF SENSOR

Figure 1.18 Temperature sensora Typical appearance. The example shown is a coolanttemperature sensor from an engine management systemb Wiring for a temperature sensor

Figure 1.19 Analogue signal voltage for a typical temperaturesensor circuit. Note the progressive change in voltage as thetemperature rises and falls

[647] Chapter 01 25/8/06 15:36 Page 13

The analogue signal voltage produced by sensors with athermistor progressively increases or decreases withchanges in temperature. Because it is common practiceto use NTC sensors, where the resistance reduces as thetemperature increases, the signal voltage will generallyalso reduce as the temperature increases. The typicalsignal voltage from a temperature sensor circuit rangesfrom approximately 4.5 volts when the temperature islow, down to approximately 0.5 volts when thetemperature is high. More specific values are quoted inChapter 3, which describes how these sensors are usedin a fuel injection system.

Figure 1.19 shows the typical analogue outputsignal voltage from a temperature sensor circuit whentemperature changes occur. Note that because thesignal is analogue, the change in voltage is progressive.

1.5.2 Rotational speed sensorsVariable reluctance typeRotational speed sensors are used to detect speed orrevolutions per minute (rev/min) of a component; twocommon examples are an engine crankshaft and a roadwheel. In both cases, the rotational speed information isrequired to enable the ECU to perform its calculations.For an engine system, the crankshaft speed informationis used for the calculation of fuel and ignitionrequirements, as well as for emission control. The wheelspeed information is used to enable calculations foranti-lock braking, wheel spin control and other vehiclestability systems. The wheel speed information can ofcourse also be used to calculate road speed or distancetravelled; this information is then displayed to thedriver or can be used to calculate fuel consumption andother information.

In most cases, rotational speed sensors work on asimple principle, similar to that of an electrical

generator: when a magnetic field is moved through acoil of wire it generates an electric current. Therotational speed sensor uses an adaptation of thisprinciple, which relies on altering the strength of themagnetic field (or magnetic flux). This is achieved bypassing a ferrous metal object (iron or steel) close to orthrough the magnetic field. The strength of themagnetic field or flux increases or decreases when themetal object is moved close to or away from themagnetic field; this change in magnetic flux causes asmall current to be generated or induced within the coilof wire. These sensors are often referred to as inductiveor magnetic variable reluctance sensors.

Rotational speed sensors are often constructed witha permanent magnet located inside or adjacent to a coilof wire. When a metal component (reluctor) passes closeto the sensor, the magnetic field or flux is altered.However, the reluctor often takes the form of a disc,which has one or more ‘teeth’, each of which acts as areluctor. Therefore, as each tooth passes the sensor, itcauses an electric current to be produced within the coilof wire.

As shown in Figure 1.20a, a crankshaft speed sensorcan be located adjacent to the front or back of thecrankshaft, and a disc with one or more teeth, mountedon the crankshaft, can be used as the reluctor disc. Forwheel speed sensors, a similar arrangement is used, butthe reluctor disc is located on the rotating portion of thewheel hub, and the sensor is mounted so that it is closeto the reluctor disc. Figure 1.20b shows a similar sensorused to measure wheel speed rotation (ABS wheelspeed sensor); note that the reluctor has a large numberof reference points.

Rotational speed sensor analogue signalWhen each reluctor tooth passes the sensor, the changein the magnetic field or flux produces a small lowvoltage electric current. As each tooth passes the sensor,


Figure 1.20 Typical arrangement for simple rotational speed sensor.a Crankshaft speed sensor with a number of reluctor teeth(reference points)b Wheel speed sensor

[647] Chapter 01 25/8/06 15:36 Page 14

the voltage increases and decreases, resulting in acontinuously changing voltage as the crankshaft orwheel rotates. In fact, the current flow oscillates oneway and then the other within the circuit, and thevoltage oscillates from positive to negative. The voltageincrease and decrease is shown in Figure 1.21; note thatthe highest voltage is produced when the reluctor toothis approaching the pole of the sensor magnet, and thelowest voltage is produced when the reluctor tooth isleaving the magnet pole. If there is no movement of thereluctor tooth, there will be no current or voltageproduced, irrespective of the position of the reluctortooth. The signal voltage progressively increases anddecreases with the rotation, so the signal is in analogueform. The ECU, which has an inbuilt timer or clock, istherefore able to count the number of pulses over agiven time, and thus calculate the speed of rotation.

It should be noted that there are variations in theway in which some rotational position sensors operate.Some sensors use an ‘exciter coil’ which has a smallvoltage applied to it, allowing a stronger signal to beproduced. Other types use a Hall effect system toproduce a signal. Both of these types of sensor arediscussed in Hillier’s Fundamentals of Motor VehicleTechnology Book 3.

Rotational angular position sensorIn some cases, it is beneficial to be able to calculate orassess the position of a rotating component such as acrankshaft. If there is a means by which the ECU candetermine the position of the crankshaft during itsrotation, it is possible to control accurately the timingof ignition and fuelling. By adapting the previouslydescribed rotational speed sensor system, it is in factrelatively easy to provide an angular position reference.

If, for example, the crankshaft reluctor disc has onlyone reluctor tooth, this tooth could be the reference tocrankshaft angle and could therefore indicate topdead centre (TDC) for piston number 1. In fact, thissingle tooth could also provide the speed reference aswell, although the signal will only be produced oncefor every crankshaft rotation. It is, however, commonpractice to provide a number of teeth around thereluctor disc (60 teeth is not uncommon), and foreach tooth to represent a particular angle ofcrankshaft rotation. If there were 60 reluctor teeth,each tooth would represent 6º of crankshaft anglerotation. However, to establish a master reference ormaster position point, it is normal practice either tomiss out one tooth or make one tooth a substantiallydifferent shape from the other teeth (Figure 1.22).Whichever method was used, the signal from thesensor would contain one voltage change that wasdifferent from the rest of the signal, and thereforeprovide a master reference point such as TDC fornumber 1 piston.

With a possible 60 reference points (or more in somecases), the ECU is now able to calculate crankshaftspeed and the rotational position of the crankshaft veryaccurately. In fact, the ECU can assess any increase ordecrease in crankshaft speed as each tooth passes thesensor. Assuming there were 60 teeth or referencepoints on the crankshaft reluctor disc, this would enablethe ECU to assess the change in crankshaft speed atevery 6º of crankshaft rotation. Control of ignitiontiming, fuelling and emissions would therefore be farmore accurate than if only one reluctor reference toothwere used.


Figure 1.21 Analogue signal produced by a rotational speedsensor. Note that the voltage progressively increases anddecreases as the reluctor tooth approaches and leaves the poleof the sensor magnet

Figure 1.22 Variable reluctance crankshaft position/speed sensorwith master reference pointa Crankshaft reluctor disc with a master position reference point(missing tooth)b Note the different shape of the signal created by the missingtooth

[647] Chapter 01 25/8/06 15:36 Page 15

Many engine management systems have a positionsensor, which indicates the rotational position of thecamshaft, in addition to a crankshaft speed/positionsensor. The camshaft sensor is included because acrankshaft TDC position reference usually relates tomore than one cylinder, e.g. cylinders 1 and 4, orcylinders 1 and 6, so the ECU is not able to calculatewhich cylinder is on the compression stroke and whichcylinder is on the exhaust stroke, whereas a camshaftonly rotates once for every engine operating cycle, i.e. amaster reference for any of the cylinders will pass thesensor only once for every engine cycle. Therefore acamshaft position sensor can indicate to the ECU theposition of cylinder 1 only (or any other cylinder chosento be the master reference cylinder), so it is possible forthe ECU to control injectors individually, timing themaccurately to each cylinder. It is also necessary to have acylinder reference signal for the modern generation ofignition systems that use individual ignition coils foreach cylinder (there is no distributor rotor arm todistribute the high tension (HT) to each spark plug).

Rotational speed/angular position sensor (Halleffect)Although performing a similar task to the variablereluctance type sensors described above, the Hall effectsensor provides a digital signal as opposed to ananalogue signal.

Hall effect principleFigure 1.23a shows a Hall integrated circuit (IC) or Hallchip. When a small input electrical current is passedacross chip terminals A to B (input current), and the chipis exposed to a magnetic field (magnetic flux), a smallcurrent is then available across C to D (output current).A permanent magnet is located close to the Hall chip,but the magnetic flux can be prevented from reachingthe Hall chip if a metal object is placed between themagnet and the chip. On the example shown in Figure1.23a, the metal object that is used to block themagnetic flux is in fact a rotor or trigger disc, which ismounted on a rotating shaft. The rotor disc has anumber of vanes and cut outs which, when the rotor isturning, alternately block and allow the magnetic flux toreach the Hall chip. The result is that the flow of currentacross the chip terminals C to D will be switched on andoff in pulses. This pulsed signal can provide a speedreference signal to an ECU. Figure 1.23b shows a typicaldigital signal produced by a Hall effect pulse generator.

Hall effect ignition triggerOn some earlier generations of electronic ignitionsystems, but also on some engine management systems,a Hall effect pulse generator was located in the ignitiondistributor body (Figure 1.23c). The rotor disc had thesame number of cut outs and vanes as cylinders. Therotor disc was mounted on the distributor shaft and


Figure 1.23 Hall effect pulse generatora Hall effect pulse generatorb Digital signal produced by a Hall effect pulse generatorc Hall effect system located in an ignition distributor

[647] Chapter 01 25/8/06 16:42 Page 16

rotated at half engine speed, i.e. one complete rotationof the rotor for every engine cycle, which is twocrankshaft rotations. If the rotor had four vanes and cutouts (for a four-cylinder engine), it would provide fourpulses for every engine cycle. The pulsed digital signalwould be passed to an ignition amplifier or to an ECU,which would then switch on and off the ignition coilcircuit, thus producing the high voltage for the sparkplugs (see Chapter 2).

1.5.3 Position sensors for detectingsmall angles of movement

The rotational position sensors described above aredesigned for use on fast rotating components such ascrankshafts. However, there are a number ofcomponents that may only partially rotate, and not infact do so continuously. A very common example is athrottle butterfly or throttle plate. The throttle butterflyis located on a spindle and may rotate through less than90 degrees, from idle through to the fully open position.On engine management systems and on older fuel andignition systems, the ECU requires information relatingto the throttle position to make accurate calculations forfuelling and ignition timing, as well as for some othercontrol functions.

Almost all modern throttle position sensors (seeFigure 1.24) use a potentiometer (variable resistance),which is usually connected to the throttle butterflyspindle, although some types are connected to thethrottle pedal or throttle linkage. The potentiometerprovides a signal voltage that increases and decreaseswhen the throttle is opened and closed, equipping theECU with information about the angular position of thethrottle butterfly. Additionally, the ECU can detect therate at which the voltage increases or decreases,enabling the ECU to calculate how quickly the driver isintending to accelerate or decelerate. Information aboutrate of change of throttle position enables the ECU toprovide more accurate fuel and ignition timing control.

Throttle position sensor analogue signalThe throttle position sensor provides a progressivelyincreasing and decreasing voltage when the throttle isopened and closed. As with many other sensors, thethrottle position sensor requires a reference voltage,typically around 5 volts. The voltage is applied to thepotentiometer resistance track, and a wiper or movingcontact moves across the track when the throttle isopened or closed. Because the resistance along the trackincreases from a low value (possibly as low as zeroohms) to a high value, the voltage at one end of thetrack could be 5 volts whilst at the other end it could beas low as zero volts. As the wiper moves along the track,the voltage at the contact point (wiper onto the track)will change as the wiper moves. The wiper moves withthe movement of the throttle; therefore differentthrottle positions will result in different voltages at the

wiper contact point (see Figure 1.25). The wiper is thenconnected back to the ECU, which uses the voltagevalue as an indication of throttle position.

Although there are variations in the construction ofthrottle position sensors and the signal voltages, it isquite common to have a low voltage of around 0.5 voltsto indicate the throttle closed position and a highervoltage around 4.5 volts to indicate that the throttle isfully open.

Note that some throttle position sensors, especiallyolder designs, have contacts that open and close whenthe throttle is opened and closed. In these sensors one


Figure 1.24 Throttle position sensor and potentiometer schematiclayout

Figure 1.25 Analogue signal produced by a throttle positionsensor compared with angle of throttle opening

[647] Chapter 01 25/8/06 15:36 Page 17

set of contacts is arranged so that they close when thethrottle is fully closed. A second set is also used toindicate when the throttle reaches a certain openingpoint, e.g. 60% open, an indication that the driver isaccelerating or requires more power. Some throttlesensors have a combination of contacts and apotentiometer, although this type is now becoming lesscommon.

There are other components fitted to ECU controlledvehicle systems that also use position sensors similar tothe throttle position sensor, and these are dealt with inthe relevant chapters.

1.5.4 Pressure sensorsThere are generally two main types of pressure sensor: amechanical type and an electronic type.

Mechanical typeOne simple mechanical type makes use of either adiaphragm or capsule, which is exposed to the pressure,or depression (Figures 1.26a and 1.26b).

For example, a pressure sensor can be used to senseengine intake depression (often referred to as enginevacuum). Because engine intake depression varies withengine load and throttle position (and other factors),the sensor can pass a signal to the ECU that indicatesthe engine load. As a result, the ECU can control fuelquantity and ignition timing, although in factinformation is required from other sensors (includingengine speed and throttle position) to enable the ECUto calculate the true engine load accurately.

If the diaphragm type sensor (Figure 1.26a) wasused to sense engine intake depression, the lowerchamber would be exposed to atmospheric pressure andthe upper chamber would be exposed to enginedepression (a lower pressure unless the engine has aturbo or supercharger). When the upper chamberpressure alters (with engine operating conditions) itwill cause the diaphragm to deflect or move within thecasing. The diaphragm can be connected to a lever,which acts on a potentiometer, causing a voltagechange in the potentiometer circuit (using the sameprinciple as the throttle position sensor potentiometerdescribed in the previous section). The signal voltagefrom the potentiometer is passed to the ECU, which isthen able to control functions such as fuelling orignition timing in response to the pressure changes.Note that on the diaphragm type sensor with apotentiometer, the signal is analogue and wouldprogressively change in the same manner as a throttleposition sensor, but in this case the changes occur withchanges in engine intake pressure.

The diaphragm type sensor is in most cases toosimple and inaccurate to be used for modern vehiclesystems such as an engine management system;however, the principle of operation is used for someapplications. A more widely used type in the past wasthe capsule type, whereby a capsule is sealed and

therefore kept at a fixed pressure, and is subsequentlyexposed to the vacuum or depression; when thepressure outside the capsule is lower, the capsulecontracts, moving the rod and potentiometer slider.There are other mechanical methods for convertingpressure change into an electrical signal, althoughmechanical pressure sensors are rarely used on modernvehicles.

Electronic typeElectronic pressure sensors are much more reliable andaccurate than mechanical sensors and have no movingparts (Figure 1.27). A solid state component or siliconchip is exposed to the pressure or depression, whichputs the chip under a strain; the strain alters withpressure change. The change in strain causes a minorchange in length or shape of the crystal. The change inshape or length alters the resistance of the chip;therefore, if the chip forms part of an electrical circuit,the result will be a change of voltage in that circuit.Note that, on some electronic types, the componentunder strain is effectively a thin diaphragm made ofsilicon.


Figure 1.26 Pressure sensors and potentiometer circuitsa Diaphragm typeb Capsule type

[647] Chapter 01 25/8/06 15:36 Page 18

Pressure sensors can be used to measure theatmospheric pressure, fuel line and fuel tank pressure.

Pressure sensor analogue or digital signalElectronic type sensors can produce an analogue or adigital signal, depending on their design. The analoguesignals are generally simple voltage changes thatincrease and decrease according to changes in pressure.Typically, a voltage of around 0.5 volts would indicate astrong engine intake depression (low pressure) such aswould occur at idle speed or low load conditions(throttle closed or almost closed). When the throttle isinitially opened this allows the intake pressure to rise(almost no depression), which results in an increase involtage to approximately 4.5 volts.

Note that not all analogue pressure sensors operatein the same way; therefore voltage values may differ.Some sensors may provide a high voltage when thedepression is strong, and a low voltage when there isalmost no depression.

Digital pressure sensors generally provide a digitalpulse, which has a frequency that changes with thechange in pressure. In effect, the signal provided to theECU is a simple one consisting of many on/off pulses.The ECU effectively counts the pulses and comparesthem against the in-built clock or timer within the ECU.When the pressure changes, the frequency of pulsesprovides the ECU with a reference to the pressure.

Refer to section 1.6 for examples of analogue anddigital signals.

MAP sensorsIt is general practice to refer to the atmosphericpressure as being zero; this is often the value shownwhen a pressure gauge is not connected to a pressuresource, i.e. the pressure gauge is not being used. Wetherefore refer to this as gauge pressure. However, theatmospheric pressure is of course not zero, but is in factapproximately 1 bar (approximately 14.5 lb/in2 or101 kilopascals), even though a gauge may indicate thisas being zero. Therefore a gauge pressure of zeroindicates a pressure of around 1 bar. Note, however,that some gauges are calibrated so that they indicatethe actual or ‘absolute’ pressure.

Absolute pressure is therefore the true pressure valueas opposed to the traditional gauge pressure. If a gaugereading indicates 2 bar, this would in fact be 2 bar aboveatmospheric pressure (which is already at 1 bar); theabsolute pressure is therefore 3 bar.

The same applies to a pressure that is lower thanatmospheric pressure. If the gauge pressure readingwere lower than zero, e.g. a negative value such as‘minus 0.25 bar’, this would be equivalent to anabsolute pressure of 0.75 bar (1 bar minus 0.25 bar).

When a complete vacuum is formed (i.e. there is nopressure at all) the absolute pressure is zero. For thisreason we should not refer to engine intake depressionas being a vacuum. Intake manifold depression is a lowpressure but it is not a true vacuum.

Sensing manifold absolute pressurePressure sensors that are used to sense engine intakedepression generally now measure absolute intakepressure. These sensors are therefore referred to asmanifold absolute pressure sensors (MAP sensors). Theintake pressure is dependent on a number of factorsincluding: throttle opening angle, engine load, airtemperature and density, engine speed, etc. Enginecondition affects the intake pressure; therefore thisfactor also affects the sensed pressure value. Thereforethe absolute pressure value provides a more accurateindication of engine operating conditions.

MAP sensors are generally of the electronic type andmay still provide either an analogue or digital signal.

1.5.5 Airflow sensingAs an alternative to the MAP or pressure sensor methodof assessing engine load, many engine managementsystems and older fuel and ignition systems usedairflow sensors. There are two types of commonly usedairflow sensors: mechanical or electrical/electronic.

MechanicalMechanical airflow sensors are usually referred to asflap or vane type airflow meters. A hinged flap isexposed to the airflow; because the flap is spring loadedto the ‘closed’ or stationary position, increasing theairflow will cause the flap to open to a greater angle(Figure 1.28). The flap is connected to a sophisticatedpotentiometer; as with a throttle position sensorpotentiometer, when the flap moves it results in achange in voltage at the potentiometer wiper contact. Amore detailed explanation is provided in Chapter 2.

When an engine draws in increasing volumes of airon the induction strokes, this causes an increase in theair volume passing through the intake trunking, whichis where the airflow meter is located. Therefore changesin throttle position and engine speed or load will affectthe airflow, thus enabling the airflow sensor to providea relevant voltage signal to the ECU. The ECU is thenable to calculate the engine load and provide therequired amount of fuel.


Figure 1.27 Electronic type MAP sensor

[647] Chapter 01 25/8/06 15:36 Page 19

Vane type airflow sensor analogue signalAs described above, the airflow sensor contains apotentiometer, which provides a signal voltage thatprogressively rises and falls as the vane or flap is movedby the increasing or decreasing airflow. The signal istherefore an analogue signal and is similar inappearance to the signal produced by a throttle positionsensor (Figure 1.25).

Measuring air volume not air massIt is important to note that the flap type airflow sensormeasures air volume but not air mass. For a givenvolume of air, the mass can increase or decrease alongwith temperature and pressure changes. The greater themass of air, the greater the amount of fuel required tomaintain the correct air:fuel ratio. Through measuringonly the volume, the flap type sensor is slightly limitedin its capacity to provide totally accurate information tothe ECU. As an example, if for a given volume ofmeasured air the density were to reduce, this changewould not be registered by a flap type sensor and wouldnot therefore result in a reduction in fuel delivered tothe cylinders; in effect the mixture would be too rich.The inaccuracies are quite small, but because emissionregulations demanded tighter controls, the flap typesensor became less popular and was largely replaced bythe electrical/electronic types of airflow sensorsdescribed below.

Electrical/electronicElectrical/electronic airflow sensors generally operateon what is referred to as the hot wire principle. Hotwire sensors are affected by air density and cantherefore provide an indication of airflow, whichaccounts for the mass of air rather than just the volume.These airflow sensors are often called mass airflowsensors; an example is shown in Figure 1.29.

The principle of operation relies on the fact that, whenair passes across a pre-heated wire it will have a coolingeffect. As the temperature of the wire changes, so doesits resistance. On mass airflow sensors, the sensing wire


Figure 1.28 Flap or vane type airflow sensor: cutaway view and picture/drawing

Figure 1.29 Hot wire airflow sensor

[647] Chapter 01 25/8/06 15:36 Page 20

is heated by passing a current through the wire. Whenchanges in airflow cause a change in the temperatureand therefore changes in the resistance of the wire, thevoltage then changes in the electronic circuitrycontained within the sensor assembly. This circuitry(explained in Chapter 3) compensates for the change insensing wire resistance and applies increased ordecreased current to the wire to maintain the desiredtemperature. The change in required current flow isconverted to a voltage signal that can be monitored bythe ECU, i.e. the airflow mass or a change in the airflowmass results in an appropriate voltage signal passingfrom the sensor to the ECU.

On some types of hot wire system, the wire isoccasionally heated when the engine is switched off to amuch higher than normal temperature, which burns offany contamination or deposits on the wire that couldotherwise affect measurement accuracy. A variation onthe hot wire system is a hot film sensor. The operationis much the same as for the hot wire sensor but anintegrated film type heated sensing element is usedinstead of the heated wire.

1.5.6 Oxygen (lambda) sensors

Reducing pollutants in the exhaust gasOxygen sensors (Figure 1.30) are used on modern motorvehicles for a very specific task: measuring the oxygencontent of the exhaust gas. Whilst the oxygen sensor isnot critical to the direct efficiency of the engine, it iscritical to the efficiency of the exhaust emissions controlsystem (the control of which is generally integrated intothe engine management system). The catalytic converterplays the major part in reducing the pollutants containedwithin the exhaust emissions; the converter, in simpleterms, creates a combustion process. For a catalyticconverter to work efficiently, it must be fed with exhaustgases that contain the required amount of oxygen. Theoxygen sensor is used to measure the oxygen contentand provide a signal to the ECU which will in turncontrol fuelling to ensure that the exhaust gas has thecorrect oxygen level.

Correct air/fuel mixtureAs detailed in Hillier’s Fundamentals of Motor VehicleTechnology Book 1, efficient combustion in an enginerelies on the air and petrol mixture (air:fuel ratio) beingcorrect. The theoretically correct mixture isapproximately 14.7 parts of air to 1 part of petrol (by

weight); this was generally referred to as thestoichiometric air:fuel ratio, but is now referred to aslambda 1.

Although the air:fuel ratio varies under differentoperating conditions, e.g. cold running, light cruise orload conditions, modern engines do operate close to theideal air:fuel ratio for much of the time. On a modernengine, the engine management system uses theinformation from various sensors to enable the ECU tocalculate the required amount of fuel, thus keeping theair:fuel ratio as close as possible to the desired value.

The catalytic converter provides a furthercombustion process (for those exhaust gases that havenot been completely burned within the engine’scombustion process), this additional combustionprocess also requires a correct air:fuel ratio. Theunburned or partially burned gases within the exhaustcontain unburned or partially burned petrol; therefore ifan amount of oxygen is added and the temperaturewithin the converter is high enough, those unburnedand partially burned gases will combine and ignite,hopefully creating a complete combustion of thosegases (thus reducing the pollutants).

Monitoring the oxygen in the exhaust gasIn reality, the exhaust gas can contain enough oxygen toenable the unburned and partially burned fuel to ignite.However, to ensure that the correct amount of oxygen ispresent in the exhaust gas, the air:fuel ratio supplied tothe engine must be precisely controlled, e.g. an excessof petrol (rich mixture) would lead to reduced amountsof oxygen being passed to the exhaust gas. The oxygensensor therefore senses the oxygen content of theexhaust gas and passes a signal back to the ECU, whichif necessary can alter the fuelling to correct the air:fuelratio, thus resulting in the exhaust gas having thecorrect oxygen content.

Although the previous explanation provides a briefunderstanding of the purpose of the oxygen sensor, theoperation of the catalytic converter and the oxygensensor are in fact much more complex. These topics aretherefore explained in greater detail in Chapter 2,dealing with petrol engine emissions control systems.

Oxygen measurement(Refer to Chapter 3 for additional information.)

A typical oxygen sensor is illustrated in Figure 1.30.The sensor uses a natural process that, when specificquantities of oxygen are passed through a certainmaterial, a small voltage is produced. Zirconium oxideis one commonly used material for an oxygen sensorelement.

When the sensor is located in the exhaust pipe, oneside of the sensing element is exposed to the exhaustgas whilst the other side is exposed to the atmosphere.Around 20.8% of the atmosphere consists of oxygen,whilst the exhaust gas typically has around 0.1% to0.8% oxygen; therefore there is a substantial differencein the oxygen levels on the two sides of the sensingelement, causing a small voltage to be produced. The


Figure 1.30 Typical appearance of an oxygen sensor

[647] Chapter 01 25/8/06 15:36 Page 21

exact voltage will depend on the amount of oxygen inthe exhaust gas. The voltage produced by the sensor isthen passed to the ECU, which can alter the fuelling asnecessary to ensure that the oxygen content is correct.The process is almost continuous: the sensor monitorsthe oxygen level and passes a signal to the ECU, whichcorrects the fuelling; this fuel correction then changesthe oxygen level which is again monitored by theoxygen sensor, and so the process continues in a loop.This kind of process is often referred to as a closed loopoperation.

Note that for the sensors to operate efficiently, theymust be at a high temperature (typically above 350°C).The exhaust gas will provide heat but some sensorshave electrical heating elements built in to the sensorbody to speed up and stabilise the heating process.

Because the oxygen sensor is effectively monitoringwhat is now referred to as the lambda value, the oxygensensors are commonly referred to as lambda sensors.However, different manufacturers (of vehicles andsensors) do use different terminology. One example isthe widely used Ford term ‘heated exhaust gas oxygen’(HEGO) sensor.

Pre-cat controlAs detailed above, the combination of the lambdasensor and the ECU effectively controls the fine tuningof the air:fuel ratio to enable the catalytic converter tooperate efficiently. The lambda sensor is locatedupstream (in front of) the catalytic converter and istherefore able to measure the oxygen level in theexhaust gas passing into the converter. The position ofthe lambda sensor in front of the catalytic converter isreferred to as pre-cat control because the combinaton oflambda sensor and ECU controls the oxygen contentbefore it reaches the catalytic converter. Thisarrangement is shown in Figure 1.31.

Post-cat monitoringEuropean legislation (and legislation in othercontinents) demands that an additional function is nowincorporated into emission control systems. Thisfunction is part of a broad range of on-boarddiagnostic (OBD) functions. One aspect of OBD is thatsome form of monitoring should take place to ensurethat the catalytic converter is performing efficiently.

This can be achieved by placing a second oxygen sensorafter or downstream of the catalytic converter (post-cat). This arrangement is shown in Figure 1.32.

If the catalytic converter is not working, the samelevel of oxygen will exit the converter as entered it. Thesecond lambda sensor signal (post-cat) will therefore beidentical to the pre-cat lambda sensor signal. In suchcases the ECU will establish that the catalytic converteris not working and will illuminate the dashboardwarning light. A fault related code or message wouldalso be accessible from the ECU using appropriatediagnostic equipment.


Figure 1.31 Arrangement of catalytic converter and oxygensensor with pre-cat exhaust gas monitoring

1.6 OBTAINING INFORMATION FROM ANALOGUE AND DIGITAL SENSORSIGNALS

Figure 1.32 Arrangement of catalytic converter with two lambdasensors for pre-cat measurement and post-cat monitoring andoxygen sensor

As discussed in section 1.3.3, a modern ECU uses digitalelectronic processes. However, many sensors mightprovide only an analogue signal, which must beconverted by the analogue to digital converter that iscontained within the ECU. Analogue signals producedby sensors vary quite considerably, although essentially

they all provide a progressive change in voltage and cantherefore be treated in a similar way by the analogue todigital converter (A/D converter).

Some examples of typical analogue signalsproduced by some sensors are shown and discussed inthis section.

[647] Chapter 01 25/8/06 15:36 Page 22

Obtaining information from analogue and digital sensor signals 23

1.6.1 Temperature sensor signals(analogue)

Temperature sensors generally provide a signal voltagethat changes progressively with the change intemperature (section 1.5.1). Therefore, when thetemperature increases, the voltage will either decreaseor increase (depending on whether the sensor is an NTCor PTC type). The voltage levels on a temperaturesensor circuit generally range from a maximum ofapproximately 5 volts to a minimum of zero volts(although for normal operation a typical range isapproximately 4.5 to 0.5 volts).

In converting the analogue signal into a digitalsignal, the ECU can use a number of voltage thresholdpoints as reference points, which in effect divide theoperating voltage range into steps (Figure 1.33). Whenthe temperature changes and the voltage consequentlydecreases or increases, each step up or down could becounted to give the ECU an indication of temperature. Ifeach step of 0.5 volts represented a 10° rise intemperature, the ECU would be able to count thenumber of steps up or down and relate this to atemperature value, thus enabling changes in fuellingand ignition timing, etc. In reality, if a greater numberof reference points or steps can be created between themaximum and minimum voltages, the ECU is able toassess smaller changes in temperature, thus providingimproved accuracy.

It is also of interest to note that if the typical sensorsignal voltage is between 0.5 volts and 4.5 volts (whenthe engine and sensor are operating correctly), then anyvoltage above or below those values could be regardedas incorrect. An incorrect voltage is most likely to occuras a result of a faulty component (sensor) or wiringfault. The ECU could therefore be programmed toilluminate a fault light on the dashboard andfurthermore to provide some form of coded message,which could be read or interpreted by diagnosticequipment.

1.6.2 Throttle position sensor signals(analogue)

As highlighted earlier (section 1.5.3), a throttle positionsensor is used to indicate the angle of opening of thethrottle butterfly. Although some earlier throttleposition sensors relied on switches and contacts, almostall modern types use a potentiometer (variableresistor). The form of the output signal from apotentiometer is very similar to that from a temperaturesensor, i.e. it progressively increases and decreases.Therefore when the throttle is opened and closed, thevoltage increases and decreases.

Assuming that the progressive or analogue increaseand decrease in voltage is converted to a digital orstepped signal (in the same way as a temperaturesensor analogue signal is converted into voltage steps),the ECU can establish the angle of opening of thethrottle and the rate at which the throttle is opened andclosed (Figure 1.34). The ECU can count the up ordown steps in voltage to calculate the angle of opening,but can also calculate the speed at which the stepsoccur, thus providing an indication of how quickly thethrottle position is changing. The ECU can then providethe appropriate adjustments to fuelling, ignition timing,etc.

1.6.3 Airflow sensors and MAPsensors (analogue)

Airflow sensors and MAP sensors can provide analogueor digital signals depending on their design. Theanalogue types produce a voltage that increases anddecreases when the airflow volume or mass changes(airflow sensors) or when the manifold intakevacuum/pressure changes (MAP sensors). As withtemperature and throttle position sensors, progressiveincreases and decreases in voltage are converted into adigital or stepped signal so that the ECU can monitorthe changes. The ECU can therefore adjust the fuelling,ignition timing and other functions as necessary, whenairflow, air mass or intake manifold pressures change.

The analogue signals and the subsequent converteddigital signals are therefore similar to those created bythe throttle position sensor (Figure 1.34), although, forthe airflow sensor, it is the change in the airflow thatcauses a change in the voltage.

1.6.4 Crankshaft/camshaft speedand position sensors(analogue)

As described in section 1.5.2, this type of sensorgenerally uses the principle whereby moving or alteringa magnetic field or magnetic flux generates a smallvoltage. Reluctor teeth are located on a rotatingcomponent such as a crankshaft or camshaft, so when

Figure 1.33 Analogue temperature sensor signal with conversionto a digital signal

[647] Chapter 01 25/8/06 15:36 Page 23

the teeth approach or leave the magnetic field (createdby the magnet within the sensor), positive or negativevoltages are generated. The voltage changes form theanalogue signal that is then passed to the ECU. As withother analogue signals, the A/D converter changes thesignal into a digital format that can then be used by theECU (Figure 1.35).

The ECU is able to count the number of pulses, and,because it has a clock or timing device, is then able tocalculate the speed of rotation of the crankshaft orwhatever rotating component is used to generate thesignal. To achieve this speed calculation, there needsonly to be one tooth on the reluctor disc. However, if anumber of teeth are located around the reluctor disc,including a master tooth (a missing or differentlyshaped tooth), the ECU is then able to monitor each ofthe individual pulses generated by each of the teeth.The ECU is able to calculate how many degrees thecrankshaft has rotated from the master position. If, forexample, the master position is TDC for cylinders oneand four, the ECU can assess how many degrees ofrotation the shaft has rotated from TDC. This couldenable the ECU to implement other control functionsthat are crankshaft position dependent, such as openinga fuel injector.

It is also possible for the ECU to assess the speed ofthe crankshaft as each tooth passes the sensor. When acylinder is on the power stroke, the crankshaft speedwill increase, but when the cylinder is on thecompression stroke, the speed will decrease.Additionally, if a particular cylinder has a fault whichreduces its combustion efficiency, then the accelerationof the crankshaft during the power stroke will be lessthan for a good cylinder, leading the ECU to assumethat a fault exists which could prevent petrol fromburning (causing high emissions). The ECU cantherefore switch off the fuel injector for that cylinder.

Note that the ECU will also have information fromthe oxygen sensor, which might indicate that the

oxygen content is too low, i.e. there is excessive unburntfuel. The ECU can use this information, along with thecrankshaft acceleration/deceleration information, todecide whether the fuel injector for the defectivecylinder should be switched off.

1.6.5 Wheel speed sensors(analogue)

Most wheel speed sensors are identical in operation tothe crankshaft speed/position sensors. The maindifference is that, although the rotating disc or reluctordisc contains a number of reluctor teeth, there is nomaster reference tooth. The ECU counts the pulsesgenerated by the teeth; by combining this informationwith the in-built clock information, the speed andacceleration or deceleration of the wheel can becalculated. An ECU on an ABS system is therefore ableto establish whether a wheel is accelerating ordecelerating at a different rate from the other wheels,which would indicate that a brake was locking onewheel. Many other vehicle systems use the informationfrom the wheel speed sensors: these are discussed in therelevant sections of the book.


Figure 1.34 Analogue throttle position sensor signal withconversion to a digital signal

Figure 1.35 Crankshaft speed/position sensor signal withconversion to a digital signala Signal produced by a crankshaft speed sensor with a singlereluctor toothb Signal produced by a crankshaft speed sensor with manyreluctor teeth and one missing master reference tooth

[647] Chapter 01 25/8/06 15:36 Page 24

Note that the analogue and converted digital signalsproduced by a wheel speed sensor are virtually identicalto the crankshaft speed/position sensor (Figure 1.35b).However, there is no master reference tooth; thereforethe signal from the wheel speed sensor is a continuousseries of pulses.

1.6.6 Engine knock pressure sensors(analogue)

Although not previously covered in this chapter, theengine knock sensors are electronic structure bornevibration sensors. A solid state component or siliconchip (usually referred to as a piezo chip or crystal) canbe used to sense pressure changes (section 1.5.4). If thistype of chip is built into a sensor that is attached to theengine (cylinder head or cylinder block), it can be usedto detect high frequency vibrations in the enginecasings when ignition knock occurs (Figure 1.36).

Ignition knock is caused when isolated pockets ofspontaneous combustion occur within the combustionchamber, as opposed to the progressive and controlledcombustion process that should occur. Because modernengines operate very close to the limits at whichcombustion knock can occur, any small variations in fuelquality or hot spots within the combustion chamber canvery quickly cause knock to occur: in effect, the ignitiontiming may be slightly advanced for the conditions atthat time. The knock sensor detects the knock andpasses a signal to the ECU, which in turn slightly retardsthe ignition timing until the knock disappears.

Knock sensors are discussed in detail in Chapter 2,but in simple terms, the sensor produces a smallelectrical signal, which is dependent on the frequency ofthe vibrations; this signal is then used by the ECU tocontrol the ignition timing. The signal provided by theknock sensor is analogue but it is very irregular becausethere is not a consistent rotation or movement of acomponent to create the signal. Although the enginedoes produce regular vibrations, the combustion processalso causes irregular vibrations to occur. The sensorsignal therefore contains voltage spikes caused by allvibrations, which are filtered by the ECU so that it is ableto analyse correctly combustion knock should it occur.

Note that some knock sensors must be tightened tothe correct torque setting when fitted to the engine;over- or under-tightening can affect the capacity of thesensor to detect the appropriate vibration frequencies.

1.6.7 Oxygen (lambda) sensor signal(analogue)

Owing to the complex nature of the oxygen (lambda)sensor signal and the interpretation of the signal by theECU, a full explanation of the signal and how the ECUresponds to the signal is provided in Chapter 3 on petrolengine emissions control systems.

Sensors convert physical quantities into signals

Position sensing is often achieved using a simplepotentiometer

A knock sensor is an accelerometer

1.6.8 Hall effect pulse generator(digital)

As briefly described in section 1.5.2, Hall effect sensorsproduce a digital signal that consists of on/off pulses.Hall effect sensors can therefore be used to providespeed or position related information to the ECU. Suchsensors are used on some ignition systems, where thesensor is located in the distributor body, the sensorhaving one cut out and plate for each cylinderreference. Hall effect sensors are also used as camshaftposition sensors; in such cases, the rotor might containonly one cut out or plate, which would result in onemaster reference signal being passed to the ECU.Because the sensor is mounted on the camshaft, theECU can determine the position (e.g. TDC) of one ofthe cylinders on a multi-cylinder engine. This is notpossible with a crankshaft sensor, because a masterTDC reference on a crankshaft will usually representTDC on two cylinders, e.g. cylinders one and four orone and six.

Key

Poin

ts

Obtaining information from analogue and digital sensor signals 25

Figure 1.36 Knock sensora Knock sensor located in the engine blockb Signal produced by knock sensor

[647] Chapter 01 25/8/06 15:36 Page 25

The signal produced by a Hall effect ignition trigger onolder systems needs only to provide a trigger signal forspark timing. Therefore one pulse of the signalcorresponds to the ignition timing point for eachcylinder; there is no requirement for a master reference(Figure 1.37a). On a four-cylinder engine, the ignitioncoil would produce four high voltage outputs (to createa spark at a spark plug), but the distributor rotor armwould direct the spark to the appropriate cylinder.

On later ignition systems (usually integrated into anengine management system), the distributor is nolonger used; there is often one individual coil for eachcylinder. The ECU therefore needs to be giveninformation regarding the position of one of thecylinders, for example, which cylinder is on thecompression stroke. Once the ECU has established areference to one of the cylinders, it can provide theignition coil control for that cylinder; then the ECU cancontrol the rest of the coils in turn at the appropriateintervals of crankshaft rotation. Remember that theECU will be receiving speed and angular positioninformation from a crankshaft sensor. However, toprovide the master reference for one of the cylinders, aHall effect pulse generator, attached to the camshaft, isoften used. The camshaft rotates once for every enginecycle, so the sensor needs only to provide a single pulse(Figure 1.37b), which indicates that the chosen cylinderis on the compression stroke (or any other stroke orposition, so long as the ECU is programmed with thisinformation).


Figure 1.37 Digital signals produced by a Hall effect pulsegeneratora Signal produced by a Hall effect pulse generator with fourpulses per engine cycle (four-cylinder engine) which is used on asimple ignition system as a trigger reference signal for the fourignition sparksb Signal produced by a Hall effect pulse generator with onepulse per engine cycle. The signal is used as a master referencefor ignition or sequential injection timing

Figure 1.38 ECUs and actuatorsa ECU controlled circuit with a single sensor and single actuatorwhich performs a mechanical taskb ECU controlled circuit with a single sensor and single actuatorwhich performs an electrical task

1.7 ACTUATORS: PRODUCING MOVEMENT AND OTHER FUNCTIONS

1.7.1 Completing the computercontrolled task

If we re-examine the purpose of ECU controlledsystems, the objective is to control a function or taskusing the speed and accuracy that a computer or ECUprovides. Therefore, when the ECU has received therequired information and made the appropriatecalculations, the ECU will provide a control signal to acomponent, which will then perform a task. In general,those components that receive a control signal and thenperform a function or task are referred to as actuators.

Mechanical and non-mechanical actuatorsThe term actuation is generally assumed to mean thatsomething is moved or actuated, and, in a highpercentage of cases with ECU controlled systems, this istrue. The ECU control signal that is passed to theactuator causes some form of movement of acomponent, such as opening an air valve or moving alever (Figure 1.38a). However, there are some caseswhere mechanical movement does not occur, such as

Note that injection system control can also rely on acamshaft located Hall effect trigger. If the injectors areoperated in sequence, i.e. in the same sequence as thecylinder firing order, the ECU will also require a masterreference signal.

[647] Chapter 01 25/8/06 15:36 Page 26

when a light bulb is switched on or off, or when anignition coil is switched on or off (Figure 1.38b).Another example of non-mechanical actuation is wherethe ECU provides a signal to a digital dashboard displayto enable the driver to view engine and vehicle speed aswell as other information. However, even when nomechanical movement takes place, when an ECUprovides a control signal to a component, thatcomponent will usually be referred to as the actuator.

Communication signals between different ECUsAnother example where an ECU provides a controlsignal that does not result in mechanical movement isthe communication of one ECU with another, or withanother electronic device.

An example of ECUs communicating is when theengine management system ECU provides output signalsto an automatic gearbox ECU (Figure 1.39); the enginemanagement ECU might provide a digital informationsignal to the gearbox ECU that indicates engine loadinformation. The engine management ECU is able tocalculate engine load conditions because it receivesinformation from sensors such as the airflow sensor, thethrottle position sensor and temperature sensor.Therefore the engine management ECU can provide asingle ‘engine load’ signal to the gearbox ECU thatprovides sufficient information for the gearbox ECU tomake its own calculations (also using information fromother sensors on the gearbox system). In this example,the engine management ECU is not directly providing anactuator signal but it is providing a signal which assiststhe gearbox ECU to make its own calculations, so that itcan provide a control signal to a gearbox actuator. Inreality, the engine management ECU is still providingcontrol signals to the engine management systemactuators, but the information signal that is being passedto the gearbox ECU is an additional function thatreduces the need for the gearbox system to duplicate thesensors used in the engine management system.

On many vehicles where the engine managementECU passes information to the gearbox ECU, the reversealso applies: the gearbox ECU passes information backto the engine management ECU. For instance thegearbox ECU might inform the engine managementECU that a gear change is taking place, e.g. third tofourth gear. The engine management ECU can thenmomentarily reduce the engine power, which makes thegear change smoother. The engine management ECUcan achieve this by slightly retarding the ignition timingor slightly reducing the amount of fuel injected, and insome cases (if the ECU also controls the throttle

opening electronically) by slightly closing the throttle.Each of these actions would result in a momentaryreduction in engine power.

1.7.2 Actuators and magnetismThere are essentially two types of mechanical movementactuators: one type is the solenoid and the second is theelectric motor. There are a number of variations insolenoids and electric motors, but, in general, solenoidsare used to achieve linear movement and motors areused for rotary movement (although it is possible formotors to be used to create linear movement, via amechanical mechanism, or it is possible for solenoids tocreate rotary movement, via a linkage).

The operation of mechanical actuators (solenoidand electric motor types) relies on magnetism. Hillier’sFundamentals of Motor Vehicle Technology Book 3explains in detail the way in which magnetic fields arecreated and used for electric motors, solenoids andgenerators, etc. However, the essential fact is that,when a current is passed through a coil of wire, amagnetic field is created around that coil of wire. Themagnetic field can then be used to create movement.

Solenoid type actuatorsIn a simple solenoid (Figure 1.40a), a soft iron plungeris located within the coil, but the plunger is free to movewith a linear motion. When an electric current is passedthrough the coil of wire and the magnetic field iscreated, this will cause the plunger to be attractedtowards or through the coil. When the current isswitched off, the spring will return the plunger back tothe start or rest position. Different designs and

Actuators: producing movement and other functions 27

Figure 1.39 Communication between engine management andautomatic gearbox ECUs

Figure 1.40 Simple solenoidsa Simple solenoidb Double acting solenoid

[647] Chapter 01 25/8/06 15:36 Page 27

constructions of solenoids allow many different tasks tobe performed. For example, the double acting solenoid(Figure 1.40b) uses two coils of wire. One coil creates amagnetic field, which moves the plunger in onedirection, and the other coil creates a magnetic field,which moves the plunger in the opposite direction.

It is also possible for the ECU to regulate or controlthe average current flow and voltage passing throughthe coil of wire by altering the duty cycle and frequencyof the control signal pulses (see section 1.8). With thiscontrol process, it is possible to control or regulate thestrength of the magnetic field. If the plunger is movingagainst a physical resistance such as a spring, it can bemoved further by increasing the strength of themagnetic field. Reducing the magnetic field will resultin the plunger moving back slightly. Additionally, whena double acting solenoid is used, the plunger movementcan be controlled in both directions; in fact onemagnetic field can be used to oppose the other. Thisallows an ECU to move and position the plunger withreasonable accuracy.

Solenoid plungers can be connected to a number ofdifferent types of mechanisms or devices that willperform different tasks or functions; various solenoidactuators are covered in the relevant chapters withinthis book.

Electric motor type actuatorsA simple electric motor operates on similar principles tothe solenoid, but instead of the magnetic field causing aplunger to move with a linear motion, the magneticfield forces a shaft to rotate. Figure 1.41 shows a simpleelectric motor, which in this example has a permanenthorseshoe shaped magnet with a north and south pole.A single loop of wire, which would normally beattached to a rotor shaft, is fed with an electric current,thus creating an electromagnetic field around the loop


Figure 1.41 Simple electric motor. Note that the primary andsecondary windings are wound around a soft iron core toconcentrate and intensify the magnetic fielda Current passes from A to B creating north and south poles onthe electromagnet. The like poles will cause the shaft and wireloop to rotate

b When the rotor has turned through 180°, the commutatorarrangement causes the current to flow in the reverse directionaround the wire loop (from B to A), therefore changing the northpole to a south pole and the south pole to a north pole. The likepoles will again repel and cause the shaft to rotate throughanother 180°

of wire. When the electromagnetic field is created,north and south poles will exist around the loop of wire.These north and south poles will either be attracted toor repelled from the north and south poles of thepermanent magnet. Remember that like poles repeleach other and unlike poles attract each other.

When the current is initially passed through the wireloop, e.g. from connection A to connection B on thewire loop, if the electromagnet north pole is adjacent tothe permanent magnet north pole (and the two southpoles will also be adjacent to each other), this will forcethe shaft to rotate (Figure 1.41a). When the shaft thenrotates through 180º, the north poles will be adjacent tothe south poles, and because unlike poles attract eachother, the motor will not rotate any further.

However, in the diagram it can be seen that the pairof semi-circular segments (or commutator) is attachedto the ends of the wire loop and therefore rotates withthe loop. The electric current passes from the powersupply to contact brushes which rub against thesegments as the shaft rotates. Therefore, when the shaftand the segments have rotated through 180º, the twosegments are now not in contact with the originalbrushes, but they are in contact with the opposingbrushes. This means that the electric current will beflowing from connection B to connection A (Figure1.41b), which is in the opposite direction around thewire loop. The result is that the north pole of theelectromagnet is now a south pole, and the south pole isnow a north pole, which will cause the shaft and wireloop to rotate another 180º; the process is thenrepeated.

The simple electric motor in Figure 1.41 shows howmagnetism can provide continuous rotary movement;the resulting rotary motion can operate various devices.Simple examples include fuel or air pumps, and wipermotors operate on the same principles.

[647] Chapter 01 25/8/06 15:36 Page 28

However, many of the electric motors used on ECUcontrolled vehicle systems are often more complex andsophisticated in the tasks they have to perform, and intheir design and construction. Many of the motors donot in fact perform a complete rotation, or they may becontrolled so that they rotate in small angular steps.These types of motors are controlled by using differenttypes of wire loops (usually coils of wire) and usingdifferent designs of commutator. In addition, byapplying control signals from the ECU that havechanging duty cycles, pulse widths and frequencies, it ispossible to rotate motors partially so that they start andstop in any desired position. The partial rotation can beprogressive from one position to another, or it can beachieved in a series of steps.

The capacity to control the rotation of motorsaccurately allows them to be used for a variety of taskssuch as opening and closing air valves in smallincrements (used for idle speed control). Otherexamples of ECU controlled motors are dealt withindividually in the following sections and in otherchapters of this book.

Magnetism and non-mechanical actuatorsThere is one main actuator used on motor vehicles thatuses the effects of a magnetic field but does not producemechanical movement – this is the ignition coil.

An explanation of how an ignition coil works isprovided in Chapter 2 of Hillier’s Fundamentals of MotorVehicle Technology Book 1. It is sufficient here to highlightthe basic principles of ignition coil operation, which relyon the movement of a magnetic field or magnetic flux toinduce an electric current into a coil of wire.

When a current is passed thorough a coil of wire, itcreates a magnetic field; this is the same principle asused in electric motors. Additionally, as is the case withan electrical generator, when a magnetic field movesthrough a coil of wire (or the coil is passed through amagnetic field) it causes an electric current/voltage tobe generated within the coil of wire. The faster themagnetic field moves relative to the wire, the greaterthe voltage produced. An ignition coil relies on bothprocesses.

On most vehicles, the voltage in the vehicleelectrical system is only around 12 volts, which is notsufficient to create a spark or electric arc at the sparkplug gap. The ignition coil must provide a way toincrease the voltage from 12 volts to many thousands ofvolts. A principle that is used in electrical transformersis also used for ignition coils: there are two coils of wire,one of which has many more windings than the other.In an ignition coil a secondary coil can typically have100 times more windings than the primary coil (seeFigure 1.42).

The processThe process relies on current (using the vehicle’s 12 voltsupply) passing through the smaller coil or primarywinding to create a magnetic field. The build up of themagnetic field is relatively slow, but once the magnetic

field has been established at full strength, it can bemaintained for a very brief period so long as the currentcontinues to flow. However, when the current isswitched off, the magnetic field collapses extremelyrapidly, in fact very much more quickly than the speedat which it was created.

Whilst the magnetic field is collapsing, the lines ofmagnetic force are collapsing across the same coil ofwire that created it (primary winding); this causes acurrent/voltage to be produced within the primarywinding. Because the speed of collapse of the magneticfield is very rapid, it causes a much higher voltage to beproduced within this coil of wire, sometimes as high as200–300 volts. Therefore, the speed of collapse is usedto step up the voltage from 12 to typically 200 volts.However, 200 volts are still not sufficient to provide thespark at the spark plug under the conditions that existin the combustion chamber (high pressure and otherfactors make it difficult for an arc to be created at theplug gap).

To achieve the desired voltage necessary to createthe spark, a secondary winding is used, as mentionedabove. The secondary winding can be adjacent to theprimary winding, although one winding is oftenwrapped around the other. When the magnetic field iscreated, the secondary winding is also exposed to themagnetic field. Therefore, when the magnetic fieldcollapses, as well as creating a voltage in the primarywinding, it also creates a voltage in the secondarywinding. Because the secondary winding may have 100times the number of turns or windings, 100 times thevoltage can in theory be produced. If 200 volts could beproduced in the primary winding (owing to the rapidspeed of collapse of the magnetic field), then in thesecondary winding it should theoretically be possible toproduce 20 000 volts (100 times greater).

Actuators: producing movement and other functions 29

Figure 1.42 Simple construction of an ignition coil

[647] Chapter 01 25/8/06 15:36 Page 29

For most petrol engines, the required voltage to producea spark at the spark plug (under operating conditions)is around 7 000 to 10 000 volts; therefore a coil that isable to produce 20 000 volts is more than capable ofproducing a spark. There is therefore sufficientadditional voltage available to overcome many minorfaults such as a plug gap that is too large orcontaminated.

Actuators convert electrical signals into actions

Common actuators, such as fuel injectors, aresolenoid operated

Key

Poin

ts


1.8 EXAMPLES OF DIFFERENT TYPES OF ACTUATORS

1.8.1 Solenoid type actuatorsThere are many different types of solenoid typeactuators used on motor vehicle systems, a number ofwhich are covered within this book. The followingexamples deal with two types that are used for totallydifferent tasks.

The first example is a fuel injector, which providesvery rapid opening and closing of a small valve with theresult that fuel flow into the engine (into the intakemanifold or combustion chamber) can be accuratelycontrolled; the amount of movement required to openand close the injector is very small.

The second example is the use of a solenoid as an airvalve. In this example, the valve forms part of apressure/vacuum circuit which is used to control aturbocharger wastegate. The valve does not have tooperate at the same speed as the injector, but it willrequire greater movement.

Fuel injectorFuel injectors are high precision components used tocontrol the flow of fuel into the engine. The injectorsare usually located in the intake manifold and thereforeinject fuel in the region of the intake valves. On somemodern petrol engines, the injectors are located so that

fuel is injected directly into the cylinder. Modern dieselengines that now use electronic control for the fuelsystem also use electronically controlled solenoidinjectors that inject fuel directly into the combustionchamber.

Figure 1.43 shows a typical construction for asolenoid type petrol injector. The injector has a 12 voltsupply from the vehicle’s electrical system, which isusually a permanent supply (via a relay) whilst theignition is switched on (engine running). The earthcircuit for the injector passes through the ECU, whichacts as the control switch.

The injector must open and close very rapidly and athigh frequency. The opening and closing time can oftenoccur in around three thousandths of a second(3 milliseconds or 3 ms), and injectors might open andclose more than 7 000 times a minute.

Solenoid air valveThe example shown in Figure 1.44 is a relatively simplesolenoid that is used to control the pressure acting on adiaphragm. The pressure is produced by a turbocharger,which causes the intake manifold to be subjected topressure (when the turbocharger is operating) as wellas the normal vacuum levels for low load engineconditions (when the turbocharger is not operating).

Figure 1.43 Solenoid type petrol injector and basic wiring

[647] Chapter 01 25/8/06 15:36 Page 30

When the pressure produced by the turbochargerbecomes too high for engine safety, the ECU will causethe solenoid to operate and thus open the valve. Thiswill allow pressure from the intake manifold to act onthe diaphragm in the wastegate, which in turn will openand allow pressure from the turbocharger to escape(often into the exhaust system or other separate pipesthat lead to the atmosphere).

To switch the air valve, the solenoid receives apermanent power supply whilst the engine is running,and the ECU controls the earth circuit. The solenoid airvalve does not have to operate at the same speed andfrequency as the fuel injector, but the movement of thevalve is usually much greater.

1.8.2 Examples of electric motortype actuators

Electric motors used on modern vehicles systems can becategorised into three main types: full and continuousrotation; full rotation with controlled positioning; andpartial rotation with controlled positioning.

● Continuous rotation motors are effectivelyconventional electric motors; the example used inthis section is a motor that is used to drive a fuelpump.

● Full rotation motors with controlled positioning areused to position a mechanism or device such as anair valve or a throttle butterfly. In most cases, themotor may rotate through more than one completeturn, but it can be stopped at a desired position.Some stepper motors used for idle speed controloperate on this principle.

● Partial rotation motors use the same principles ofoperation as a normal motor but the angle ofrotation is limited. The example used in this sectionis a motor that is used to control an air valve, which

in turn controls the volume of air passing into theengine at idle speeds.

Continuous rotation fuel pump motorThe example shown in Figure 1.45 is a conventionaltype electric motor, which is used to drive a fuel pump.In this example, the motor and pump assembly aremounted outside the fuel tank, although for manyapplications, an adaptation of this type of pump islocated inside the fuel tank.

The pump will receive a power supply, which isusually fed via a fuel pump relay (often forming part ofan engine management system relay). The pump willusually have a permanent earth connection.

Examples of different types of actuators 31

Figure 1.44 Solenoidoperated air valve

Figure 1.45 Electric motor driven fuel pump

[647] Chapter 01 25/8/06 15:36 Page 31

Idle speed stepper motor with full rotation andcontrolled positioningIn the example shown in Figure 1.46, the motor is ableto rotate fully (possibly for more than one completerotation) but it can be positioned electrically byswitching on and off the current passing into the motor.The motor contains more than one set of magnets andelectromagnets. This construction enables the ECU toswitch on and off each electromagnet, which enablesthe motor to rotate in small steps in either direction.The different control signals to the stepper motor willtherefore have a series of on/off pulses which can bepositive or negative to achieve clockwise or anti-clockwise rotation of the motor.


Figure 1.46 Stepper motor Figure 1.47 Rotary idle valve using partial rotation motor

1.9 ECU/ACTUATOR CONTROL SIGNALS

Idle speed partial rotation motor (rotary idlevalve)In this design (Figure 1.47), the motor armature isrestricted by mechanical stops from rotating throughmore than approximately 60°. Connected to the end ofthe armature is an air flap or air valve assembly which,when opened and closed, will regulate the air passinginto the engine, thus enabling idle speed to becontrolled.

In the simplest type, a spring keeps the motorarmature rotated against one of the mechanical stops.However, when an electric current is applied to themotor (creating electromagnets), this will cause thearmature to rotate against the spring. The ECU controlsthe average current flowing in the circuit by altering theduty cycle of the control signal. The greater the averagecurrent, the more the armature will rotate against thespring force. By continuously altering the duty cycle it isthen possible to alter the angular position of thearmature.

1.9.1 ECU functioning as a switch ina circuit

The control signal provided by the ECU to an actuator ismost commonly a digital signal, which effectivelyswitches the actuator on or off; this is achieved in mostcases by making the ECU a part of the actuatorelectrical circuit. The ECU is therefore acting as asophisticated switch that makes or breaks (switches onor off) the actuator circuit (Figures 1.48a and 1.48b).

As previously described (see the text about amplifiers insection 1.3.3), the ECU usually contains a final stagepower transistor, which is effectively the actuator circuitswitch. The low voltage signal from the ECU’smicroprocessor simply controls the power transistor,which then replicates or copies the control signal. Butbecause the power transistor is the switch within theactuator circuit, when the microprocessor control signalis on or off, it causes the power transistor to switch onor off, thus making or breaking the actuator circuit.

[647] Chapter 01 25/8/06 15:36 Page 32

transistor to switch on and off the actuator circuit.However, it is not just simple on or off control that isprovided by the ECU: most control signals will causethe actuator to switch on or off for different lengths oftime and at different speeds or frequencies (measuredin hertz (Hz)).

When examining the control signal, the duration ofthe on or off period can be referred to as pulse width. Itis however general practice that the pulse width refersto the on time only.

Figure 1.49a shows an on/off control signal wherethe on time or pulse width is 1⁄2 second, and the off timeis also 1⁄2 second. In this case the frequency is 1 hertz,which means that the actuator is switched on and offonce every second.

Figure 1.49b shows a similar signal, but the on timeis 1⁄4 second, with the off time being 3⁄4 second. Thefrequency is therefore still 1 hertz but the on and offtimes are different.

Figure 1.49c shows a control signal with equal onand off times but the frequency is 10 hertz.

The completion of the on and off process is onecomplete cycle of operation or 1 cycle. Therefore, whenthe signal completes one on and one off pulse, this isalso referred to as 1 cycle. If there are 10 cycles withinone second, this is a frequency of 10 cycles per second,which is referred to as 10 hertz (10 Hz).

If the durations of the on and off times are the same,this is referred to as a duty cycle of 50%, i.e. the on timeis 50% of one cycle. If however the on time is 1⁄4 of thetotal cycle time then this is referred to as a duty cycle of25%.

Figure 1.50a shows two control signals, each with a50% duty cycle. Although the durations and frequenciesof the two signals are different, the duty cycles are 50%in both cases. Figure 1.50b shows two control signals,

ECU/actuator control signals 33

Figure 1.48 Switching an actuator circuita Normal switch controlling an actuator circuitb ECU acting as the switch and controlling the actuator circuit.Note that the power transistor in the ECU directly switches thecircuit in response to the signal from the ECU microprocessor

Note that for most ECU controlled actuator circuits, theECU (power transistor) forms part of the earth or returncircuit (negative path). The positive path from thepower supply (the battery) can be directly connected tothe actuator or it may contain a switch such as anignition switch. Fuses and relays are also generallyconnected into the positive side of the circuit. The ECU,which provides the controlling function, is thereforemaking and breaking (switching on and off) the earthor negative side of the circuit.

Most control signals provided by the ECU aretherefore simple on/off pulses that cause the power

Figure 1.49 ECU control signal duration and frequencya ECU control signal with equal on and off duration of 1⁄2 secondand a frequency of 1 hertzb ECU control signal with on duration of 1⁄4 second and offduration of 3⁄4 second but with a frequency that is still 1 hertzc ECU control signal with equal on and off duration, but with afrequency of 10 hertz

[647] Chapter 01 25/8/06 15:36 Page 33


Figure 1.50 Duty cyclesa Control signals with 50% duty cycleb Control signals with 25% duty cycle

each with a 25% duty cycle, and again, although thedurations and frequencies are different, the duty cyclesare the same.

Important note: In the control signal examplesillustrated, the off time is shown as the higher portionof the pulse, i.e. as a voltage level. The on time istherefore shown as zero volts. When a switch (in thiscase the ECU) is connected into the earth or negativepath of an actuator circuit, the earth circuit will in factbe at zero volts when the circuit is switched on and atbattery level voltage when the circuit is switched off. Itis important to note when using test equipment, suchas multimeters or oscilloscopes, that themeasurements displayed may be the reverse of theexpected readings, e.g. the duty cycle could be shownas 25% instead of 75%.

1.9.2 Using the signal to control theactuator

By understanding that the duration (pulse width), dutycycle and frequency of the control signal can be altered,it is possible to understand how an actuator can becontrolled so that the task it performs can be varied. An

example is a fuel injector, which can be provided with acontrol signal where the duty cycle or pulse widthvaries. This means that the injector can be opened forlonger or shorter time periods, thus allowing differentquantities of fuel to be delivered to the engine.

The control signals affect how the actuator operatesin different ways because the actuator is altering thecurrent flow in the circuit. It was explained previouslythat the ECU is effectively an on/off switch, but this isonly part of the whole story.

Altering the control signal duty cycleAltering the duty cycle or pulse width has the effect ofaltering the average current flow and applied voltage ina circuit.

As an example, a simple 12 volt light circuit isswitched on and off by a simple switch (Figure 1.51a).When the circuit is switched on, the voltage on thepower supply to the bulb will be 12 volts. Because thelight bulb has a 2 ohm resistance, the current willtherefore be 6 amps and the bulb will produce itsmaximum light output. However, when the switch isoff, the voltage and current will both be zero and thebulb will produce no light.

If the light switch could be switched on and offvery rapidly, for example at 100 times a second

[647] Chapter 01 25/8/06 15:36 Page 34

(Figure 1.51b), and the duty cycle was 50%, i.e. the onand off pulses were both 50% of 1 cycle (of equalduration), the result would be that the light would beon for only half of the time. This means that theaverage voltage, the average current and the averageamount of light produced by the bulb would also be50% of the maximum value had the bulb beenswitched on all the time.

ECU/actuator control signals 35

Figure 1.51 Altering the duty cycles to affect the average voltageand currenta Simple 12 volt light circuit and switchb Average voltage and current in a light circuit with a 50% dutycyclec Average voltage and current in a light circuit with a 25% dutycycle

In this example where the bulb is rapidly switched onand off, the average voltage on the power supply circuitis 6 volts because this is 50% of the maximum supplyvoltage (50% of 12 volts). The average current istherefore 3 amps (50% of 6 amps). The total amount oflight produced by the light bulb should in theory be50% of the light that would have been produced if thebulb had been illuminated for all of the time.

If the duty cycle was changed so that the on timewas only 25% of the total cycle (Figure 1.51c), then12 volts would be available to power the bulb for only25% of the time, but zero volts would be supplied for75% of the time. The average voltage would thereforebe 25% of 12 volts, i.e. 3 volts. The average currentwould therefore also be 25% of the maximum 6 amps,i.e. 1.5 amps. The average amount of light producedwould therefore in theory also be 25% of the maximum.

If this same process of altering the duty cycle isapplied to a control signal that is being used on anactuator such as an electric motor, it is then possible toalter the power produced by the motor. The sameapplies to any actuator control signal, where alteringthe duty cycle will influence the way in which theactuator functions.

Altering the control signal timing and frequencyIf the control signal consists of simple on and off pulses,an actuator will also be switched on and off. It istherefore possible to provide the on and off pulses at aspecified time. A common example is when a fuelinjector used on a modern fuel injection/enginemanagement system is required to open at a certaintime in the engine operating cycle. The injectors onsome modern systems will open just before, or at thestart of, the intake stroke (possibly just before or just asthe intake valve opens). A sensor (usually the camshaftposition sensor) is used by the ECU as a timingreference to calculate when the intake stroke is about tostart, allowing it to provide the on pulse in the controlsignal at the right time.

The frequency of the control signal also affects howan actuator behaves. For instance, a simple solenoidcould be used to open and close a small valve (whichcould be allowing fuel to pass through a pipe). If thecontrol signal had a 50% duty cycle, and provided onand off pulses that occurred very slowly, e.g. every 10seconds, the solenoid would open the valve for 10seconds and close the valve for 10 seconds. Althoughthis would regulate the flow of fuel in the pipe, it is nota very effective means of control. If, however, thecontrol signal pulses occurred 100 times every second(100 hertz), this would mean that the solenoid wouldbe trying to open and close 100 times a second. Thesolenoid would in fact adopt a half open position i.e. itwould never reach the fully open or fully closedpositions. Therefore, altering the duty cycle will affectthe average opening time of a solenoid controlled valve,but it is more effective if the frequency is high (suchas 100 hertz) than it is if the frequency is low.

[647] Chapter 01 25/8/06 15:36 Page 35

Web linksEngine systems informationwww.bosch.comwww.sae.orgwww.imeche.org.ukwww.picotech.comwww.autotap.comwww.visteon.com www.infineon.comwww.kvaser.com (follow CAN Education links)

Teaching/learning resourcesOnline learning material relating to powertrainsystems:

www.auto-training.co.uk


[647] Chapter 01 25/8/06 15:36 Page 36

ENGINE MANAGEMENT –SPARK IGNITION

Cha

pter

2

2.1 EMISSIONS, RELIABILITY AND DURABILITY


Emissions, reliability and durability

Electronic ignition systems (early generations)

Computer controlled ignition systems

Distributorless and direct ignition systems

Spark plugs

This section relates to systems covered in chapters 3, 4and 5.

2.1.1 Emissions legislationThe introduction of electronic engine control systemswas a result of a number of factors, most of which stillapply. As noted in Chapter 1, affordable electonicsenabled vehicle manufacturers to make increasing useof electronic components and electronically controlledsystems. However, in the 1960s through to the early1990s, it was in many cases still less expensive to fitmore traditional fuel and ignition systems, such ascarburettors and contact breaker ignition. At some stagetherefore, vehicle manufacturers needed somemotivation to fit electronic systems that in the earlystages were still more expensive to produce and todevelop than the existing components at that time.

Probably the single biggest factor in the increasinguse of electronic systems was the introduction ofemissions legislation. It is generally accepted that theUSA was the leading country in introducing legislationthat forced a reduction in emissions levels produced byengines and vehicles in general. Legislation forcedvehicle manufacturers to develop and fit electronicsystems to engines. This process really started towardsthe end of the 1960s, when electronics had just reacheda level of capability and cost that enabledmanufacturers to start to build systems that usedelectronic components.

What did make things slightly difficult was thatdifferent states in the USA had different problems;therefore they had different requirements anddifferent legislation. Smog is one particular problem

(Figure 2.1) that captured everyone’s attention. Smog isa term that became commonly used with reference tofog that was not naturally formed in the atmosphere: itwas created from smoke that had been produced byfactories, houses and of course cars. Burning fossilbased fuels such as coal, petrol, diesel and even woodproduces smoke, and many towns and cities around theworld suffered with smog. One of the most famous isLondon, which had for many years had a serious smogproblem. In fact, ever since entertainment films havebeen made, it has been common to depict London (evenin the 1800s) as having a serious smog or fog problem.

The smog in London had been present for too manyyears for it to be blamed entirely on motor vehicles, butwhen certain weather conditions existed (which couldhave produced normal fog), smoke produced by coal orwood fires in houses and smoke from the factorychimneys added to the problem. Although motorvehicles must have started to contribute to the problem,the relatively low number of vehicles in use up until the1960s was not sufficient to be the major cause.

Figure 2.1 Smog in a city

[647] Chapter 02 25/8/06 15:37 Page 37

The smog problem that occurred in Los Angeles,California, which became very serious in the 1950s and1960s, occurred because Los Angeles lies in a valleywhere there is often little or no wind and a lot of heat.In isolation, this could be regarded as a naturalproblem, but in reality smog developed because ofemissions from the burning of fossil fuels. Los Angeleshad a high concentration of motor vehicles that, evenin the 1960s, spent a lot of time in traffic jams in whatwas a confined area. Therefore a combination oflocation, weather conditions and burning fossil fuelscaused extremely serious problems.

Whilst smog can be quite thick and therefore makesdriving hazardous, the real problem is a health issue.Serious illness, such as respiratory problems, can becaused by smog and many deaths have also beenattributed to smog. The acids contained within vehiclerelated smog cause damage to buildings as well aspeople. Athens in Greece is a particular case whereacids are destroying many of the ancient buildings.

Motor vehicles were a contributing factor to smogproblems so they became a focus of attention forlegislators. Legislation dealing with factories and firesin houses has also been brought in. In the UK, the CleanAir Act restricted the use of many types of fossil fuels inboth homes and factories, and London is nowsignificantly free of serious smog.

Vehicle emissions are a broad subject and smog isonly one of a number of problems influenced or createdby such emissions. Therefore, within this chapter,emission problems are referred to and explained whereapplicable, but particular reference should be made tothe emissions section.

If we appreciate that vehicle emissions can cause orcontribute to serious health and environment problems,then we must accept that any effort to reduce theproblem is justifiable, even at a cost. The Los Angelesproblem, amongst others, was without doubt a majortalking point that captured the attention of the public:the consumers that buy motor vehicles. Therefore anyreasonable added cost for the vehicle became relativelyacceptable. The vehicle manufacturers were thereforetasked by legislators to reduce the level of vehicleemissions. It is perhaps coincidental or fortunate that ata time when emissions problems were a major focus ofattention for legislators, electronics were becoming verymuch more capable and affordable.

Note: In the USA, diesel engines are very rarely fitted topassenger vehicles. The cost of petrol was and remainslow compared to most countries and the petrol enginewas more acceptable as a means of powering largeAmerican cars. Emissions legislation and technicaldevelopments in the USA were therefore focused on thepetrol engine rather than the diesel engine. Europe andother regions were therefore able to take advantage oflegislation and technical changes made in the USA, andit was not until more recent times in Europe that dieselengines have become a target for substantial emissionsreduction (and therefore technical change).

Engine maintenanceEngine design in the 1960s had not really changed toomuch for many years. Although improvements hadbeen made, such as overhead valves instead of sidevalves, the main objective was to improve engineperformance, which at that time primarily meant morepower. In the USA, in particular, vehicles were generallymuch larger than in the rest of the world, and becausepetrol was very inexpensive in the USA, large fuelthirsty engines fitted in large heavy vehicles wereaccepted as normal.

Engines in the USA at that time were usually of V8configuration with typical capacities of 4 litres to7 litres. Emissions levels from these large engines werevery high, especially when the engine was at idle speed(when measured as a percentage of the total exhaustgas, some pollutants were at their highest levels at idlespeed). It was therefore obviously going to be verydifficult to change vehicle and engine design suddenly,so the changes were generally planned over a relativelylong period of time.

However it was recognised that one major factorthat could help to reduce emissions was to ensure thatregular maintenance was performed correctly, butideally, the need for regular maintenance on the fueland ignition systems should be reduced or eveneliminated. For those readers who have never workedon older vehicles with carburettors and contact breaker(points) ignition systems, it may be difficult toappreciate that it was necessary to clean and adjust thecarburettor regularly and to adjust or replace the

38 Engine management – spark ignition Fundamentals of Motor Vehicle Technology: Book 2

The formation of photochemical smog

H2OWater

HN3Nitric acid

O3Ozone

Ultravioletradiation

Solarradiation

OAtomicoxygen

NO2Nitrogen dioxide

O2Molecular

oxygen

Figure 2.2 Heat and exhaust gases can cause smog and otherpollution, especially in certain types of geographical location

[647] Chapter 02 25/8/06 15:37 Page 38

contact breaker in ignition systems. Ignition or fuelsystems would not suddenly become inefficient at aparticular interval, such as 16 000 km or 10 000 miles:the reality was that ignition and fuel systemsprogressively deteriorated. Therefore, immediatelyafter ‘perfect’ maintenance had been carried out,systems would progressively become less and lessefficient, until a time was reached when theinefficiency was unacceptable. Hopefully, regularmaintenance was performed before systems becametoo unacceptable.

See Hillier’s Fundamentals of Motor VehicleTechnology Book 1 for an explanation of carburettorsand contact breaker ignition systems.

Loss of efficiency of fuel and ignition systemsCarburettors would progressively accumulate a buildup of deposits in the airways and petrol jets (smallholes through which fuel passed). This build up ofdeposits would then alter the air/fuel mixture, whichcould result in reduced combustion efficiency and highemissions. It was also common for the carburettor torequire slight adjustment of the idle mixture setting, tofurther compensate for build up of deposits, and even achange in the weather. Another major factor that madeit necessary to adjust the idle mixture setting was thedeterioration in the way the engine was performing,which was caused by wear of the contact breakers inthe ignition system.

Progressive wear on the contact breakers was anacceptable part of the design, and it was possible toadjust them to compensate for wear until such time asthe wear was too excessive. Contact breakers on veryold vehicles may have required replacement after aslittle as 5 000 miles (8 000 km) but even theimproved versions still required regular adjustmentand were traditionally replaced at main serviceintervals, which could be between 5 000 and10 000 miles (8 000 and 16 000 km).

One particular problem related to wear on thecontact breakers is that it can reduce the quality orstrength of the spark, because a worn set of contactbreakers (or an adjustment that is out of specification)will not allow sufficient time for the ignition coil tobuild up a strong magnetic field. This results in weakelectrical output from the ignition coil and therefore aweak spark. The result is that the combustion of theair–fuel mixture will not be efficient; this results inhigh emissions of pollutants.

A second problem with worn contact breakers isthat, as the contact breakers wear, the ignition timingchanges. See Hillier’s Fundamentals of Motor VehicleTechnology Book 1 for a full explanation of this.However, in simple terms, wear in the contact breakermechanism causes the contact breakers to open earlieror later. The opening time of the contact breakerscauses the ignition coil to provide the electrical energythat in turn causes a spark at the plug, so wear in thecontact breakers affects spark timing.

Incorrect ignition spark timing will reduce combustionefficiency, with the result that emissions of pollutantsincreases, power is reduced and more fuel is wasted.When ignition timing is incorrect (usually retarded orlater than specified), the engine will not idle smoothly.However, it is possible on older vehicles to adjust theair/fuel mixture at idle speed, which helps to smoothout the way the engine is operating. This is generallyachieved by making the mixture richer than normal, i.e.an excess of fuel. This in turn can also create higheremissions.

A third problem with contact breakers is that, eachtime the contacts open, a small arc can occur across thecontacts. Although this problem was very much reducedby the introduction of a condenser or capacitor in thecontact breaker circuit, arcing progressively damagedthe two contact points on the contact breakers.

Other systems requiring maintenanceIt was more than just simple carburettor and contactbreaker maintenance that resulted in increasingemissions. Older engine designs included manycomponents and systems that required regularmaintenance, which involved cleaning, adjustment orreplacement. Those areas requiring regular maintenanceincluded:

● valve operating clearances● spark plugs, which required regular changing or

resetting of the plug gap● engine oil, which required regular changing● engine breather systems● contact breakers and ignition timing● fuel systems (carburettors).

Each of the above listed items would progressivelywear, or their operating performance wouldprogressively deteriorate. In turn, this would eitheraffect combustion efficiency (which would increaseexhaust emissions), or would result in excessive engineoil fumes and emissions. Without regular and correctmaintenance, many of those items listed above wouldend up operating ‘out of specification’, which meansthey would be operating outside their intended designlimits. The net result was that the engine would beoperating very inefficiently and excessive emissionswould be produced.

2.1.2 Reliability and durabilityEven if an engine was maintained at the recommendedregular intervals, wear and progressive deterioration inoperating performance still occurred between themaintenance intervals, thus causing an increase inemissions. Ideally therefore, those adjustable items, oritems that wear and deteriorate between maintenanceintervals should be redesigned to avoid any reduction inoperating performance between maintenance intervals(or for longer if possible). In effect, designers were

Emissions, reliability and durability 39

[647] Chapter 02 25/8/06 15:37 Page 39

trying to create much improved reliability anddurability of engine components and systems so thatemissions did not increase significantly betweenmaintenance intervals. In fact, one piece of legislationcalled for ignition systems to be able to operate withoutmaintenance for 50 000 miles (80 000 km).

Ignition system developmentsA number of design changes were made over a period oftime to various engine components and systems.However, one particular engine system benefited fromthe use of electronics: the ignition system.

Early designs of electronic ignition (fitted as originalequipment) were designed to eliminate the contactbreakers (see section 2.2). An electronic modulefunctioned as the on/off switch for the ignition coilcircuit, so the current passing through the coil circuitwas no longer being controlled by a mechanical switch;arcing at the contacts was therefore eliminated.Additionally, higher currents could be passed throughthe electronic module, which enabled more magneticand electrical energy to be created within the ignitioncoil.

With the contact breakers eliminated as a means ofswitching the ignition coil circuit, the spark still had tobe provided at the correct time. The electronic moduletherefore needed a reference or trigger signal so itwould be able to switch off the ignition coil circuit atthe appropriate time, thus creating the spark at thespark plug. Most earlier ignition systems used an‘inductive’ or magnetic sensor, which was locatedwithin the distributor body (effectively in the sameplace as the previously used contact breakers).

The inductive or magnetic sensor (often referred toas a pulse generator) operated in the same way asmodern speed/position sensors or rotational speedsensors (see section 1.5.2), but usually used onereluctor tooth for each cylinder. When each reluctortooth passed the sensor magnet, it caused a smallelectrical pulse to be induced into the coil of wire(adjacent to or wound around the magnet). Theelectrical pulses were used by the electronic module asa reference point for each cylinder, thus allowing theelectronic module to switch the ignition coil circuit atthe appropriate time. The ignition system deliveredthe high voltage to a rotor arm (as was the case withcontact breaker systems), and the rotor arm directedthe voltage to the appropriate spark plug.

Figure 2.3 shows a comparison between a typicalcontact breaker ignition system and an early generationelectronic ignition system, whilst Figure 2.4 shows atypical inductive ignition timing sensor and an earlytype ignition module.

On the early generations of electronic ignition, theautomatic timing advance and retard mechanisms wereidentical to those used on contact breaker systems. Thecentrifugal ‘bob weight’ system was used to advance thetiming with increase in engine speed and the vacuumoperated system was used to retard or ‘back off ’ the


Figure 2.3 Schematic layout of ignition systemsa A contact breaker ignition system b An early generation electronic ignition system

timing when a high load was applied to the engine. Theadvance and retard mechanisms were linked to theinductive sensor base plate and to the rotor shaft (as oncontact breaker systems) enabling relative angularmovement between the inductive sensor magnet andthe reluctor teeth, which affects the triggering timing(and therefore the ignition timing).

2.1.3 Progress in electronic systemcapability

Electronic ignition systems were certainly much morereliable and efficient than contact breaker systems.What we would now regard as simple electronicsallowed considerable improvements to be made in theignition system (as described above) but mechanicaldevices were still relied on to alter the ignition timingwhen engine speed and load changed. Fuel systems,however, even into the 1980s (certainly in Europe)continued to rely on the carburettor as the means bywhich fuel and air were mixed in the correctproportions. Designers had made dramaticimprovements to the capability and accuracy ofcarburettors, by adding various mechanical devicesand some electronic control functions onto the basiccarburettor. The carburettor actually developed into acomplex and often unreliable device. There wastherefore a growing demand to find an alternativemethod of delivering fuel to the engine.

[647] Chapter 02 25/8/06 15:37 Page 40

Emissions legislation in the USA, Europe and manyother countries was becoming increasingly tough forvehicle manufacturers to comply with, but electronicswere developing at a very rapid pace, which enabledmany design changes to be made to ignition and fuelsystems. While the early use of electronics improvedreliability and durability, electronics latterly developedto the level of being able to control the engine systems;this was a fundamental turning point in vehicletechnology.

The remaining sections within this chapter detailmore modern ignition and petrol systems, which aregenerally now integrated as part of an enginemanagement system. Ignition timing is now controlledelectronically instead of mechanically, and fuel systemsare also electronically controlled (via electronic fuelinjection). Although some fuel injection systems usedin the late 1970s (through to the early 1990s) weremechanically based, even these systems were improvedby the use of electronic control. In the end, however,mechanically based ignition or fuel systems wereeffectively no longer able to provide the accuracy andcontrol (at a cost effective price) that are nownecessary to maintain the low emissions levelsdemanded by legislation.

Other benefitsAlthough emissions reductions are often regarded as theonly motivation for using electronics on engine systems,the truth is that there are many other benefits. In

general, when engine efficiency is improved, thisusually results in better fuel economy and higher enginepower, as well as improved engine smoothness andreliability. Using electronically controlled systemsallows engine designers to change certain designfeatures so that there are fewer compromises.

For example, when ignition timing is electronicallycontrolled, compression ratios can be increased to apoint where they are almost at their extreme limits(which improves combustion efficiency). Highcompression can result in combustion knock or pre-ignition in the engine (especially if the fuel quality ispoor); this problem can be accelerated if the ignitiontiming is only very slightly incorrect. Because the oldmechanical timing controls were relatively inaccurate, itwas not possible to risk damage that could be caused byhigh compression. Therefore ignition timing throughoutthe speed and load ranges of an engine were generallyset on the ‘safe side’, slightly retarded from the idealvalue.

However, with electronic control and monitoring,the ignition timing can be more accurate, which reducesthe risk of detonation and knocking, and, if knocksensors detect any combustion knock, then the timingcan be retarded slightly to reduce the problem.

The above is just one instance where engine designcan be improved through the application of electronics.The net result is that engine efficiencies are improvedso that power and economy as well as emissions are allsubstantially better than was the case with older non-

Emissions, reliability and durability 41

Figure 2.4 Inductive pulse generator and ignition modulea Two types of inductive pulse generator that were located in the distributor bodyb Typical appearance of ignition amplifier/module

[647] Chapter 02 25/8/06 15:37 Page 41

electronic systems. Although we do tend to focus onignition and petrol injection, other engine systemshave also been improved through electronic control;examples include variable valve timing and evenengine cooling systems.

Whilst it is often the case that existing mechanicalsystems are improved through electronic control,sometimes totally new systems are introduced whichmay not have been possible with mechanical control. Agood example is electronic diesel injection which is notan evolution of mechanical diesel injection, but a newway of delivering diesel fuel to the engine withaccuracies and control that were not previouslyachievable.

Growing similarities between diesel and petrolsystemsPerhaps it is ironic to note that the modern electronic‘common rail’ type diesel injection systems bear a closerelationship to petrol injection systems, with the use ofelectronic injectors and many sensors that are almostidentical to those of a petrol system. Conversely, somemodern petrol engines make use of direct petrolinjection, whereby the injectors are positioned in thecylinder rather than in the traditional location of the

intake ports. Diesel engines have of course traditionallyalways had direct injection into the cylinder (or moreaccurately: into the combustion chamber).

Whilst the fundamental difference between petroland diesel engines remains the way in which ignitionoccurs (spark for petrol and heat generated from highcompressions for diesel), the systems used for fueldelivery are now very similar.

The continuous pace of developmentMany electronically controlled engine systems arecovered within the rest of this chapter, but such is thepace of development that innovations are introduced ona regular basis. However, if the reader has anunderstanding of the fundamental aspects of engineoperation along with an understanding of electroniccontrol, it is relatively easy to embrace any newdevelopments.

A key driver for emissions has been changes inregulations

Regular maintenance and accurate settings ofignition and fuel systems reduces emissions andimproves fuel consumption

Key

Poin

ts


2.2 ELECTRONIC IGNITION SYSTEMS (EARLY GENERATIONS)

This section deals initially with those ignition systemsthat were not necessarily integrated into enginemanagement systems. However, the design of theseseparate ignition systems incorporates many featuresand components that were then used for ignitioncontrol on engine management systems. By studyingmany of these components and design featuresindependently, it becomes easier to understand thecomplete operation of petrol engine managementsystems, which are covered in section 3.2.

2.2.1 Disadvantages and limitationsof mechanical ignition systems

Those readers who are not totally familiar withmechanically based ignition systems (contact breakersystems with mechanical advance and retardmechanisms) should refer to Hillier’s Fundamentals ofMotor Vehicle Technology Book 1.

There are many disadvantages associated withmechanical ignition systems; the main disadvantagesare:

● the contact breaker mechanism wears (causingincorrect ignition timing and low ignition coiloutput)

● at higher engine speeds, there is insufficient time(ignition dwell time) for the ignition coil to build up

a strong magnetic field thus reducing ignition coiloutput; this is made worse by higher engineoperating speeds

● there is arcing at the contact breaker contact faces(causing reduced ignition coil output)

● maximum current flow passing across the contactbreakers is limited because excessive current willpromote arcing and cause the contact breaker facesto burn away

● ignition timing control is inaccurate, which restrictsthe potential for engines to operate close to theirlimits of efficiency

● at high engine speeds, contact breakers are not ableto open and close quickly and accurately: there is atendency for contact breaker bounce (pointsbounce) to occur (this causes the contact breakers tobounce open before they should do, thus causingincorrect spark timing and reduced coil output).Contact breaker systems are therefore not suited tothe high engine speeds that are now typical for themodern engine

● weaker fuel mixtures can be more easily ignitedwith larger spark plug gaps but this requires moreenergy to be produced by the ignition coil, and suchlevels of energy are not available from contactbreaker systems under all operating conditions; thefollowing sections explain how greater energy levelsare produced.

[647] Chapter 02 25/8/06 15:37 Page 42

All of the above limitations and problems reduce engineefficiency, leading to high emissions, reduced enginepower and higher fuel consumption. It was necessarytherefore to replace the mechanical contact breaker withan electronic switch for the ignition coil primary circuit.

Figure 2.5 is a schematic diagram of a contactbreaker ignition system. Note that, as with almost everydesign of coil ignition system, the primary circuit for theignition coil is switched on the earth circuit, in thisexample by the contact breakers.

2.2.2 Main requirements ofelectronic ignition systems

The fundamental requirements of an ignition systemhave not changed much since the day that ignitionsystems were first used on a petrol engine. The primaryrequirements are to provide a spark or arc at a sparkplug that is strong enough to ignite the air–fuel mixture;this spark must occur at the correct time in theoperating cycle. There have been other requirementsintroduced over the years such as suitable interferencesuppression, higher engine speed operation, etc., butthe basic requirements are much as they always were.

What has changed however is the quality of sparkand the standards of performance (reliability andtiming accuracy). These changes continue through totoday’s ignition systems where they form part of anengine management system. We can therefore look atthe overall requirements of an ignition system at thisstage, and as the reader progresses through this ignitionsection (and the engine management section), it is thenpossible to see how the later generations of ignitionsystem are able to provide improved quality andstandards of performance.

High voltageA fundamental requirement of the ignition system isthat it should produce a sufficiently high voltage fromthe ignition coil at all speeds to enable the air–fuelmixture to be initially ignited under cylinder pressure(compression pressure). The spark or arc produced atthe spark plug must produce sufficient heat to causeignition of the mixture. Many thousands of volts(kilovolts, or kV) are used to create or initiate the spark.

Voltage requirement (firing voltage)A typical voltage requirement for a modern engine is inthe region of 7 kV to 12 kV or slightly higher (assumingall components are good), while on an older engine, therequirements were slightly lower at around 6 kV to10 kV. Note however that, irrespective of how muchvoltage the coil can produce, the voltage delivered bythe coil is dependent on the conditions that exist in thehigh voltage ignition circuit and in the combustionchamber. Remember that electricity will take the easiestpath to earth, so if the circuit from the coil passeddirectly to earth (in effect a short circuit), the energy orvoltage requirement would essentially be zero, becausethis is the easiest route without resistance or barriers toelectrical current flow. If, however, gaps and resistancesexist in the circuit, a higher voltage will be required forthe flow of electricity to reach earth. The major factorsaffecting ignition systems are listed below.

● Plug gap – with a gap in the electrical HT circuit(high tension or high voltage circuit), the energyrequired for the electrical flow to jump the gap andreach earth will be large. The larger the gap, thehigher the voltage requirement; plug gaps aregenerally larger than in the past to assist in ignitingweaker fuel mixtures. It is also true that the pluggap can become fouled or contaminated, which in

Electronic ignition systems (early generations) 43

Figure 2.5 Contact breaker ignition

[647] Chapter 02 25/8/06 15:37 Page 43

some cases will make it more difficult to create aspark, this also increases the voltage requirement.

● Resistance – if we add resistance to the plug leadsand possibly also to the spark plug (to providesuppression), again this will mean that a highervoltage is required. Additionally, as plug leads andother HT components deteriorate, the resistancegenerally increases, which increases the voltagerequirement.

● Cylinder pressures – with modern highcompression engines the high pressure at the pluggap makes it more difficult for electricity to reachearth and a higher voltage is again required. It isalso interesting to note that if the ignition timing(spark timing) is slightly retarded so that the sparkoccurs at TDC instead of slightly before, the cylinderpressures at this time will be higher; this will alsocreate a higher voltage requirement. A worn enginethat has a low cylinder pressure will thereforerequire a lower voltage.

● Air:fuel ratio – although a minor factor, it is alsointeresting to note that it is more difficult for a sparkto be generated across air than it is across vaporisedpetrol. When a spark is created, it ionises themixture. Vaporised petrol ionises much more readilythan air, so it is more difficult to create a spark in aweak mixture; again, a higher voltage is required.

The important thing to remember is that electricity iseffectively ‘lazy’, but it will always try to reach earth,i.e. complete the circuit. If the voltage requirement islow, then the voltage delivered by the ignition coil willbe low. If however, there are resistances and gaps, etc.in the circuit, these will result in a higher voltage beingdelivered by the coil. Because electricity will alwaysattempt to reach earth, even if there are restrictions orresistances, the voltage delivered by the coil willincrease in line with the requirements, until such timeas there is insufficient energy in the coil.

Figure 2.6 shows the output voltage produced by anignition coil during one ignition cycle for one cylinder.Note that ‘firing voltage’ is at a high voltage value,which is necessary to initiate the spark at the spark pluggap (under operating conditions). The rest of thevoltage levels are detailed in the following paragraphs.

Maintaining the spark (spark duration)Since the introduction of emissions regulations, becauseengines generally operate on weaker or leaner mixtures(a greater proportion of air), it is more difficult initiallyto ignite the air/fuel mixture. Additionally, when aweaker mixture is used, flame spread throughout themixture can be less effective at igniting all of themixture. To help overcome these problems, a slightlyhigher initial spark voltage is required, but a spark of


Figure 2.6 Output voltage from an ignition coil through one ignition cycle

[647] Chapter 02 25/8/06 15:37 Page 44

longer duration is also required to maintain thetemperature at the spark plug, thus helping to ignite ormaintain the combustion of all of the mixture.

Modern ignition systems produce a spark that canlast typically around 2 ms (2 milliseconds or 2thousandths of a second) or more, whilst older engines(with relatively rich mixtures) had spark durations ofaround 0.5 ms. The coil needs to produce sufficientenergy or voltage to create the spark initially, but theremust also be sufficient energy available from the coil tomaintain the spark for longer periods than was the casewith older systems that ignited richer mixtures.

The spark voltage and spark duration periods shownin Figure 2.6 illustrate that once the spark has beeninitiated, a much lower voltage is then required tomaintain the spark: once the ionisation process hasbeen started, maintaining that process does not requiresuch a high voltage.

When almost all of the coil energy has been used toinitiate and maintain the spark, the energy remaining inthe coil will be insufficient to keep the spark going;therefore the spark will extinguish. The small amount ofremaining energy then tends to oscillate backwards andforwards within the system until there is no usableenergy remaining.

Coil charge time (dwell period)If the coil is visualised as a container that storeselectrical energy, it is clear that a fully charged coil willbe able to maintain a spark for longer than a partiallycharged coil. Because the coil energy is now used toinitiate and then maintain the spark for longer periods,modern ignition coils must be able to build up energy(charge up) much faster than those on older systems.Current flows through the coil primary winding areincreased by reducing the resistance of the coil primarywinding, which enables faster charging of the coil.Additionally, the charge up period (previously referredto as the ignition dwell period) is controlled onelectronic systems such that there is a longer timeavailable to build up the coil energy.

As well as ensuring the coil provides enough energyto initiate and maintain the spark for longer, higherengine speeds add another demand because they meanless time for the coil to build up energy, so ignitionsystems must be able to charge the coil (build up the coilenergy) in even less time.

In Figure 2.6, the indicated dwell period is theperiod when current flows through the primarywinding, allowing energy to be built up within the coil.A switch located in the primary circuit is used to switchon and off the flow of current. On old systems thisswitch was the contact breaker, and when the contactbreakers closed, this would complete the circuit andallow current to flow. On modern systems, the contactbreakers are replaced by an electronic switch, which isusually part of an amplifier or module (or in manycases it is incorporated into the engine managementECU).

On older type contact breaker ignition systems it wouldtake as long as 10 ms for the current to reach itsmaximum flow rate (typically 4 A maximum). As anexample, on an old type four-cylinder engine operatingat 6000 rev/min (Figure 2.7), the whole ignition cyclefor a cylinder (the time between the spark on onecylinder to the spark on the next cylinder) would takeonly 5 ms (5 thousandths of a second). However, it isnot possible to allocate the whole of the ignition cycleto building up coil energy because some of that timemust be available for the coil to deliver the energy toproduce the spark. Typically, around half of the totalignition cycle was used for building up coil energy; inour example, this would mean 2.5 ms (2.5 thousandthsof a second). If it takes 10 ms to reach maximumcurrent flow in the coil primary circuit, then at highengine speeds there would only be around one-quarterof that time available, which would restrict coil energy.

On an engine operating on relatively rich mixtures,the low coil energy was not so much of a problem buton modern high speed engines with weaker mixtures,the coil energy must be higher than on older ignitionsystems. Modern coils operating with modernelectronically controlled ignition systems can build upsufficient energy in around 3 or 4 ms or less. Althoughthe coils may not be fully charged at high enginespeeds, the charge level is sufficient. However, manymodern engines now use one coil for each cylinder,which on a four-cylinder engine means that a coil hasfour times as long to build up the energy as enginesusing a single coil for all cylinders. This is especiallyimportant on an engine with six, eight, ten or 12cylinders, because the more cylinders there are, the lesstime is available between ignition cycles. Remember, ona four-cylinder engine, there are two sparks for everycrankshaft revolution. On a 12-cylinder engine there aresix sparks for every crankshaft revolution, which meansthere is only one third of the time available compared toa four-cylinder engine.

Dwell angles and dwell percentageOn old ignition systems that used contact breakers toswitch on and off the primary current the contactbreakers were set so that they were closed for typicallyaround 50% to 60% of the ignition cycle, i.e. the dwellperiod. On an old type four-cylinder engine with contactbreakers and a distributor, the distributor shaft wouldrotate through 90° for each ignition cycle (for onecylinder). A dwell period of 50% would equate to 45° ofdistributor rotation. It was therefore common in the pastto quote the charge-up time as being the ‘dwell angle’.

When contact breakers were fitted and adjusted, theobjective was to achieve the specified dwell angle; byadjusting the position of the contact breaker within thedistributor, the correct dwell angle could be obtained.Because the mechanical setting and operation of thecontact breakers dictated this percentage, it would notchange throughout the speed range of the engine. If thepercentage were any larger, at slow engine speeds, the


[647] Chapter 02 25/8/06 15:37 Page 45

current flow in the primary circuit would be too longand the circuit and coil would overheat. If thepercentage were any smaller, at high engine speeds therewould not be enough time to charge up the ignition coil.The percentage used was therefore a compromise.

However, once the correct dwell angle was set, itwould not change with engine speed, so when theengine speed increased, the whole ignition cycle had totake place within less time. Therefore there was lesstime to charge up the coil.

Modern electronically based ignition systems areable to control the dwell period. In effect, the dwelltime is controlled so that it is ideally the same durationat all engine speeds. Therefore, if the ignition coil takesonly 2 ms to build up the required energy, then thedwell time would be 2 ms at all engine speeds. Thereare situations where at high engine speeds there maynot be quite enough time available but a number ofother design changes have overcome this problem (asdiscussed in the following sections).

Spark timingThe spark must occur at the correct time, at all enginespeeds and loads. When the mixture is initially ignited, itmust have sufficient time to combust and for the gasesto begin to expand before the piston has moved too fardown the cylinder on the power stroke. Ideally, themixture should initially be ignited just before the pistonreaches TDC on the compression stroke; this shouldallow for the mixture to begin to combust and then for

the heat to cause the gases to start to expand just as thepiston reaches TDC. The subsequent expansion of thegases then forces the piston back down the cylinder.

If the spark occurs too soon (with over-advancedtiming), the gases will start to expand before the pistonreaches TDC and the expansion will try to force thepiston down the cylinder before it has reached TDC,which means that some of the energy created by the gasexpansion will be pushing against the rising piston; thiscan result in combustion knock occurring because of thepremature ignition timing.

If the spark occurs too late (with over-retardedtiming), the expansion of gases will be late and thepiston may be already on its way down the cylinder (onthe power stroke). The effect of the gas expansion willtherefore be wasted.

Speed related timing advanceIf we accept that it takes a certain period of time for thefuel to ignite and then combust or burn sufficiently tocreate the heat and expansion of the gases, then ingeneral this time period will theoretically not changesignificantly when the engine speed changes (assumingall other conditions remain the same). The burn timedoes in fact change with engine speed, but for thefollowing example, it is assumed that the time remainsconstant.

If we assume that the time allowed for the burningprocess to take place and create the maximum pressurein the cylinder is around 4 ms (4 thousandths of a


Figure 2.7 The charge up time (dwell period) on a four-cylinder engine

[647] Chapter 02 25/8/06 15:37 Page 46

second), then the spark must occur 4 ms before themaximum pressure is required. In reality, the maximumpressure created by the gas expansion is usuallyrequired just after the piston has passed TDC, i.e. it isjust beginning the downward stroke.

If in the following example (see Figure 2.8a) we assumea burn time of 4 ms, then at 1000 rev/min thecrankshaft would rotate through 24° during the 4 msburn time. If we also then assume that maximum gaspressure should occur at 14° of crankshaft rotation afterTDC, then the spark should occur at 10° before TDC.

If the engine speed is increased to 2000 rev/min, i.e.doubled, 4 ms will still be needed for the burn time(Figure 2.8b). Because the crankshaft is now rotating attwice the previous speed, it will rotate through 48°during the 4 ms burn time. Therefore, the spark mustoccur 48° before the maximum pressure is required(which remains at 14° after TDC). The spark musttherefore occur at 34° before TDC.

Further increases in speed would therefore alsorequire additional advances in the ignition timing, butbecause the burn time does in fact not remain constant(conditions within the cylinder change with speed), theamount of additional advance required for increases inengine speed gradually reduces. Figure 2.8c shows anapproximate advance curve related to engine speed.Note that different engine designs and combustionchamber designs will have different advancecharacteristics, so the advance curve illustrated shows atrend, rather than indicating exact values of sparktiming advance.

As previously mentioned, older type mechanicaladvance mechanisms are not accurate enough toprovide the exact advance curve needed for modernengines. Therefore electronic systems now control thetiming advance process as described later in section 2.3.

Load related timing advance/retardWhen an engine is operated at light load, it is possible tooperate on weaker mixtures than when the engine isoperated under high load conditions. On modernengines, the mixture is controlled by the enginemanagement system using the oxygen sensormonitoring process (as discussed in sections 1.5.6 and3.2), and there is not so much change between themixture settings for light and heavy load conditions(compared with older engines). However, when weakermixtures are used for light load conditions this causes arequirement for a different ignition or spark timing.

A weaker mixture takes longer to burn, so the sparktiming will require additional advance. Additionally,under light load conditions, the throttle valve(butterfly) is only partially open, thus restricting theflow of intake air; this means that a lower volume of airis drawn into the cylinder, and during the compressionstroke, the cylinder pressures and temperatures arelower, which also has an effect on the burning process.Under these conditions, it is therefore necessary toadvance the timing slightly to account for the slowerburn time.

When the engine is again placed under load and themixture is no longer weak (cylinder pressures will alsobe higher), the amount of timing advance can bereduced back to the load setting.


Figure 2.8 Ignition timing advance related to engine speeda Burn time takes 4 ms. At 1000 rev/min, 4 ms gives 24° ofcrankshaft rotation. If the maximum pressure must occur 14° afterTDC, then the spark must occur 10° before TDC.b Burn time takes 4 ms. At 2000 rev/min, 4 ms gives 48° ofcrankshaft rotation. If the maximum pressure must occur 14° afterTDC, then the spark must occur 34° before TDC.c Typical ignition advance requirements

[647] Chapter 02 25/8/06 15:37 Page 47

On older ignition systems with mechanical advancemechanisms, the speed related timing advanceprovided the main advance setting and when theengine was under higher loads, a vacuum operatedsystem would retard the timing; this type of loaddependent system was referred to as a timing retardsystem. In effect there was a speed related advancesystem and a load related retard system.

Incorrect timingIf the ignition timing is incorrect, as previouslymentioned, it can cause two problems: first, over-advanced timing can cause combustion knock: thepremature expansion of the gases will be wastedbecause the piston is still rising on the compressionstroke when the gases are expanding; second, over-retarded timing will result in the maximum pressureoccurring when the piston has already travelled too fardown the cylinder: therefore the expansion of the gaswill again be wasted. In effect, both over-advanced andover-retarded timing will cause a reduction in power.

The ignition timing can also affect the emissions;over-advanced timing can cause incomplete combustionof the mixture which will result in high levels ofunburned or partially burned fuel entering the exhaustgas. In addition, higher temperatures will be created inthe cylinder (the pressure rise will be higher), whichcan result in increases of oxides of nitrogen (see section3.5). However, over-retarded timing can in some caseshelp to reduce emissions, although this might also causepoor combustion: some older emission control systemsused retarded timing under certain operating conditionsto ensure that combustion continued later in the enginecycle, helping to burn some of the partially burnedexhaust gases.

Interference suppressionIgnition systems create considerable electrical noise;this is the effect of providing a high voltage to a sparkplug, which causes radio frequency energy around thewires carrying the high voltage. Suppression is achievedby using resistances built into the high tension (HT)cable and/or the spark plugs and rotor arm (wherefitted on older systems). Apart from the fact that theinterference causes electrical noise (crackles, etc.) onthe vehicle radio, television and other audiovisualequipment can be affected (even if it is someconsiderable distance from the vehicle). Legislationlimits the levels of interference.

2.2.3 Electronic switching of the coilprimary circuit

The coil primary circuit carries a relatively highcurrent for an extremely short period of time. On acontact breaker system, the resistance of the coilprimary winding was typically around 3 ohms, whichwould mean the current flow in the 12 volt primarycircuit would reach a maximum of 4 A (which was the

maximum that could be used on a contact breakersystem without causing excessive arcing anddamage). On older slow running engines, this currentlevel was sufficient to enable the coil to build up areasonably strong magnetic field and thereforeproduce an acceptable output voltage. Note, however,that fuel mixtures were relatively rich (containing anexcess of petrol) which enabled the mixture to beignited with a relatively weak spark. However, if astronger spark is required, this can be generated byeither increasing the current flow in the ignition coilprimary circuit (lowering the resistance of the coilprimary winding) or maintaining the current flow inthe primary circuit for longer periods (a longer dwelltime). Both options would result in accelerateddamage to the contact breakers. In addition, higherengine speeds and multi-cylinder engines reduce theavailable time for increasing the dwell period (thecoil charge time).

Using an electronic switch instead of a mechanicalswitch provides a number of benefits. One is improvedreliability because there are no moving parts and noarcing at the contacts, but an electronic switch alsoenables higher current flows to exist for longer periods.The simple type electronic switch, which is a powertransistor, simply switches on and off the coil primarycircuit at the appropriate time, but note that thetransistor will perform the switching task only when anappropriate electrical trigger signal is provided.

Figure 2.9 shows a simplified example of how atransistor functions as a switch by comparing atransistor to a water valve. Note that the main waterflow from the collection point C cannot flow throughthe valve and be emitted at E when the valve is closed.However, if a small flow of water is allowed to pass intothe base of the valve B this will cause the valve to openand allow the main water flow to pass from C to E.

For the transistor, the principle is much the same.The main electrical circuit cannot flow through thetransistor from the collector to the emitter (C to E) untila small voltage or current is applied at the base B. It istherefore possible to provide a small electrical signal atB (a low current or voltage) to control a much highercurrent and voltage, which are passing through thetransistor from C to E.

See Hillier’s Fundamentals of Motor VehicleTechnology Book 3 for a more detailed explanation ofhow transistors operate.

2.2.4 Electronically assisted ignitionSome early generations of electronic ignition systemused a contact breaker to provide the small electricsignal to the transistor. Figure 2.10 shows a simplifiedcircuit where the transistor is the switch for the ignitioncoil primary circuit. In this example, however, thecontact breaker is used only as a means of switching onand off a low current signal to the transistor. Therefore


[647] Chapter 02 25/8/06 15:37 Page 48


Figure 2.9 Principle of a transistor using a water valve as a comparisona A water valve closed and openb A transistor switching on and off

Figure 2.10 Using a contact breaker to control the switching of a transistor in an ignition circuit

the contact breaker can be operated in exactly the sameway as on a traditional contact breaker ignition system,but the contact breakers do not carry the high current. Aresistor is used in this example in the contact breakercircuit to reduce the current flow passing to the B orbase terminal of the transistor.

Where a simple transistor is used to switch the coilprimary circuit, it is often referred to as the ‘ignitionmodule’ or ‘ignition amplifier’, although in truth thetransistor is not strictly speaking amplifying the trigger

signal but merely switching the coil primary circuit inresponse to a trigger signal.

The contact breaker triggered system shown inFigure 2.10 would eliminate the high currents passingthrough the contact breaker, and higher currents couldthen be allowed to flow in the ignition coil primarycircuit. However, the accuracy of the ignition timing isstill dependent on the contact breaker, which would stillsuffer from mechanical wear. A non-mechanical triggermechanism is therefore required.

[647] Chapter 02 25/8/06 15:37 Page 49

There are a number of ways that an electrical signal canbe created that will act as a means of triggering thepower transistor in an ignition coil primary circuit. Asexplained above, a contact breaker can be used toswitch on and off a trigger circuit, but more accurateand reliable methods are generally non-mechanical.The following section covers the most common types ofnon-mechanical trigger mechanism that were used forelectronic ignition systems.

The vast majority of early generation electronicignition systems used either an inductive or Hall effecttrigger system, although a few systems did use anoptical trigger. Although we can refer to these systemsas being non-mechanical, mechanical components areused, but the important fact is that the trigger signalpassed to the ignition module is created by a non-mechanical process, i.e. there is no mechanical switch.

A number of constructional and design variationsexisted for all three trigger systems, but for each of theexamples mentioned in the following section, thegeneral operating processes form the basis for othervariants of each type.

2.2.5 Inductive ignition triggerInductive pulse generatorFigures 2.11 and 2.12 show two different types ofinductive trigger or pulse generator. Both types werelocated in the ignition distributor body and as suchphysically replaced those components originally usedfor contact breakers. On earlier generations ofelectronic ignition, both types were also connected tothe mechanical/vacuum advance mechanisms(inherited from the contact breaker systems).

Both examples operate on the same principle ofusing a reluctor with the same number of reluctor teeth(or triggering lugs) as the number of cylinders, i.e. fourreluctor teeth for four cylinders. A permanent magnetand inductive coil (coil of wire) are located adjacent tothe reluctor. When the reluctor is rotating, each of the

reluctor teeth passes the magnet and inductive coilassembly, which will produce a small electrical pulsedsignal. This electrical signal is an analogue signal. Thepulsed signal is then passed to the ignition module(power or switching transistor), which uses the pulsesas a trigger or reference point to switch the ignition coilprimary circuit.

The example in Figure 2.11 shows a magnet andinductive coil assembly located to one side of thereluctor. The iron reluctor is mounted on the distributorshaft and therefore rotates with the shaft. The exampleshown in Figure 2.12 operates in much the same way asthe example in Figure 2.11, but the construction isslightly different. The magnet is a circular disc which islocated concentrically with the distributor shaft; theinductive coil is also concentric with the magnet. Thestator, or pole for the magnet, consists of fingers (onefor each cylinder) which protrude upwards. A rotor orreluctor, which also has one reluctor tooth for eachcylinder, is located on the distributor shaft; the reluctorteeth are formed as fingers that protrude downwardsand pass adjacent to the stator fingers.

In both examples, when the reluctor teeth or fingerspass the stator or stator fingers, this causes an electricalsignal to be produced as explained below.

Generating the pulseWhen a reluctor tooth is aligned with the permanentmagnet or stator (as shown in Figure 2.11), it allows themagnetic flux to flow from the stator, across the reluctorand back again. When the distributor shaft is rotating,the reluctor teeth will inevitably move away from thestator, providing a gap between the reluctor teeth andthe stator. This gap results in a greater reluctance of themagnetic field or magnetic flux, i.e. the flow will be less.

In effect, when the reluctor teeth approach thestator, the flow of magnetic flux will increase. The flowof magnetic flux will be at its maximum when the teethand stator are in alignment, and it will reduce when thereluctor teeth move away from the stator. When theflow of magnetic flux changes, i.e. increases or


Figure 2.11 Inductive pulse generator with the magnet and inductive coil located at the side of the reluctor

[647] Chapter 02 25/8/06 15:37 Page 50

decreases (owing to rotation of the reluctor), this causesa small electrical current to be produced in theinductive coil. The voltage generated is at its greatestwhen the change in flux flow is at its greatest; thisoccurs just as the reluctor teeth are approaching orleaving alignment with the stator.

Figure 2.13a shows the voltage output from theinductive sensor, or variable reluctance sensor as it issometimes called. There are four electrical pulses (for afour-cylinder engine) produced during one completerotation of the reluctor. A positive voltage is producedas the reluctor teeth approach alignment with thestator (magnet); as the reluctor teeth leave alignmentwith the stator, a negative voltage is produced. Theoutput signal is therefore an analogue alternatingcurrent (AC).

As described above, when the reluctor teethapproach or leave the stator (Figure 2.13b) this causes alarge change in magnetic flux, which therefore produceshigher voltage (positive or negative). However, whenthe reluctor teeth are close to alignment and inalignment with the stator, the result is very little or nochange in the flux, which means that less voltage isproduced. When the reluctor is directly in alignmentwith the stator, there is no change in magnetic flux:therefore the voltage produced is zero.

Note: The gap between the reluctor teeth and the statoris effectively set during manufacture. However, on sometypes of construction it is possible to alter the gap. If thegap is not correct, this will affect the magnetic flux andthe strength of the signal produced. Reference shouldalways be made to manufacturer’s specifications.

Reference point for ignition timingIt is normal practice to use the change or ‘switch over’from positive to negative voltage, i.e. the zero voltagepoint, as the reference point for ignition timing. Theignition module will therefore use this ‘zero volt’ pointof the electrical signal as the reference to switch off theignition coil primary circuit, thus creating the highvoltage and a spark at the plug.


Figure 2.12 Inductive pulse generator with the magnet and inductive coil located concentrically around the reluctor

Figure 2.13 Rotation of the reluctora One rotation of the reluctor produces four pulses as ananalogue signalb Voltage levels produced when the reluctor teeth are indifferent positions relative to the magnet (stator)

[647] Chapter 02 25/8/06 15:37 Page 51

The ignition coil will produce its high voltage outputwhen each reluctor tooth is aligned with the stator: on afour-cylinder engine there will be four high voltageoutputs from the coil for every rotation of thedistributor shaft. The output from the coil musttherefore be distributed to the appropriate spark plugsat the correct time (when each of the cylinders is closeto TDC on the compression stroke); this is achieved bypassing the high voltage from the ignition coil to thecentre of a rotor arm located within the distributor.When the distributor shaft and rotor arm are rotating,the rotor arm will direct the high voltage to the contactsegments in the distributor cap, which allows thevoltage to pass to each of the spark plug leads and sparkplugs in turn (Figure 2.14). It is therefore important tonote the exact location of each spark plug lead on thedistributor cap to ensure that the voltage is directed tothe correct spark plug at the correct time.

Wiring circuit for an inductive pulse generatorInductive pulse generators generally have two wiringconnections to the ignition module. Effectively, thesetwo wires provide a positive and a negative path for theelectric current. However, remember that the current isan alternating current which means that the flowalternates within the wiring; each wire thereforealternately carries positive and negative flows.Figure 2.15 shows the wiring for a typical inductivesensor and ignition module.

The wiring diagram (Figure 2.15) shows two wirescarrying the pulsed signal from the inductive pulsegenerator to the ignition module. Because the moduleforms part of the earth circuit for the ignition coilprimary circuit, the power supply from the ignitionswitch passes to the coil positive terminal (usuallymarked terminal 15) and then through the coil primarywinding to the ignition module. The module functionsas the switch for the primary circuit; therefore thecircuit must pass through the power or switchingtransistor in the module before it is connected to earth.If the ignition module contained only simple passiveelectronics, no power supply would be required for themodule, but it contains active electronic componentsthat require an additional power supply and earthconnections.

On some applications, a third wire has been usedwhich is wrapped around the two signal wires. Thethird wire is connected to earth or ground, acting as ascreen or shield against interference from otherelectrical systems. Also note that when only two wiresare used, one of the wires may be wrapped around theother, which provides a form of screening.

The inductive sensors can usually be classified asself-generating, which means that no additional power


Figure 2.14 Rotor arm and distributor cap (allowing a highvoltage to be directed to the correct spark plug)a Distributor capb Rotor Figure 2.15 Wiring for a simple inductive trigger ignition system

(a)

(b)

[647] Chapter 02 25/8/06 15:37 Page 52

supply is required to enable the pulsed signal to beproduced. Therefore the two wiring connections simplyprovide a complete circuit for the inductive coil. Thereare however some examples where an electric current isused to enhance or create the magnetic field; in suchcases the same wiring that provides the current to thesensor also carries the pulsed signal.

Advantages and disadvantages of inductive pulsegeneratorsInductive pulse generators are relatively inexpensive toproduce and generally reliable and accurate, even whenworking in the harsh environment of a vehicle’s engine.Inductive pulse generators produce an analogue signal,which, when passed to an ECU that is operating withdigital electronics, will require an analogue to digitalconverter (A/D converter). When the analogue signal issupplied to the early types of ignition modules (a simplepower transistor switch), some simple circuitry withinthe ignition module is used to reshape the pulse so thatthe applicable reference points are recognised, i.e. thezero volt timing reference point.

The inductive pulse sensor operates using the sameprinciples as a conventional electrical generator and, asis the case with a conventional generator, the voltageproduced increases with the increase in rotational speedof the rotor (or reluctor). Therefore, when the enginespeed increases, the voltage produced by the inductivepulse generator also increases, ranging from around0.5 volts at slow speeds to a possible 100 volts at highspeeds. If, therefore, there is any deterioration in thestrength of the magnetic flux, and the engine is turningover very slowly during starting, it is possible that nosignal will be produced.

Also of note is that, on the example shown in Figure2.11, where the magnet and inductive coil assembly arelocated to one side, it is possible for an erratic orunusable signal to be produced if wear exists in thedistributor shaft bearings. If a distributor shaft bearingis worn, the shaft can wobble during rotation; this inturn can result in the air gap between the reluctor teethand the stator changing. The usual problemencountered is that the air gap for one reluctor tooth istoo small whilst the gap for the opposite tooth is toolarge. It is not uncommon in these cases for the pulsesignal to be missing a trigger pulse for one andsometimes two cylinders. This will mean that one ortwo cylinders may not receive a spark at the spark plug.

2.2.6 Hall effect ignition triggerAs mentioned previously, when a sensor produces ananalogue signal (such as the inductive ignition trigger),if the signal is then passed to an ECU that operates withdigital electronics, an analogue to digital converter isrequired to enable the ECU to interpret the signal. It istherefore an advantage if the sensor is able to provide adigital signal.

A digital signal also has other advantages relating to thevery defined reference points that can be provided. Inthe previous section (2.2.5) it was stated that theanalogue signal produced by the inductive sensorwould change with engine speed: the voltage producedby the sensor increases with increases in engine speed.In fact, the whole ‘shape’ of the signal changes.However, with a digital signal such as the signalproduced by a Hall effect sensor, the reference pointsare consistent, irrespective of engine speed.

Hall type systems are often referred to as Hall effectswitches or Hall effect pulse generators. Figure 2.16ashows a typical construction of a Hall effect ignitiontrigger located in an ignition distributor; Figure 2.16bshows a separate view of the Hall trigger assembly.


Figure 2.16 Hall effect pulse generatorThe rotor has four vanes which causes four pulses to beproduced (a digital signal) during one rotation of the rotora A Hall effect ignition trigger in an ignition distributor assemblyb A Hall effect pulse generator

[647] Chapter 02 25/8/06 15:37 Page 53

HTconnection

High tension

lead

Coil lowtension

leadHall effectignition module

Hall effectpower and signal

Earthconnection

Distributorwith Hall effect

trigger unit(4 cylinder)

12 Vvia

ignition switch

PrimarywindingSecondary

winding

Ignitioncoil

1541

+

–


Hall voltage

A – switch on – no vane in gap

B – switch off – vane in gap

0 Time

Vane:gap ratio = 70:30(Bosch)

Figure 2.17 Digital signal produced by a Hall effect pulsegenerator

For most Hall effect systems, there will be the samenumber of vanes as there are engine cylinders. Thismeans that on a four-cylinder engine with a singleignition coil, there will be four vanes, and the ignitioncoil will therefore produce four high voltage outputs forevery rotation of the Hall effect rotor/trigger disc. Aswith the inductive ignition trigger system, it is thereforenecessary to pass the high voltage coil output to a rotorarm in the distributor cap, which will then direct thevoltage to the four spark plugs in turn.

Wiring for Hall effect pulse generatorFigure 2.18 shows the wiring for a simple Hall effecttriggered ignition system. The Hall effect pulsegenerator initially requires two wiring connections forthe input (a power supply and earth connection) toenable it to function; Figure 2.16b shows the circuitpassing across the Hall chip terminals A and B.However, the signal produced at the Hall chip shouldalso have two connections (positive and negativepaths), but the negative or earth path is shared with theearth path of the input, which means that there is atotal of three connections between the Hall effectsensor assembly and the ignition module.

Note that it is common practice to mark theterminals on the Hall sensor connector plug with threesymbols: +, –, and 0. The + terminal is the powersupply (often stabilised at 5 or 8 volts), the – terminalis the earth terminal and the 0 terminal is the outputterminal for the digital signal.

The ignition module will require a power supply andearth connection, and the module again forms part ofthe earth circuit for the ignition coil primary circuit.

2.2.7 Optical ignition triggerOptical ignition triggers have been used on a number ofignition systems and also for some other applicationswhere a digital signal is preferred to an analogue signal;

Figure 2.18 Wiring for a simple Hall effect triggered ignitionsystem

Hall effect digital pulseThe construction and operation of the Hall effect sensoris described in Chapter 1. The signal produced is asquare wave: Figure 2.17 shows a typical digital signalproduced by a Hall effect ignition trigger. Here, thereare four pulses produced during one rotation of therotor, which would indicate that the rotor has fourvanes and would be used on a four-cylinder engine.

With a digital signal, it is possible to make use of atleast two definitive reference points on the signal. Onereference point is when the voltage increases from zeroto 5 volts (or whatever voltage is used on the sensor).The second option is when the voltage drops from5 volts down to zero. Therefore, either the rise or thefall in voltage can be used as the reference point for theignition module to switch off the ignition coil primarycircuit. When working on Hall effect systems, it istherefore necessary to refer to the manufacturer’sinstructions to find out whether the ignition timingpoint occurs when a vane is just leaving the air gap(between the magnet and the Hall chip), or when thevane is just entering the air gap, because this willdictate whether the voltage is rising or falling.

[647] Chapter 02 25/8/06 15:37 Page 54

the optical system provides an alternative to the Halleffect system.

Producing the optical trigger digital pulseA light emitting diode (LED) produces a small lightwhen an electric current is passed through it;conversely a phototransistor produces a small electriccurrent when it is exposed to light. Therefore, if an LEDis used to project light onto a phototransistor, thephototransistor will produce an electric current.

As illustrated in Figure 2.19, an LED andphototransistor can be located, as an assembly, insidethe distributor body. If a ‘chopper disc’ is located on thedistributor shaft so that the vanes or shutters of thechopper disc pass between the LED and phototransistor,the shutters will prevent the light from the LED fromreaching the phototransistor. If the chopper disc hasfour shutters (as shown in the illustration), then whenthe disc rotates (with the distributor shaft), each time ashutter blocks the light from the LED then thephototransistor will not produce an electric current.However, when the gaps between the shutters are inline with the LED and phototransistor, the light willreach the phototransistor thus producing the electriccurrent. The four shutters and gaps would result in fouron/off pulses of current being produced by thephototransistor in one rotation of the distributor shaftand shutter disc; this version would therefore be usedon a four-cylinder engine.

The optical system produces a digital pulse (Figure2.20), which can be at a lower voltage than thatprovided by the Hall system (sometimes as low as2.4 volts). However, the signal from the optical systemis sufficient for the ignition module to identify the risein voltage or the fall in voltage as the signal alternatesfrom on to off.

Wiring for optical ignition triggerThe LED requires a power supply and earth connections(Figure 2.21). The power supply will usually be passedfrom the ignition module to the LED. Note that thevoltage will be stabilised by components within themodule. The phototransistor will also require twoconnections to the module (positive and negativepaths) to carry the signal produced by thephototransistor. There will therefore be fourconnections between the optical sensor and the ignitionmodule. The module will also require its own powersupply and earth connections, and, as with inductive orHall systems, the module will form part of the earthcircuit for the ignition coil.

Advantages and disadvantages of an opticalsystemThe optical ignition trigger system provides a truedigital signal, which is especially useful when the signalis passed to an ECU that operates with digitalelectronics. Optical systems are also generally reliableand relatively inexpensive.


Figure 2.19 Optical ignition trigger assembly

V signal

+2 V

+0.2 V

1 pulse/cylinder(4 cylinder)

0 360° crank angle

Figure 2.20 Digital signal produced by an optical ignition trigger

[647] Chapter 02 25/8/06 15:37 Page 55

The main disadvantage with an optical system is theimportance of keeping the LED and phototransistorclean. A build up of dirt or oil on the components wouldreduce the efficiency or prevent the system fromfunctioning.

2.2.8 Ignition modules/amplifiers:early types

With earlier types of electronic ignition, the ignitionmodule simply functioned as the electronic switch forthe ignition coil primary circuit. Assuming that a triggermechanism such as an inductive trigger (as described insection 2.2.5) was used as a means of providing thetiming reference signal, the module simply switched theprimary circuit on and off thus forming a non-mechanical circuit switch. There were several variationsof early systems but the basic objective was to eliminateany mechanical switching, and this was achieved withan inductive trigger (or Hall effect type) and atransistor that acted as the primary circuit switch.

Ballast resistor and dwell periodEarly generations of electronic ignition systems, oftenhad a ballast resistor in the primary circuit (as did manycontact breaker systems). The resistance of the ballastresistor altered with temperature, and the temperaturechange was dictated by the length of time that currentflowed in the primary circuit. As with a contact breakersystem, the dwell period was a fixed percentage of theignition cycle, e.g. 60%. Therefore, at slow enginespeeds when an ignition cycle took a relatively longtime, the current flowed through the primary circuit for60% of this period, causing the resistor temperatureand its resistance to increase, thus reducing the current

flow. However, when the engine speed increased, thetime available for the current flow reduced thusreducing the temperature of the resistor; the resistanceof the ballast resistor therefore also reduced thusincreasing the current flow in the primary circuit.

The action of the ballast resistor therefore allowed ahigh current to flow in the primary circuit when therewas reduced time available at high engine speeds but,when the engine speed was low, the current flow wasreduced to prevent overheating of the coil and wiring(i.e. it is an output control ballast). The result was thatthe ignition coil could build up to an acceptable energylevel at high as well as low speeds.

The ignition coils would have a primary windingresistance of typically 1.5 ohms (or slightly less), whichwould allow for a rapid build up of current flow. Theballast resistor would also have a resistance ofapproximately 1.5 ohms, so the total resistance in thecircuit was approximately 3 ohms, which, with the12 volt supply, would result in a current of 4 A. Whenthe resistance of the ballast resistor increased at lowengine speeds, this would then reduce the current flow.For many systems, the dwell percentage was often morethan 60%, which meant that there was a longer periodfor charging the coil. So long as the ballast resistorfunctioned correctly, then the system would notoverheat at low speeds.

Dwell periods on the early systems were generally afixed percentage (fixed dwell angle). In the same waythat a reference point on the trigger signal was used asa reference to the timing point, it was also possible touse another reference point on the signal as a referenceto starting the dwell period. Some systems might haveused an electronic device within the ignition module tocontrol the dwell period.


HTconnection

HTlead

Coil LTlead

Optical ignitiontracer module

Supply wiresto LED

Signal wires fromphototransistor

Earth

12 Vignition switch

Primaryand

secondarywindings

Ignitioncoil

1541

Figure 2.21 Wiring for an optical ignition trigger

[647] Chapter 02 25/8/06 15:37 Page 56

Ignition timing advance and retardWith early generations of electronic ignition, the ignitiontiming advance and retard mechanisms remained muchthe same as for a contact breaker system: mechanicaladvance mechanism to alter the timing with enginespeed and a vacuum operated retard system to alter thetiming with changes in engine load. See section 2.2 inthis book and section 2.28.8 in Hillier’s Fundamentals ofMotor Vehicle Technology Book 1.

2.2.9 Ignition modules: later typeswith dwell control andconstant energy function

Improving the coil output at all engine speedsOne of the main problems with early electronic moduleswas that the dwell period was a compromise (as wasthe case with contact breaker systems); this meant thatthe dwell period was too long at slow engine speedsand too short at high engine speeds (hence the use of aballast resistor as described in the previous section).

The next generation of ignition modules thereforeprovided a facility to alter the percentage of dwelldepending on the engine speed. In effect, when theengine was at low speeds and an ignition cycle lasted arelatively long time, the dwell percentage was a smallpercentage of the long ignition cycle time. When theengine was at high speeds and the cycle time was muchshorter, the dwell percentage was increased.

Example of dwell controlAs an example, to achieve a good quality spark, if weassume that a current that flows for 2.5 ms is needed toallow the coil to build up the required amount of energy(magnetic field strength), then this in theory would bethe same irrespective of engine speed. At 1000 rev/minon a four-cylinder engine, the ignition cycle for onecylinder would last for 30 ms, so the required 2.5 mswould represent one-twelfth of this period i.e. 8.33%.

If the engine speed is then increased to2000 rev/min (twice the speed), then one ignition cyclewill now last for only 15 ms (half the time). However,the coil would still require 2.5 ms of charge up time,which represent one sixth of the total 30 ms i.e.16.66%. In effect, the charge up time remains the samebut the percentage of the whole cycle changes inproportion with the change in engine speed.

As a last part of the example, if the engine speed isnow increased to 6000 rev/min, the ignition cycle forone cylinder will last for only 5 ms. The coil charge uptime will remain at 2.5 ms, which now represents 50%of the ignition cycle.

It is therefore possible with this type of dwell timecontrol to operate an engine at high engine speeds andstill provide a long enough dwell period for the coil tobuild up strong energy levels.

The actual control of the dwell period is notnecessarily as precise as in the explanation above, and

there are several variations in the exact dwell timeprovided depending on the ignition system moduledesign. However, the objective is to ensure that thedwell time is sufficient for all engine speeds, thusallowing the ignition coil to provide a reasonablyconsistent output at all speeds.

For engines with more than four cylinders, the dwelltime will have to be slightly less because of the shortertime available for one ignition cycle for each cylinder.However, on more modern ignition systems,developments have included one coil to provide a sparkfor two cylinders and more recently, systems nowprovide one coil for each cylinder. In both cases, thetime available for each coil to charge up is considerablyincreased. These systems are explained in greater detaillater in this section.

Controlling dwell for specific conditionsAlthough the system can control the dwell to suitengine speed, there are some operating conditionswhere it is an advantage to enable the coil to providegreater energy levels than normal. The usual examplesare to provide a slight increase in dwell time at startingand at low engine speeds. Starting inevitably requires astrong spark, and at idle speed where emissions arecritical, a better quality spark helps to ensure improvedcombustion. This is especially true with enginesoperating on relatively weak mixtures, which thenbenefit from long spark durations.

Constant energy control and high energy coilsIt is obviously an advantage to use an ignition coil thathas a rapid build or charge up time, and this can beachieved by using a coil with a low primary windingresistance (low inductance). Many modern ignitioncoils have a primary winding resistance that is as low as0.5 ohms, which is one-sixth of the resistance of oldercontact breaker system coils. Therefore, the potentialcurrent flow through the primary winding on a moderncoil could be as high as 24 A (12 volts through aresistance of 0.5 ohms). In fact, such potentially highcurrent levels would be too high and could damage thewiring and the coil winding. However, another majorbenefit of the low resistance primary winding is that thebuild up of current flow is much quicker than withprimary windings of higher resistance. With this fact inmind, it is then possible to allow an initial rapid buildup of current flow but to then limit the current so that itdoes not reach levels that are too high; the coil will beable to produce high energy levels due to the rapidbuild up time but without the problem of high currentsdamaging the system components.

Later generations of electronic ignition modules andmodern ignition systems use a method of current controlor ‘current limiting’ in conjunction with low resistanceignition coils. The process of current control is generallycarried out in one of two ways, both methods relying ona ‘feedback’ or ‘closed loop’ system. In effect, the systemmonitors the current flow in the primary circuit and


[647] Chapter 02 25/8/06 15:37 Page 57

when it rises to a predetermined level, the current in theprimary circuit is then either restricted or the dwell timeis reduced; both methods will prevent overheating thatwould otherwise be caused by an excessive current levelthat lasts for long periods.

Feedback systemThe feedback systems are relatively simple in operationand rely on a resistor that forms part of the primarycircuit, i.e. it is in series with the primary winding of theignition coil. The resistor (referred to as a sensingresistor) is located within the ignition module, which isof course functioning as a switch on the earth path forthe primary circuit. The voltage drop across the resistorwill change with changes in current flow and, by passingthe voltage signal from the resistor to other circuitry inthe module, it is then possible to either control thecurrent or control the dwell time accordingly.

With the current limiting systems, the ‘Darlingtonpair’ within the module is used to switch the primarycurrent through to earth (Figure 2.22). When thevoltage signal from the resistor indicates that thecurrent has reached the predetermined level, the inputvoltage to the Darlington pair is reduced, which in turnreduces the primary circuit current flowing from C to E(collector to emitter). When the current is reduced, thevoltage at the sensing resistor is also reduced and thisvoltage is then again used as the reference to control theinput voltage to the Darlington pair. In effect, theprocess is a continuous action of monitoring andadjustment of voltages and current flow (i.e. it is aclosed loop).

With dwell control systems, the same principle is usedas for current limiting systems but the voltage from thesensing resistor is passed to the dwell control circuitrywithin the module. Dwell control therefore depends onthe voltage at the sensing resistor.

A complete constant energy systemFigure 2.23 shows a simplified layout of a constantenergy ignition system. The process is as follows.

● A trigger signal is passed from an inductive or Halleffect trigger to the ignition module.

● The signal will be processed by the pulse shapingdevice; if the trigger signal is provided by aninductive trigger, it will also be converted fromanalogue to digital.

● The processed signal is then passed through thedwell control device and peak coil current cut offdevice.

● The signal is then passed to the ‘driver’ which iseffectively a low current/voltage switching transistorthat is directly responding to the processed triggersignal. The driver responds to the trigger signal andin turn switches on and off the Darlington pair,which contains the main power transistor.

● The Darlington pair forms the final switchingelement of the ignition module. The primary currentpasses through the Darlington pair, so when theDarlington pair switches on the primary circuit,current will flow through the primary circuit thusenabling the ignition coil to build up a magneticfield. When the Darlington pair switches off theprimary current, the magnetic field in the coil willcollapse, thus providing a high voltage to the sparkplug.

● Note that the voltage either side of the currentsensing resistor is passed to a comparator, whichthen passes an appropriate signal either to thedriver (which controls the Darlington pair), or thesignal is passed to the dwell control device.

Different locations for ignition modulesThere are several variations in the design of theelectronic ignition systems so far discussed. Whilst inprinciple the systems will generally all function in thesame way, the different versions produced for differentvehicles have constructional variations that are eitherdesign preferences or are dictated by installationrequirements.

In general, there are three basic physical layouts:

● remote module – the ignition module is remotelylocated away from the trigger mechanism

● integrated with the distributor body – the module iseither located inside the distributor or mounted onthe outside of the distributor body

● located on the ignition coil – the module is mountedon the ignition coil casing.

Figure 2.24 shows some examples of different modulelocations.


Figure 2.22 Voltage feedback control

[647] Chapter 02 25/8/06 15:37 Page 58


OR

Indu

ctiv

e pu

lse

gene

rato

r

Dar

lingt

on p

air

Comparator

Referencevoltage

Currentsensingresistor

Ignitioncoilprimary

BatteryH

all

gene

rato

r

Pul

sesh

aper

Dwellcontrol

Peak coil currentcut-off

Driver

Figure 2.23 Layout of constant energy ignition system

Figure 2.24 Different physical layouts for ignition systemsa Ignition coil mounted retrofit systemb Remote mountingc Integrated distributor body mounting

(a)

+

Heat sinkassembly

Four heat sinkmounting pointsfor self tapping

screws

Ignition switchExisting wire

or block ballastresistor (if fitted)

Distributor

A

Coil

+

Positive

A

HT cablecap

–

(b)

[647] Chapter 02 25/8/06 15:37 Page 59

2.2.10 Capacitor discharge ignitionCapacitor discharge ignition systems have been used inthe past on some vehicles and were not uncommon onhigh performance engines. The capacitor dischargesystem is however not ideally suited to modern enginesand is therefore seldom used on today’s vehicles.

Capacitor discharge (CD) systems operate in aslightly different way to more traditional types ofignition. Although an ignition coil is still used, the coildoes not store energy as is the case with most systems:the ignition coil on a CD system functions as a ‘pulsetransformer’. In effect, a short and relatively highvoltage pulse (typically of 400 volts) is passed throughthe primary winding of the coil; this causes a very rapidbuild up of a magnetic flux (magnetic field) in theprimary winding. Because the secondary winding isexposed to the rapidly created magnetic field, a veryhigh voltage (typically around 40 kV) is then inducedinto the secondary winding.

To create the short 400 volt pulse, a capacitor in themodule is charged (effectively during a dwell period).However, when a trigger signal is provided, thecapacitor discharges its stored energy through theprimary winding.

A simple capacitor discharge circuit is shown inFigure 2.25. Note that a pulse generator is still used andthe trigger signal from the pulse generator is passed to apulse shaper. The processed signal from the pulseshaper is passed to the trigger stage, and the signal fromthe trigger stage will cause the capacitor to discharge.

CD systems provide a short but very high voltage coiloutput, which is typically only around 0.1 ms induration. This short duration spark is not effective inmaintaining the combustion process with weakerair/fuel mixtures. However, the high intensity 40 kVcoil output is consistent across most engine speeds andit is very effective at igniting relatively rich mixtures.One advantage of a CD system with older highperformance and racing engines was that the highvoltage at the spark plug could burn off anycontaminants. On older racing engines, it was commonon cold engines for the mechanical clearances to belarge until the engine reached operating temperatures;this allowed oil to pass the piston rings and valve guidesand therefore enter the combustion chambers. Oilwould therefore contaminate the spark plugs and causeignition and combustion problems. Additionally,because racing engines operated with relatively richmixtures, carbon fouling of the spark plugs was acommon problem. The high voltage at the spark plugproduced by CD systems was very effective at keepingthe plugs free of contaminants.

Electronic ignition systems use a pulse generatorto signal an amplifier that switches the coil on andoff

To maintain constant energy in a coil, the dwell isvaried. Dwell is the angle of distributor rotationwhen the coil is switched on. It is also given as apercentage

Key

Poin

ts


Figure 2.25 Capacitor discharge system

[647] Chapter 02 25/8/06 15:37 Page 60

2.3.1 A further improvement intiming accuracy

Eliminating the problems of mechanical timingsystemsAlthough earlier generations of electronic ignitionsystems had improved reliability and spark quality, theaccuracy of ignition timing advance and retardmechanisms was still dependent on mechanical andvacuum operated systems that had not changed formany years: the timing advance and retard mechanismswere no different to those of contact breaker systems.The problem again was tighter emissions regulations, arequirement for improved fuel consumption andcontinued demand for improved engine power.

As is often the case with emission requirements,there are conflicts between achieving the desiredemissions and also achieving economy and power.However, if the ignition timing can be more accuratelycontrolled and if the changes in timing (advance andretard with speed and load) can be more rapidlyimplemented, then combustion efficiency can beimproved under almost all conditions.

As mentioned in section 2.2.2, if the correct ignition(spark) timing can be provided at exactly the right time,

the gases will expand at exactly the right time and thiswill allow maximum possible power to be achieved(assuming all other conditions are good). However,because of minor variations in fuel quality and in otherengine operating conditions, it was general practice toset the timing slightly retarded so that combustionknock and other overheating problems did not occur.

Importantly, the relative inaccuracy of mechanicaland mechanical/vacuum based timing mechanismsprevented the timing being correct for all operatingconditions. Even when the components were new,mechanical advance mechanisms and mechanical/vacuum operated retard mechanisms could not providecorrect timing in all operating conditions.

Mechanical ‘engine speed’ related advanceFigure 2.26a shows two advance lines for enginespeed/timing advance; the curved line shows therequirements for an engine, whilst the straight lineshows what was available from a mechanical advancemechanism (bob weight and spring system). Note thatthe kinks or angle changes in the straight line arecaused by using unequal strength springs on the bobweight system (Figure 2.26b): the weaker spring actsagainst the bob weights for the lower engine speeds and

Computer controlled ignition systems 61

2.3 COMPUTER CONTROLLED IGNITION SYSTEMS

Figure 2.26 Engine speed related timing advance curve andmechanical advance mechanisma Ignition advance curvesb Mechanical advance mechanism using unequal length springs

[647] Chapter 02 25/8/06 15:37 Page 61

then, as the bob weights are flung out under centrifugalforce, the stronger spring then also acts against the bobweights.

The difference between the two advance linesshown in Figure 2.26a illustrates the inability of amechanical advance system to provide the correcttiming at all engine speeds. It was therefore necessaryto try to achieve the best compromise for timingadvance, which meant that the timing was not correctfor much of the engine speed range.

As well as not being able to provide the correcttiming at all times when the system was new, when thetiming mechanisms began to wear matters becameworse and timing was often much too far away from thedesired value to enable the engine to operate efficiently.Mechanical and mechanical/vacuum systems could notrespond quickly enough to the required changes intiming and the accuracy deteriorated over time due towear. It was therefore necessary to eliminatemechanical systems and to use electronic or computercontrol for the ignition timing functions.

Load related vacuum retardFigure 2.27 shows a simple vacuum operated timingretard system. The illustration shows the mechanismacting on a base plate onto which the contact breakerassembly were mounted.

When the throttle was closed at idle speed or duringdeceleration (upper left diagram), the vacuum wasblocked from reaching the diaphragm and therefore anytiming changes were dependent purely on themechanical bob weight system (engine speed related).

When the throttle was then partially opened (lightload conditions), high manifold depression acts onthe diaphragm, which in turn pulls the diaphragm

against the spring. The movement of the diaphragmpulls the linkage, which rotates the base plate in ananticlockwise direction, against the direction ofrotation of the distributor shaft. This would have theeffect of advancing the trigger signal and timing.

When however the throttle was then openedfurther (high load conditions), the intake depressionwould initially reduce (higher pressure), the springwould then force back the diaphragm and in turn, thiswould allow the base plate to move back in aclockwise direction (the same direction as thedistributor shaft rotation). The opening of the contactbreaker and timing would therefore retard.

Whilst the vacuum advance/retard system wasreasonably effective, it could not accurately controltiming for all the variations of load that occur andagain a compromise was inevitable. Adding thecompromise of a vacuum system to the compromiseof a mechanical advance system, resulted in aninaccurate timing control system.

For electronic systems, near the triggermechanism located in the distributor (e.g. inductive,Hall effect or optical). Refer to Figures 2.11, 2.12 and2.19. The movement of the base plate and distributorshaft altered the timing in the same way.

Changing operating conditionsApart from the lack of accuracy of mechanical andvacuum based timing systems, there are other reasonswhy greater flexibility of timing control is needed.Air/fuel mixtures need to be altered more rapidly, so theignition timing must be altered accordingly; in addition,ignition timing requirements differ with temperature. Ineffect, anything that affects the speed of combustion,i.e. the flame speed through the combustion chamber


Figure 2.27 Vacuum advance/retard mechanism

[647] Chapter 02 25/8/06 15:37 Page 62

and the burn time, will require a different timingadvance value.

Although many adaptations to basic mechanical andvacuum systems were introduced, the accuracy andspeed of change of timing remained less thansatisfactory for modern requirements.

With computer control, and the use of a number ofsensors, it became possible to obtain much moreaccurate control of ignition timing. Rapidly changingoperating conditions could therefore be sensed(including rapid changes in engine speed), and thecomputer or ECU could then alter the timing asnecessary. Systems were therefore introduced where theswitching action of the ignition module was controlledby the ECU. A trigger signal is still produced by aninductive or Hall effect trigger (or other type of pulsegenerator), but the signal is passed to the ECU, whichmodifies or ‘phases’ the signal to achieve timing control.Figure 2.28 shows the basic layout of such a system.Note that the ignition module can be separate from theECU or, in many cases, the module is integrated withinthe ECU.

The simple example in Figure 2.28 shows an ECUreceiving a trigger signal provided by an inductivesensor. A vacuum pipe connects the intake manifoldvacuum (depression) to a pressure sensitive componentin the ECU thus providing the ECU with engine speedand load information. See section 1.5.4 for informationon electronic type pressure sensors.


2.3.2 Digital timing controlIf we assume that an ECU receives a trigger signal fromsome form of engine speed sensor, and loadinformation via the vacuum sensor, the ECU can thencalculate the required ignition timing. Within the ECU isa ‘look up table’ or memory, which contains all therelevant timing data applicable to the different loadconditions; by comparing the conditions with the datain the look up table, the ECU can determine the timingadvance required.

The look up table effectively contains a three-dimensional map of the timing requirements, a simpleexample of which is shown in Figure 2.29. The sparktiming map provides the timing angle (crankshaft angle)related to engine speed (revolutions/second) and engineload (on a scale of 0 to 1). The example in Figure 2.29shows a relatively small number of speed, load andtiming reference points, but many more reference pointscan be included on a spark map.

Figure 2.30 shows the same spark timing map withthe engine speed at 32 rev/s (1920 rev/min) and theengine load at 0.5 (half load). The 32 point mark on themap is therefore followed across until it intersects withthe 0.5 load line; the intersect point identifies therequired spark advance, which in this case is 52°.

The timing advance characteristics are establishedduring many tests on development engines (beforethey go into production). Once the timing

Figure 2.28 Layout of a computer (ECU) controlled ignition system

[647] Chapter 02 25/8/06 15:37 Page 63

characteristics have been established, production ECUswill then have a memory and look up tables containingthe applicable data.

Because accurate timing is so important to engineefficiency, performance, emissions and economy,improvements in all of these factors are achieved witha greater number of timing reference points.Figure 2.31 shows a more modern and complex sparkadvance map (compared with the example inFigure 2.29).

It can now be appreciated that considerableaccuracy of spark timing can be achieved with thecomplex spark timing map. However, because thetrigger signal from the inductive or Hall effect triggers(and other pulse generators) discussed so far providesonly one reference point per cylinder, this is a limiting

factor in the accuracy of ignition/spark timing.Additionally, the trigger mechanisms so far discussedare usually located within the distributor body, so therotation of the trigger mechanisms is driven by thedistributor shaft which in turn is driven by a timingbelt or chain. Any wear or maladjustment of the beltor chain will cause inaccuracies in the timingreference.

A more reliable and accurate means of providing atrigger/reference signal would enable the spark timingalso to be more accurate. The following section (2.3.3)covers some examples of more accurate timingtrigger/reference signal systems which usually have asensor and reluctor or trigger disc located on thecrankshaft.


Figure 2.29 Simplified three-dimensional spark advance map

Figure 2.30 Simplified three-dimensional spark advance map showing the spark timing at 32 rev/s and half engine load

[647] Chapter 02 25/8/06 15:37 Page 64

2.3.3 Crankshaft speed/positionsensors

Direct triggering reference from the crankshaftMost crankshaft speed/position sensors are locatedadjacent to the crankshaft and are usually inductivesensors (Figures 2.32 and 2.33). In some cases Halleffect sensors have been used. Locating a sensoradjacent to a crankshaft allows a reluctor disc (triggerdisc with the reference points) to be mounted directlyon the crankshaft, with most versions being mountedeither at the front pulley or at the rear of the crankshaftadjacent to the flywheel. Sometimes a reluctor disc ismounted at a convenient point on the crankshaftbetween the crankshaft webs, i.e. the disc is located inthe crankcase.

The obvious advantage of locating thereluctor/trigger disc on the crankshaft is that there is nodrive linkage (belt, chain or other mechanism); thismeans that the trigger or reference signal will be

accurately identifying the crankshaft speed or angularposition without any losses of accuracy that could becaused by drive mechanisms.

Increased number of reference pointsWith a crankshaft mounted reluctor or trigger disc, it ispossible to use a larger diameter disc, which can moreeasily contain a larger number of reference points(reluctor teeth). Due to the fact that most of thecrankshaft speed position sensors are of the inductivetype, the reference points usually take the form of teethlocated around the disc.

In most cases the trigger disc is located directly atthe front or the rear of the crankshaft and can form part


Figure 2.31 Typical modern three-dimensional spark advance map

Figure 2.32 Crankshaft speed/position sensorFigure 2.33 Crankshaft speed/position sensor located adjacent toflywheel

[647] Chapter 02 25/8/06 15:37 Page 65

of the front pulley or part of the flywheel assembly.Because of these locations, especially if the disc has asimilar diameter to the flywheel, it is relatively easy tolocate a number of reference points around the disc.Some earlier examples had only a small number ofreference points, e.g. two or four, but most modernsystems have as many as 60 reference points. One of thereference points or teeth is usually either missing or is adifferent shape to the rest of the teeth; this enables amaster reference signal to be produced.

Figure 2.33 shows a flywheel located inductivecrankshaft speed/position sensor. Note the missingtooth, which is adjacent to the sensor in the illustration.In fact, the example shown in Figure 2.33 has twomissing teeth; one missing tooth is a master referencefor cylinders 1 and 4, while the other missing tooth is amaster reference for cylinders 2 and 3. On thisparticular system, with a single ignition coil, the highvoltage output from the coil is directed via a rotor armlocated at the end of the camshaft (OHC type).Therefore the ECU does not need to receive a signalrelating to each cylinder, but it does receive signalsrelating to the TDC position of each pair of cylinders (orother predefined angular position of the crankshaft).

On this system therefore, the sensor would providethe speed signal and the angular position of thecrankshaft as each tooth passes the sensor; the ECU cancount the number of signals as each tooth passes thesensor thus enabling the ECU to identify the angle ofrotation from TDC (or from the master referenceposition). It is also possible on this type of system forthe ECU to assess any changes in crankshaft speed aseach tooth passes the sensor.

Using Figure 2.28 as an example, the ECU receives aload signal from the pressure sensor (pressuretransducer), which is connected by a vacuum pipe to


TDCNo. 1cylinder

Optical pick-up

Rotor plate

Figure 2.34 Optical speed/position sensor located in distributorbody

Figure 2.35 Analogue signal produced by an inductive crankshaftspeed/position sensor

the intake manifold. This system also uses informationfrom a knock sensor (discussed in section 2.3.4). Alsonote that a coolant temperature sensor providesinformation to the ECU to enable changes in timing tooccur with changes in temperature (temperaturesensors are covered in section 1.5.1).

Note: Some speed/position sensor systems locatedwithin a distributor also have many reference points,one type being an optical system with 360 slots locatedaround a disc (Figure 2.34). However, locating a largenumber of reference points around a relatively smalldisc requires good manufacturing accuracy and theproblem still remains that any wear or maladjustmentof the drive linkage to the distributor will result inincorrect timing references.

Crankshaft speed/position sensor: operation andsignalsSection 2.2.5 provides an explanation of an inductivepulse generator, and the inductive crankshaftspeed/position sensor operates in exactly the same way.The crankshaft sensors, however, are usuallyconstructed so that the winding or coil is formedaround the magnet and this assembly is located in thesensor body, which is then bolted or secured in someway to the engine block or flywheel housing. The sensoris located so that it will be affected by the movement ofthe reluctor teeth (reference points) whilst thecrankshaft is rotating.

Figure 2.35 shows a typical analogue signalproduced by an inductive crankshaft speed/positionsensor. Note the different shaped pulse produced by themissing tooth.

The analogue signal is passed to the ECU whichthen converts it to a digital signal, thus enabling therequired speed and angular position information to beobtained.

[647] Chapter 02 25/8/06 15:37 Page 66

Other types of sensor signalAlmost all crankshaft speed/position sensors are of theinductive type, but where a Hall effect or optical systemis used, these types will provide a digital signal in thesame way as the older ignition trigger systemsdiscussed previously in sections 2.2.6 and 2.2.7. Theonly differences compared with the older ignitiontrigger systems will be the number of signal pulsesproduced which will depend on the number of referencepoints.

2.3.4 Knock sensorsA knock sensor (Figure 2.36) is effectively a vibrationsensor that responds to those vibrations in the enginethat cause pressure waves to occur in the cylinder blockor cylinder head. By detecting the vibrations or pressurewaves, a knock sensor can detect the vibrations causedby combustion knock.

The knock sensor is an electronic pressure sensor,which with a pressure sensitive crystal that produces asmall electrical pulse when it is exposed to pressurewaves (such as the engine vibrations). Vibrationscaused by combustion knock will result in a slightlydifferent signal (frequency and voltage) being producedby the sensor. When the ECU receives the signal fromthe sensor, it is able to filter out the normal vibrationsand respond to the particular part of the signal that iscaused by combustion knock.

Although ECU controlled ignition timing should provideideal timing for all operating conditions, it is possiblethat fuel quality could be poor (momentarily orcontinuously). Other factors such as the temperature inthe combustion chamber can also cause short termcombustion knock. In most cases, slightly retarding theignition timing/spark advance will reduce andeliminate combustion knock.

Therefore, when the ECU detects a combustionknock signal, it will respond by retarding the sparktiming a predetermined number of degrees. If thecombustion knock is no longer detected, the ECU willprogressively advance the timing to its correct value (solong as combustion knock does not reoccur).

An ECU can alter the timing for just the affectedcylinder. When the knock occurs (when combustionoccurs in the affected cylinder), the ECU will thenprovide the correct timing for the remaining cylinders(for example, the remaining three cylinders on a four-cylinder engine). When the affected cylinder is then dueto receive its next spark, the ECU can retard the timingfor just the affected cylinder.

The main advantage of computer controlledignition is accurate timing – that stays accurateover the life of the engine

The ideal timing setting is held in the ECU memoryin the form of a look up table

Key

Poin

ts


Figure 2.36 Knock sensor located in cylinder block

[647] Chapter 02 25/8/06 15:37 Page 67

2.4.1 Limitations of distributor basedsystems

Restricted dwell time and wasted energyAll of the ignition systems covered so far have twomajor disadvantages when it comes to providing highenergy from the ignition coil to the spark plugs; bothdisadvantages arise because the ignition systems use asingle coil to provide a spark for all of the cylinders.

Time to build up coil energyThe first disadvantage when using a single coil for allcylinders is that it limits the time available to build upcoil energy between each of the individual ignitioncycles: there is very little ‘dwell’ time available for thecurrent to flow through the primary winding of the coiland build up a strong magnetic field (magnetic flux).

As previously explained, on a multi-cylinder enginethere is very little time between one cylinder firing andthe next; the faster the engine speed and the greater thenumber of cylinders, the less time there is available forthe ignition coil to build up sufficient energy for thenext ignition cycle. On modern high speed engineswhich operate with relatively weak mixtures, it isessential that the energy available from the coil issufficient to produce a powerful and long durationspark otherwise emissions and general performancewill not be acceptable.

Distributor and rotor arm wasting energyWhen a single coil is used to produce a spark for anumber of cylinders, the energy from the coil (highvoltage) is passed via a high tension cable (HT lead) tothe distributor cap and rotor arm assembly (see section2.28 in Hillier’s Fundamentals of Motor VehicleTechnology Book 1). Note that the lead passing theenergy from the coil to the distributor cap is oftenreferred to as the ‘king lead’. The distributor capcontains a number of contact points (referred to aselectrodes), which in turn are connected to each of thespark plugs via additional HT leads. When the highvoltage from the coil passes along the king lead to thecentre electrode in the distributor cap, it is then passedto the centre of the rotor arm; because the rotor armrotates with the distributor shaft, it is then able to passthe energy to the individual HT leads and spark plugs.

Figure 2.37a shows a basic layout of an ignitionsystem with a single coil and Figure 2.37b shows a planview of the rotor arm and distributor cap.

One problem with the rotor arm system is thatvoltage is lost or wasted when the current flow flowsthrough all of the HT leads, and especially when thecurrent flows across the rotor arm tip to each of theelectrodes. There is a necessary gap between the rotorarm tip and the electrode, and this absorbs or usessome of the energy produced by the coil.

However, although Figure 2.37b shows the rotor arm inalignment with the electrode, in reality, the rotor armpasses through quite a large angle during the period oftime that the spark exists (remember that the sparkmay last 2 ms or more). When an engine is operating at6000 rev/min, the rotor arm (which rotates at halfengine speed) will rotate 50 times in one second or onerotation in 0.02 s (2 hundredths of a second). Duringthe spark duration of 2 ms (2 thousandths of asecond), the rotor arm will rotate through one-tenth ofa complete rotation i.e. 36° of rotation. There istherefore quite a substantial gap between the rotor armtip and the electrodes when the energy from the coil ispassing to the spark plug. This gap inevitably usesconsiderable amounts of valuable energy, whichreduces the energy available to maintain the spark.Additionally, the rotor arm tip and electrodes willprogressively deteriorate due to the arcing that occursas the voltage or energy flows across the gaps.

It is also important to note that, when electricityhas to jump the gap at the rotor arm tip, this createselectrical interference, which must be suppressed toprevent interference with other electrical andelectronic devices.

Using multiple ignition coils (eliminating the rotorarm)The next progression in ignition system design wastherefore based on a desire to eliminate the distributorcap and rotor arm assembly and use more than oneignition coil. Note that there were some enginesproduced (typically V8 and V12 engines) that did usetwo coils, each of which provided sparks for half of thecylinders. However, these systems still used one rotorarm for each coil and group of cylinders: in effect therewere two ignition systems.

The ultimate ignition system would have one coilfor each cylinder, and this is the general rule for modernengines, where an individual ignition coil is eitherdirectly connected to the top of the spark plug or thereis an HT lead from each coil to the spark plug.

There is however another solution, which is stillused on some systems, and this design uses a single coilto provide a spark at two spark plugs. Although thesesystems are often referred to as ‘distributorless’ ignitionsystems, the same terminology can be applied tosystems that use one coil for each cylinder. For thepurposes of differentiation, coils that provide sparks totwo cylinders at the same time are referred to withinthis book as ‘wasted spark’ systems, the reason forwhich will be made clear in section 2.4.2. Section 2.4.3covers systems that use a ‘single coil per cylinder’.

For both the wasted spark and the single coil percylinder systems, the ignition systems do not require adistributor and rotor arm assembly to distribute thespark to the different cylinders: they are both therefore


2.4 DISTRIBUTORLESS AND DIRECT IGNITION SYSTEMS

[647] Chapter 02 25/8/06 15:37 Page 68

‘distributorless’ ignition systems. The term ‘directignition’ is also used but this would generally refer tosystems where the ignition coil is directly located ontothe spark plug, i.e. there is no HT lead between the coiland spark plug.

2.4.2 Wasted spark ignition systemsPrinciple of operation of a wasted spark systemProviding a spark at two cylindersFor the majority of engines layouts where an evennumber of cylinders is used, two of the pistons will riseand fall within the cylinders at the same time. Using anin line four-cylinder engine as an example: pistons 1and 4 rise and fall together, but when cylinder 1approaches TDC on the compression stroke and isprovided with the spark, piston number 4 isapproaching TDC on the exhaust stroke. However, atthe next full rotation of the crankshaft, the situation isreversed. The same process is true for pistons 2 and 3.

Therefore, with an ignition coil that can provide aspark at the spark plugs for cylinders 1 and 4 at thesame time, if cylinder 1 receives a spark at the correcttime (i.e. at the top of its compression stroke) cylinder 4will receive the spark when it is on the exhaust strokeand therefore the spark to cylinder 4 is wasted. On the

next revolution of the crankshaft, the wasted spark willbe at cylinder 1 and cylinder 4 will receive the spark atthe correct time (i.e. at the top of its compressionstroke). Figure 2.38 shows the basic principle: a pair ofcylinders that rise and fall together (but on differentstrokes) receive a spark at the same time.

Distributorless and direct ignition systems 69

Figure 2.37 Ignition system with a single ignition coila System layoutb Distributor cap and rotor arm: plan view

Discharge circuit

Spark plugIgniter

Spark plug

Ignition coil

+B

Figure 2.38 Single ignition coil providing a spark to twocylinders at the same time

[647] Chapter 02 25/8/06 15:37 Page 69

Positive and negative sparksThe ignition coil on a wasted spark system operates inthe same way as the conventional coils so fardiscussed. However, both ends of the secondarywinding in the coil are connected to an HT lead and aspark plug (Figure 2.39), whereas on a conventionalcoil only one end of the secondary is connected to anHT lead and spark plug (the other end is connectedinternally within the coil to the end of the primarywinding, which completes the earth path for thesecondary winding).

On the wasted spark ignition coil, when themagnetic field collapses around the secondarywinding, one end of the winding will be positive whilstthe other end will be negative. The flow of current atthe positive end of the winding will pass from the coilwinding down to the spark plug and then across theplug gap to earth (the engine acts as the earth point forthe spark plugs). Note that current flow is generallyregarded as passing from the higher voltage to thelower voltage, and, if we assume that the outputvoltage from the coil in this example is 10 000 volts,current will flow from the 10 000 volts at the coil tothe zero volts at earth.

The flow of current at the negative end of thewinding will however be from earth (the earthelectrode at the spark plug) across the plug gap andthen through the HT lead to the coil. This direction offlow is caused by the fact that the voltage at thenegative end of the coil winding is a minus value, e.g.–10 000 volts, which is 10 000 volts lower than thevoltage at earth (which is zero volts); current is againassumed to flow from the higher to the lower voltage.

Note: Electron flow is actually opposite to thegenerally accepted convention: electrons flow from thelower to the higher voltage rather than from the

assumed higher to the lower voltage. However, as withmost explanations, it is assumed that conventionalflow is from the higher voltage to the lower voltage.

Irrespective of which way the current is flowing acrossthe spark plug gap, a spark will be provided that issufficient to cause combustion in the combustionchamber. However, because electrons flow naturallyfrom a hot surface, and the centre electrode is thehotter of the two electrodes, it is easier to produce thespark or arc if the current flows in a direction thatmatches this natural electron flow (because the voltagerequired is lower). Conventional coils are wired in sucha way that there is a ‘negative spark’ at the plug gap.However, with the wasted spark coils, one spark will bea negative spark and the other will be a positive spark.

The positive spark is still effective at creatingcombustion but there is a tendency (because ofincorrect electron flow) for the electrodes to operate atslightly lower temperatures, which can result in foulingof the plug electrodes. It is therefore necessary either touse different grades of spark plug for the positive andnegative sparked plugs (which is not desirable) or touse a spark plug that is effective irrespective of thepolarity of the spark.

Increased dwell timeAlthough the term ‘dwell’ is perhaps not reallyapplicable to modern electronic ignition systems, it stillused to refer to the period of time available forbuilding up coil energy, i.e. the time for current to flowthrough the primary winding. We can thereforeexamine the ‘dwell time’ available for a single coilsystem and for a wasted spark system for a four-cylinder engine to assess the increased amount of dwelltime available.

1 When a single coil is used to serve all fourcylinders, the coil is required to provide a sparkwhen each of the cylinders approach TDC. Thismeans that in two revolutions of the crankshaft, thecoil will be required to provide four sparks, which istwo sparks per revolution of the crankshaft. Thisequates to one spark or one complete cycle ofoperation for the coil for every half rotation of thecrankshaft.

2 When a wasted spark system is used on a four-cylinder engine, there would be two wasted sparkcoils: one coil for cylinders 1 and 4, and one coil forcylinders 2 and 3. If we examine the operation ofthe coil that serves cylinders 1 and 4, it will berequired to provide a spark each time that the twopistons approach TDC, which equates to onecomplete cycle of operation for each revolution ofthe crankshaft.

With a wasted spark system on a four-cylinder engine,the ignition coils now have twice as much timeavailable for the cycle of operation (between one sparkand the next), which means that it would be possible to


Figure 2.39 Schematic view of a wasted spark ignition coil

[647] Chapter 02 25/8/06 15:37 Page 70

provide twice as much dwell time (if necessary) in orderto build up the energy in the coil. Note that, whenever awasted spark system is used, there will be twice asmuch time available compared with the same engineusing a single coil with the same number of cylinders.

Typical operation and layout of a wasted sparksystemOn a wasted spark system, the coils are switched inexactly the same way as on a single coil system: the ECUcalculates the ignition timing in the same way and theECU will cause the ignition module to switch each of thewasted spark coils at the appropriate time. Because thereare effectively two coils on a four-cylinder engine, therewould need to be two switching modules, i.e. the ECUwill be controlling two modules, each of which willswitch the primary circuit for one of the coils. In reality,the modules are usually integrated into the ECU and infact the ECU will contain two power stages (one for eachcoil), with all other functions such as current and dwellcontrol being managed by the ECU.

The ECU will require information (from a crankshaftspeed/position sensor) to indicate when one pair ofcylinders is approaching TDC; in effect a masterreference. Once this master reference has beenestablished, the ECU can make one of the coilsprovide the sparks to one pair of cylinders. If thecrankshaft sensor is providing many reference pointsin addition to the master reference (because there is alarge number of teeth on the reluctor disc), the ECUcan then calculate the angle of rotation of thecrankshaft and make the other coil or coils providesparks to the other pair of cylinders. As with otherECU controlled systems, a load sensor and possiblyother sensors, e.g. temperature sensors, can provideinformation to enable the ECU to more accuratelycalculate the correct ignition advance angle (sparktiming).

Figure 2.40a shows a basic layout of a wastedspark system, with an example of a typical wastedspark coil shown in Figure 2.40b (note that therewould be one of these coils for each pair of cylinders).


Figure 2.40 Wasted spark ignitiona Layout of systemb Example of a wasted spark ignition coil

[647] Chapter 02 25/8/06 15:37 Page 71

Alternative construction and operation of wastedspark system with diode controlA development of the wasted spark system used a ‘coilpack’ with two primary windings but only onesecondary winding (Figure 2.41).

The ECU functioned as on the previous system byswitching the two primary windings alternately. Thepolarity applied to the two primary windings isdifferent; therefore this will cause the magnetic fieldcreated in the primary and secondary windings toalternately reverse polarity. In turn this will cause theoutput current from the secondary winding to alsoalternately reverse polarity.

In effect, the HT voltage at point A on the secondarywinding (Figure 2.41) will alternately be positive thennegative, depending on which of the primary windingsis active on each of the ignition cycles; the samesituation will therefore occur at point B, so that whenthe HT coil output at A is positive, it will be negative atpoint B. This situation will reverse when the otherprimary winding is active on the next cycle.

If we assume that, on one of the cycles the current isflowing through the secondary winding from A to B(from earth through A to B and then back to earthagain), then the diode at spark plugs 1 and 4 willprevent current flow. However, the diodes at sparkplugs 2 and 3 will allow current to flow, thus allowing aspark to occur at spark plugs 2 and 3.

On the next cycle, the other primary winding will beactive and this will cause the polarity at A to benegative and B to be positive. Therefore the diodes onspark plugs 1 and 4 will allow current to flow andsparks to occur at those plugs. However the diodes onspark plugs 2 and 3 will prevent current flow andtherefore there will be no spark at plugs 2 and 3.

This design of system allows for a more compact coilassembly that will contain the two primary windingsand a single secondary winding, as well as the diodes.The ECU still performs the same task as for thepreviously described wasted spark system, in that itcontrols the switching of the primary circuits at theappropriate time (spark timing). The ECU also stillrelies on the crankshaft speed/position sensor and a

load sensor (usually a vacuum sensor) for the requiredinformation.

2.4.3 Single coil per cylinder and coilon plug ignition systems

The logical developmentA high percentage of modern petrol engines are fittedwith ignition systems that use one coil for eachindividual cylinder. The basic principle of operation foreach coil remains the same as for previously describedcoils. However, unlike wasted spark systems, each coilcan provide a negative spark.

One logical reason for using a single coil for eachcylinder is the fact that in recent years there have beenseveral engine designs with odd numbers of cylinders(five cylinders and three cylinders). A wasted sparksystem cannot effectively be used on engines with oddnumbers of cylinders, since such a system is suited toengines where two pistons rise and fall at the sametime. Older distributor cap and rotor arm systems areinefficient, so, on engines with odd numbers ofcylinders, it is necessary to use one coil for eachcylinder.

Another advantage of single coil per cylindersystems is that the ignition coil can be mounted directlyonto the spark plug. This direct connection eliminatesthe need for an HT lead and therefore reduces thepotential for lead and connection failure. Electricalinterference is also greatly reduced. However, somesingle coil per cylinder systems do locate the coilsremotely from the spark plugs and therefore need an HTlead. A basic layout for a single coil per cylinder systemis shown in Figure 2.42.

With single coil per cylinder systems, it is perhapsobvious that the time available to build up coil energy isgreater than when a single coil is used to provide aspark to all cylinders. As an example, on an eight-cylinder engine with eight ignition coils, each coil willhave eight times longer to complete one ignition cyclecompared with the same engine using a single coil forall cylinders. The available ‘dwell’ time is therefore alsoup to eight times longer.

The operation of the system is much as for wastedspark systems except that the ECU will now have onepower stage to switch each of the ignition coils. TheECU will receive information from a crankshaftspeed/position sensor and other information from loadsensors, etc. In most cases, single coil per cylindersystems form part of an engine management systemand the ECU is usually the same ECU that controlsfuelling and other functions.

Cylinder recognitionSingle coil per cylinder systems require one additionalitem of information from sensors compared with otherignition systems and this is referred to as ‘cylinderrecognition’.


Figure 2.41 Wasted spark ignition system with diode control

[647] Chapter 02 25/8/06 15:37 Page 72

Although the ECU can receive a master reference signalfrom the crankshaft speed/position sensor, the masterreference cannot identify an individual cylinder; on afour-cylinder engine, a TDC reference for number 1cylinder will also indicate TDC for number 4 cylinder.Therefore a separate sensor is required to identify whenone particular cylinder is on a particular part of theengine operating cycle, e.g. TDC on the compressionstroke for number 1 cylinder.

Because the camshaft rotates once for everycomplete engine cycle and the crankshaft rotates twice,the camshaft is the best place to locate a cylinderidentification sensor. The sensors are generally eitherinductive or Hall effect and operate in exactly the sameway as previously described inductive and Hall typepulse generators. However, whichever type of sensor is

used, it requires only one reference point to provide therequired information to the ECU.

It is quite common for a trigger lug to be located onthe camshaft; this lug can look very similar to a camlobe but it functions as the reluctor tooth for theinductive sensor; the sensor is therefore located on thecamshaft cover or housing in line with the trigger lug.Each time that the trigger lug passes the sensor, asingle electrical pulse will be passed to the ECU, whichcan then assess which cylinder is on which part of thecycle. It does not matter which cylinder or which strokeof a cylinder’s cycle the signal identifies, so long as thenecessary information is programmed into the ECU.

The ECU uses the cylinder identification signal toestablish when a particular cylinder is on thecompression stroke, and the spark will then beprovided to that cylinder at the appropriate time. TheECU then uses the information from the referencepoints on the crankshaft speed/position sensor tocalculate when the sparks should be provided to theremaining cylinders.

Cylinder recognition sensors are used for mostmodern petrol injection systems so, as is the case withmost sensors, the information is provided to the ECUwhich is then able to control ignition and fuel systems.

Distributorless ignition uses the ‘lost or wastedspark’ principle

Direct ignition systems use one coil for eachcylinderK

ey P

oin

ts


42

1

1 Ignition lock2 Ignition coil3 Spark plug4 ECU5 Battery

5 3

Figure 2.42 Single coil per cylinder ignition systemNote that the module can be located within the ECU or separatefrom the ECU. In some cases the module is located with theignition coil.

2.5 SPARK PLUGS

2.5.1 FunctionThe sparking plug provides the gap across which thehigh tension current (coil energy) jumps, thus creatingan arc or spark that will then ignite the petrol–airmixture. Since the Frenchman Etienne Lenoir inventedthe anti-flashover ribbed insulator sparking plug in1860, many detailed changes have been made, but thebasic construction has remained the same: a highlyinsulated electrode is connected to the HT cable, and anearth electrode joined to the plug body.

2.5.2 Spark plug requirementsThe basic requirement is that a spark of sufficientenergy should be produced across the electrodes at alltimes, irrespective of the pressure and temperature ofthe gases in the combustion chamber. These two factorsof temperature and pressure create a very hostileoperating environment, as highlighted in the followingparagraphs, but in addition to these basic requirements,

the plug must be: resistant to corrosion, durable, gastight and inexpensive to produce.

PressureBesides withstanding an operating pressure of about70 bar during combustion, the plug must be also able toproduce the high energy spark when the gas pressurewithin the cylinder is about 10 bar or more (at the top ofthe compression stroke). The voltage required to do thismay be as high as 30 kV, so adequate insulation isrequired to prevent leakage of electrical energy to earth.

TemperatureThe plug must be capable of withstanding temperaturesof between 350°C and 900°C for long periods of time.The spark plug construction should ensure thatelectrode temperature remains between these limits,because if these limits are exceeded the plug will fail.

Above 900°C the high temperature of the electrodescauses pre-ignition, whereas below 350°C carbon willform on the insulator; this can cause fouling whichallows the electrical energy to find an easier route toearth via the carbon rather than jumping the plug gap.

[647] Chapter 02 25/8/06 15:37 Page 73


2.5.3 ConstructionFigure 2.43 shows the construction of a typical sparkplug. The example shown consists of an alloy/steelcentre electrode and an aluminium oxide ceramicinsulator, which is supported in a steel shell. Gas leakagepast the insulator is prevented by ‘sillment compressedpowder seals’ and leakage between the cylinder headand the plug shell is prevented either by a gasket (oftena copper or alloy gasket) or by using a tapered seating(where the shell contacts the cylinder head).

An earth electrode (usually of rectangular crosssection) is welded to the shell. Whilst most plugs havetraditionally used a single earth electrode, there aremany designs where more than one earth electrode isused. There is normally a hexagon formed on the shell toenable a socket to be used for installing and removing(tightening and loosening) the spark plug.

Ribs are formed on the outside of the insulator,which increase the length of the flashover path, and alsoimprove the grip of the HT lead end covers (or coil endcovers) that are used to prevent moisture or dirtgathering around the insulator.

Plug terms and identificationThe length of the thread that screws into the cylinderhead is called the ‘reach’ and the diameter of thethreaded part indicates the ‘plug size’; common sizesused are 10, 12, 14 and 18 mm.

Spark plug manufacturers use their own codes toidentify their products and the variations; the letters

and numbers stamped on the insulator give thefollowing information:

● diameter and reach● seat sealing and radio interference features● centre electrode features, such as the incorporation

of a resistor or an auxiliary gap heat range● configuration of the firing end of the plug.

Figure 2.44 shows the position of the plug when it isscrewed into the cylinder head. The gasket (whereused) creates a difference between the seating height Aand the plug reach.

Heat rangeHeat range indicates the temperature range in which aplug operates without causing pre-ignition and withoutcausing plug fouling due to carbon or oil deposits onthe insulator.

Figure 2.45 shows the heat limits and the effect ofroad speed on the temperature of a typical plug. Inaddition to pre-ignition and carbon fouling, the graphshows that an operating temperature in excess of about

Figure 2.43 Spark plug construction

Figure 2.44 Spark plug screwed into cylinder head

Figure 2.45 Spark plug operating temperature

[647] Chapter 02 25/8/06 15:37 Page 74

750°C causes oxide fouling of the insulator andexcessive burning of the electrodes.

The operating temperature of a plug depends on thefour features shown in Figure 2.46, these are:

1 Insulator nose length. This is the distance from thetop of the electrode to the body. The length of thisheat flow path governs the temperature of theinsulator nose; therefore if the path is made short,the plug will run relatively cool.

2 Projection of insulator. The amount that theinsulator protrudes into the combustion chambergoverns the amount of cooling obtained from theincoming air/fuel charge.

3 Bore clearance. The clearance between theinsulator and the shell governs the amount ofdeposit that can be accepted before the plugelectrodes are shorted out (the spark finds a shorterroute to earth).

4 Material. Rate of heat transfer depends on thethermal conductivity of the materials used,especially the material used for the insulator.

Figure 2.47 shows two plugs with different heat ranges.The hot (or soft) plug has a long heat transfer path andis recommended for cool running, low compressionengines, and other engines that are used continually atlow speed for short journeys. Unless this type of plug isused on these engines, carbon will build up on theinsulator, thus causing misfiring to occur after a shortperiod of time.

At the other end of the heat range is the cold (or hard)plug. The cold plug has good thermal conductivity,and is therefore used on engines where highcompressions and high combustion temperatureswould cause excessive temperatures at the spark plug.This type of plug is therefore frequently used in highperformance engines or engines with high poweroutputs for their size.

Spark plug manufacturers offer a wide range ofplugs, thus enabling engines to be fitted with anappropriate type of plug to suit the operatingtemperatures, etc.

2.5.4 Electrode featuresMaterialsSpark plugs have traditionally used nickel alloy for theelectrodes, which give good resistance to corrosiveattack by combustion products, and also goodresistance to the erosion that is caused by the highvoltage arc. Both electrodes must be robust to withstandvibration from combustion effects and they must also becorrectly shaped to allow a spark to be produced withminimum voltage (Figure 2.48a). Under normalconditions, erosion eats away the electrodes, so after aperiod of time the earth electrode becomes pointed inshape (Figure 2.48b); in this state it requires a highervoltage to produce a spark.

Spark plugs 75

Figure 2.46 Spark plug features that affect temperature

Figure 2.47 Spark plug heat range

Figure 2.48 Electrode wear

[647] Chapter 02 25/8/06 15:37 Page 75

As previously stated, nickel alloy has traditionally beenused for the electrodes. Although increasing theinsulator nose length reduces the risk of carbon foulingwhen the vehicle is operated on short journeys, the plugcan then overheat when high vehicle speeds aremaintained for long periods. This temperature problemcan be overcome by using more expensive materials andalloys for part of the construction of the electrodes,which can include materials such as platinum, iridium,silver or gold–palladium alloy. However, a lessexpensive material is copper, which can be used for thecore of the central electrode and which provides goodthermal conductivity (Figure 2.49).

Platinum and other materials can be used for theelectrodes to overcome the problems of erosion causedby high voltage arcing. These expensive materials mightbe used only as a coating over the base material, so careshould be taken not to scratch the coating.

Plug gapsSpark plug gaps used to require regular adjustment, andthe electrodes would require regular cleaning and evenfiling to ensure that their tips were the correct shape (anincorrect shape would be caused by erosion). This wasvery important on vehicles with older ignition systemsbecause the eroded electrode tips would often result inlarger gaps that required additional voltage for an arc toform. However, older ignition systems (primarilycontact breaker systems) would produce lower voltagesdue to wear and maladjustment of the contact breakers.It was therefore common for poor starting and misfiresto occur if spark plug maintenance and ignition systemmaintenance were not carried out.

On older engines, plug gaps were typically around0.6 mm (0.24 in) but gaps on modern engines andspark plugs are more likely to be 0.8 mm and larger.These larger gaps are more suitable for enginesoperating on mixtures that are weaker than in the past.Of note is the fact that the larger gaps will inevitablyrequire greater voltages to initiate the arc (spark) and tomaintain the arc. It was therefore essential that themodern generation of ignition systems was developedto create and sustain the arc at the spark plug.

Electrode polarityA lower voltage is needed to produce a spark at the plugelectrodes when the centre electrode is negative inrelation to the HT circuit polarity. A hot surface emitselectrons and, because the centre electrode is the hotterof the two electrodes, there is a natural flow of electronsfrom the centre electrode to the earth electrode.Therefore, if the circuit is connected to give the samedirection of electron flow as the natural flow ofelectrons, this will assist in producing a spark for alower voltage.

Because the direction of electron flow in thesecondary winding depends on the polarity of theprimary winding, it is important that the primarywinding is connected correctly. Most coils will havemarkings to identify the positive and negative terminalsand these are usually marked as ‘15’ or ‘+ve’, and ‘1’ or‘–ve’. A correctly connected primary winding willtherefore provide a ‘negative’ spark.

Combustion takes place at different temperaturesin different engines. Spark plugs are designed tooperate at set temperatures so the correct heatrange plug must be used

Spark plug centre electrodes use materials such ascopper or even silver to aid temperatureconduction


Teaching/learning resourcesOnline learning material relating to powertrain systems:


Key

Poin

ts


Figure 2.49 Copper core central electrode

[647] Chapter 02 25/8/06 15:37 Page 76

ENGINE MANAGEMENT –PETROL

Cha

pter

3

3.1 INTRODUCTION TO ELECTRONIC PETROL INJECTION SYSTEMS


Introduction to electronic petrol injection systems

Petrol injection system examples (multi-point injection)

Single-point (throttle body) petrol injection

Direct petrol injection

Emissions and emission control (petrol engines)

Engine management (the conclusion)

Engine system self-diagnosis (on-board diagnostics) and EOBD

The following sections deal with sensors, ECUs andactuators that are covered in detail in other sections; seethe relevant sections in Chapters 1 and 2, as well asother sections within Chapter 3.

Electronic injection systems have progressedthrough many developments and variations, so itwould not be possible to cover all of these within thisbook. However, section 3.1 provides a generalunderstanding of the systems and their components,and sections 3.2–3.4 give examples of specific injectionsystems along with the latest developments. Note thatlater injection systems form part of enginemanagement systems, which are covered separately insection 3.6.

3.1.1 Fuel system developmentsFrom the carburettor to electronic injectionAs with ignition systems (see sections 2.2–2.4), fuelsystems have evolved progressively since the motorvehicle first appeared, but, with the introduction ofelectronic control, fuel injection has become thedominant method of fuel delivery for the petrol engine.

The carburettor (covered in Hillier’s Fundamentalsof Motor Vehicle Technology Book 1) was almostuniversally used on petrol engines through until thelate 1970s, when fuel injection systems began toappear on mass produced vehicles. Electronic injectionwas, however, used in the late 1960s to overcome

emission control problems with some vehicles intendedfor the American market. In the 1970s a Boschmechanical/hydraulic system (Bosch K-Jetronic)gained favour with many European manufacturers: thissystem tended to lead the way until the early 1980s,when a new generation of electronic injection systemsprogressively became more common. The Bosch K-Jetronic and its ECU controlled variant the KE-Jetronicsystem are dealt with in Hillier’s Fundamentals of MotorVehicle Technology Book 1.

It is suggested that the first application of fuelinjection was on the engine that was used by the WrightBrothers for the first manned flight of an aeroplane.However, simple carburettors were very much the onlypetrol delivery systems used on mass production vehicleengines for many years. Although diesel engines used amechanical injection system until fairly recently, petrolengines relied on carburettor systems because of cost,simplicity and the fact that there was no need for highpressure delivery of petrol.

A number of cars did use petrol injection, but thesewere generally racing cars and not production vehicles.One notable exception, however, was in the mid-1950s,when Mercedes-Benz used an adaptation of amechanical diesel pump on its racing engines and alsoon a limited production sports car (Mercedes 300SL); itis interesting to note that these systems wereadaptations of diesel mechanical pump systems, whichinjected fuel directly into the cylinder and not into theintake system, which has been the general principle for

[647] Chapter 03 25/8/06 15:37 Page 77

most electronic injection systems until recently.Mercedes continued to develop its use of fuel injectionwith petrol engines and, in the early 1960s, several ofthe company’s production vehicles were available withpetrol injection.

So, although fuel injection was not widely used,Mercedes and a number of vehicle and racing enginemanufacturers continued to develop the use of fuelinjection for petrol engines. Various mechanical andelectronic systems used from the early 1960s through tothe 1980s were developed to improve emissions forvehicles sold in the United States, although there wasan increase in the use of injection systems for theEuropean market for higher performance vehicles andfor more expensive vehicles.

It was, however, the introduction of Europeanemissions regulations that effectively forced the use offuel injection systems on almost all petrol engines bythe early 1990s. By this time, electronic control wasbecoming less expensive and it was therefore inevitablethat any mechanical systems would be replaced byelectronic injection systems.

Fuel injection and engine managementAlmost all modern petrol engines for light cars areequipped with engine management systems, whichcontrol the fuel injection, the ignition and many otherfunctions. The engine management system is thereforeeffectively a number of different systems controlled by asingle ECU (in most cases). The main systems (ignitionand fuelling) are covered separately within this book;the integration of these systems is dealt with in theengine management section (section 3.6).

3.1.2 Advantages of electronic petrolinjection

Improved efficiency and controlCompared with the carburettor, there are numerousbenefits provided by a fuel injection system; most ofthese will become obvious in the following sections.However, almost by way of a conclusion, it is certainthat an electronic fuel injection system provides anoverall efficiency of fuel delivery and control of fuelquantity that could not be achieved with a carburettor;the result is improved combustion efficiency, improvedengine performance (power), improved economy andreduced emissions.

Even when compared with later types of mechanicalinjection systems, electronic control provides a superiorcapacity to control fuel quantity and to embrace anychanges in fuelling needed to suit changing conditions.

However, a fully electronic fuel injection system alsoprovides the facility for integration and communicationwith other vehicle systems, such as the ignition andemission control systems.

Some specific advantages of electronic fuel injectionare covered in the following sub-sections.

Controlled pressure differenceA carburettor operates by using ‘pressure difference’. Inbasic terms, fuel in the carburettor float chamber (thefuel reservoir) is exposed to atmospheric pressure.Then, when air flows through the carburettor body, thiscreates a low pressure area around the venturi (locatedin the carburettor body). Therefore, the fuel in thechamber, which is at a higher pressure, flows to thelower pressure area. The airflow through to the enginethen carries the fuel with it thus resulting in a mixing ofair and fuel in the combustion chambers. In effect, apressure difference is created by the air flow, so varyingquantities of petrol can be drawn into the engine,depending on

● the speed of airflow● the size of the holes or jets through which the petrol

flows● the throttle opening (the angle of opening of the

throttle butterfly).

A fuel injector works on a similar principle of pressuredifference, but the fuel at the injector is at a higherpressure than that of the atmosphere and thereforemuch higher than in the intake manifold or in thecylinder on the induction stroke. The fuel pressure iscreated using a pump controlled by some form ofregulator, so it is always at a controlled pressure. Thereis, therefore, no need to create a low pressure by usinga venturi, because the fuel pressure is always higherthan that of the intake system or cylinder at the timewhen the fuel is delivered. Even on a turbocharged orsupercharged engine, where the intake system pressurecan be higher than atmospheric pressure, the fuelpressure will always be higher by a ‘controlled’ pressuredifference. Fuel therefore flows into the intake systemor into the cylinder in a controlled way due to thispressure difference.

Electronic petrol injection system pressures vary,but typically they are in the region of 2.5 to 3 bar. Thispressure, forcing the petrol through the injector nozzle,then assists in creating good atomisation of the petrol:mixing of air and petrol is therefore much moreeffective.

Figure 3.1 shows a simple carburettor and a fuelinjector, both of which rely on pressure difference as ameans of delivering fuel.

Intermittent injection to individual cylindersWith a carburettor, the flow of air creates the lowpressure that causes the petrol to flow from thecarburettor to the intake system. In theory, the petrolmixes with the air: therefore, as the air enters thecylinder, the petrol also enters the cylinder. However,there is an inevitable delay from the time that theinlet valve opens (the start of airflow of that cylinder)to the time when the increasing airflow draws petrolfrom the carburettor; for this and other reasons, it isnecessary to operate with an excess of petrol in themixture.

78 Engine management – petrol Fundamentals of Motor Vehicle Technology: Book 2

[647] Chapter 03 25/8/06 15:37 Page 78

With the more commonly used mechanical injectionsystems (Bosch K and KE-Jetronic) the injector is in theintake port just ahead of the intake valve (Figure 3.2),so petrol is delivered directly to each individual intakeport. However, the flow of petrol from the injectors onthe K and KE systems is continuous all the time that theengine is running, so most of the petrol is being injectedwhilst the inlet valve is closed. The petrol is therefore‘waiting’ in the intake port until the inlet valve opensand the air starts to flow into the cylinder. In reality, thepetrol flowing from the injector is atomised sufficientlyfor it to mix with the air in the port, so this is aconsiderable improvement over the carburettor.

One big advantage of electronic injection is that thepetrol injectors (located in the same place as inmechanical injection systems) are opened and closed atspecific times, which in theory reduces the waiting timebefore the petrol is drawn into the cylinder. Althoughon earlier generations of electronic injection, there wassome ‘waiting’ time, most modern systems inject atprecisely the correct time so that petrol typically leavesthe injectors just before the intake valve opens. Notethat some modern systems inject petrol direct into thecylinder during the intake stroke.


Figure 3.1 Pressure difference causing a flow of fuela in a carburettor b in a fuel injection system

Figure 3.2 Injector located just ahead of the inlet valve

The injection on fully electronic systems is notcontinuous, since the injectors open and closeintermittently at predetermined times, so it issometimes referred to as ‘intermittent injection’.

[647] Chapter 03 25/8/06 15:37 Page 79

Precise fuel controlElectronic injection systems are controlled by acomputer (ECU), which not only switches the injectorson and off at the appropriate time, but is also suppliedwith a wide range of information (by various sensors)that enable it to calculate the required fuel quantity forall operating conditions. The ECU is therefore able todeliver the correct fuel quantity at the correct time, andchange that quantity as the operating conditionschange. A schematic layout of a simple injection systemis shown in Figure 3.3.

The injection system ECU can also communicatewith other electronic systems, such as the ignition andemissions systems. In fact, there is considerablecommunication between the engine systems and chassissystems such as ABS and transmission, as well as withsystems such as the air conditioning. This level ofcommunication enables the fuel injection ECU to assessmany aspects of a vehicle’s operation, thus helping toimprove the accuracy and efficiency of the engine. Inturn, the fuel injection ECU can pass information toother vehicle systems, thus improving the efficiency ofthose other systems.

Management of these systems can be integrated sothat they are all controlled by a single ECU: this processis now almost universal for engines, where an enginemanagement system ECU controls virtually all enginefunctions. A similar philosophy is used for chassissystems, where the braking, and vehicle stabilitysystems are controlled by a single ECU. Since this ECUcommunicates with the engine management ECU, thenext step is to use a single ECU for all vehiclefunctions.

3.1.3 Main components and layoutof a multi-point, port typeelectronic system

Note: This section deals with the main componentsrequired for a simple multi-point injection system.Additional components used for emission control and

for other functions are covered in subsequent sectionsin this chapter. Single-point injection systems (oftenreferred to as ‘throttle body’ injection), where a singleinjector is used to deliver fuel to all of the cylinders, arecovered in section 3.3. Direct injection, where theinjectors deliver fuel directly to the cylinders, is coveredin section 3.4.

Two sub-systemsAn electronic petrol injection system effectively consistsof two sub-systems: an electrical/electronic system anda fuel delivery system. This section deals with the sub-systems for multi-point injection systems, where anindividual injector is used to deliver fuel to eachcylinder.

The main components of both sub-systems are listedbelow and are also illustrated in Figure 3.4.

Electrical/electronic system (see section 3.1.4)● Injectors – electrically operated fuel valves that,

when open, allow petrol to flow into the engine.● ECU – the computer that calculates the required

amount of petrol and then opens the injectors forthe appropriate amount of time.

● Sensors – provide the necessary information to theECU to enable it to calculate the fuel required fordifferent operating conditions.

Fuel system (see section 3.1.5)● Fuel pump – moves the fuel from the fuel tank to

the injectors; the pump provides an excess of fuel,which results in pressure being developed in the fuelsystem.

● Fuel filter – filters the fuel to remove dirt particlesthat could damage the system components or blockthe injectors.

● Fuel pressure regulator and fuel rail – the regulatorcontrols the pressure of the fuel; the fuel rail acts asthe distribution pipe to pass fuel to the injectors.

Note that, in addition to the main sub-systems, an idlespeed control system forms part of many injectionsystems. These systems are covered separately insection 3.1.6.


Figure 3.3 Layout of a simple electronic fuel injection systemThe schematic layout shows the ECU receiving information fromsensors and then controlling the fuel injectors.

[647] Chapter 03 25/8/06 15:37 Page 80


Figure 3.4 Simple electronic fuel injection system, showing thefuel and electrical/electronic sub-systems

3.1.4 Electrical/electronic system

Injector solenoid valvesPetrol injectors used on electronic injection systems arefuel valves that open and close to control fuel delivery.The injectors are solenoids with a needle valve attachedto the solenoid armature, so that, when current flows inthe solenoid winding, the magnetic field moves thearmature, which in turn moves the needle valve off itsseating and allows the fuel to flow through the nozzle.An example of an injector with a needle valve is shownin Figure 3.5a, and a different type of injector, with adisc rather than a needle is shown in Figure 3.5b. Notethat a fine mesh filter is used to filter out the very smallparticles that can damage the injector nozzle seating.

The injector solenoid valve is connected to a fuelsupply rail (Figure 3.6), or in some cases is locatedwithin the fuel rail. The fuel within the rail is regulatedat a predetermined pressure, which is altered to suitoperating conditions. However, the quantity of fueldelivered is largely controlled by opening the injectorsfor differing lengths of time.

Creating an atomised fuel sprayFuel is fed to the injector under pressure (typicallyaround 3 bar); because the fuel is under pressure, whenit flows out through the injector nozzle a spray of finelyatomised fuel is formed that is able to mix easily withthe air. To further assist with creating a spray ofatomised fuel, the injector needle and needle seatingare designed so that the fuel is forced to exit the injectorin a particular spray pattern.

In general, the fuel exiting the injector nozzle isdirected so that it sprays against the back of the inletvalve. There are now many different designs of nozzleused to create a spray pattern for the fuel as it flowsthrough and exits the nozzle. Depending on the locationof the injector in the inlet port, spray patternrequirements will differ: wide and narrow angle spraypatterns are used to suit the different engineapplications. Some injectors provide a dual spraypattern, designed to suit engines with two inlet valvesper cylinder; each of the fuel sprays is directed to theback of each of the inlet valves.

[647] Chapter 03 25/8/06 15:37 Page 81

Many petrol injection systems now inject petrol directlyinto the cylinder rather than into the intake port. Thesesystems are covered in section 3.4.

Speed of operation and opening timeWith most systems, the injectors will open either onceor twice for each operating cycle of a cylinder (seeinjector timing in the following paragraphs). Therefore,if an engine is operating at 6000 rev/min, each cylinderwill complete 3000 cycles in one minute or 50 cycles inone second. An injector might therefore open and closeas many as 50 times a second (once a cycle) or 100times a second (twice a cycle).

An injector needs to open for sufficient time to allowthe required amount of fuel to enter the intake port(or enter the cylinder with some types). Depending onthe amount of fuel required (for example, low loadand engine speed or high load), the injectors willtypically be open for durations of 1.5 ms to as much as15 ms.

Electronic control unit (ECU)As with an ECU controlled ignition system, the fuelinjection electronic control unit is the ‘brain’ of thesystem (Figure 3.7). The ECU controls the fuel injectorsin response to the information received from the sensors.


Figure 3.6 Injectors connected to the fuel rail

Injector connector

Solenoid coil

From deliverypipe

Rubber coilPlunger

TE type nozzle

Discharge orifices

(b)

Figure 3.5 Fuel injectorsa with a needle valveb with a disc valve

[647] Chapter 03 25/8/06 15:37 Page 82

An ECU contains a programmed memory, which, in aninjection ECU, contains data on how much fuel shouldbe injected under different operating conditions. Wheninformation is received from the sensors, the ECU refersto the programmed data and switches on the fuelinjectors so that they deliver the required amount ofpetrol (Figure 3.8). See sections 1.2 and 1.3 for anexplanation of ECU operation and construction.

Fuel map and basic fuel programSection 2.3.2 describes how a ‘map’ is used to provideignition timing values on modern ignition systems. Thesame process is used for ECU controlled fuel injectionsystems: a three-dimensional map provides the ECUwith the necessary references for the required quantityof fuel.


Figure 3.7 Typical appearance of a fuel injection ECU

Measuredvariable

Intake airquantity

Intake airtemperature

Batteryvoltage

Startingsignal

Engine speed

Engine temperature

Idle tofull load

Sensors Actuator Fuel supplysystem

Fuel tank

Fuel pump

Filter

Pressureregulator

Air flowsensor

Air temperaturesensor

Engine speedsensor

Engine temperaturesensor

Fuel injectionvalves

To engine

Signal processing

Throttle valveswitch

Figure 3.8 ECU processes in an injection systemIntake air quantity per unit of time and engine speed are the basic measured variables to which corrections are applied.

[647] Chapter 03 25/8/06 15:37 Page 83


Figure 3.9 Fuel mixture map

The map shown in Figure 3.9 gives references forfuelling based on engine load and engine speed; thefuelling references (vertical scale) are air:fuel ratios,which are expressed as lambda (λ) values. Lambdavalues are explained in section 3.5.1.

As noted above, the fuel map provides the ECU withreferences for fuel quantities for different operatingconditions. However, there are some basic orfundamental trends that dictate the overall mapping orprogram strategy.

In theory, the amount of fuel injected is matched tothe mass of air injected so that the stoichiometricair:fuel ratio is provided (the stoichiometric ratio is theideal ratio of air and fuel to provide completecombustion). However, minor variations in air:fuel ratioare necessary for different operating conditions, so theECU controls the injector opening time to suit thevarious conditions as listed below.

● Light load conditions – The injection duration islong enough to provide the quantity of fuel neededto give the theoretical stoichiometric ratio. Minorincreases or decreases in air mass (air drawn intothe engine) will result in minor changes ininjection duration.

● Acceleration and high load – The ECU willincrease the injection duration so that the fuelquantity increases to match the increase in air mass.However, under heavy load and acceleration, aslight excess of petrol is usually required (a richmixture), so the injection duration increases toslightly more than would be required to achieve theideal stoichiometric ratio.

● Cold running – When the engine is cold, the coldsurfaces of the intake port and combustion chambercan cause slight condensation of the fuel andprevent complete mixing of the fuel vapour and air.

The injection duration is therefore increased slightlyto provide a rich mixture, thus ensuring thatsufficient fuel is available to mix with the air.

● Idle – When the engine is idling, the air:fuel ratio onmodern engines is controlled at aroundstoichiometric or lambda 1. It was, however, normalon older engines for a slightly rich mixture to beprovided, which helped the engine to developsufficient power and to run smoothly.

● Deceleration – During deceleration, no power isrequired from the engine so fuel injection can becompletely cut off. Depending on engine speed andwhether the throttle is partially or completely closed(indicated by the throttle position sensor), the fuelinjectors can be completely switched off, or theinjection duration reduced, so that very little fuel isinjected. Careful programming of the ECU map isnecessary because, when the throttle is reopened,there can be a tendency for the engine to hesitate. Aprogressive cut-off and reapplication of fuelinjection are necessary to ensure a smooth transitionfrom deceleration to acceleration.

ECU switching the injectorsThe ECU contains the ‘power stages’ or powertransistors that are used to switch the injector electricalcircuits on and off. As with most computer controlledsystems, the ECU forms part of the earth circuit for theinjectors, so the ECU is switching on and off the earthpath. The injectors receive a battery voltage supply fromthe battery via a relay.

Having made the necessary calculations, the lowcurrent (and low voltage) microchips within the ECUwill provide an appropriate signal to the power stage,which will cause the power stage to complete the earthcircuit to the injectors, thus switching them on andallowing fuel to be delivered.

[647] Chapter 03 25/8/06 15:38 Page 84

Injector timing on earlier systems(simultaneous injection)The ECU will switch on the injectors (by completing theearth circuit) at a predefined time in the engineoperating cycle. On many earlier electronic injectionsystems (typically through until the early 1990s), theinjectors were all opened at the same time (on four-cylinder engines), which is referred to as ‘simultaneousinjection’. With six-cylinder engines the injectors weregenerally operated in two groups of three injectors; witheight-cylinder engines the injectors were operated intwo groups of four; and with 12-cylinder engines therewere four groups of three injectors. All of the injectors ina group would open and close at the same time.

It was also usual for all of the injectors to be openedtwice for every engine cycle, so half of the requiredquantity of fuel was delivered each time the injectorsopened. On these older systems, the injector timingwas therefore not perfect because, while one cylindermight have its injector opening when the inlet valvewas open, on the rest of the cylinders, the inlet valveswould be closed. As previously noted, the injectedpetrol would therefore be ‘waiting’ for a short periodbefore it was drawn into the cylinder. Figure 3.10shows a four-cylinder engine where the injectors areopened simultaneously twice for every engine cycle.

On earlier systems, the injector opening would betriggered by the ignition system, but the injection ECUwould switch on the injectors on alternate ignitionpulses: i.e., if the firing order was 1,3,4,2, then theinjectors would open when the ignition system wasfiring numbers 1 and 4 cylinders or numbers 3 and 2cylinders. Therefore, in a complete engine cycle (twocrankshaft revolutions), in which all cylinders wouldhave fired once, the injectors would have opened twice.

Injector timing on later systems (sequentialinjection)With modern systems the injectors are usually openedindividually in sequence (to match the engine firingorder); this is known as sequential injection. Theinjectors are typically opened just prior to the inletvalve opening. All the required fuel is thereforedelivered in one ‘opening’ of the injector (Figure 3.11).However, there are occasions where a very largequantity of fuel is required, for example during full loadacceleration, where the injectors can be opened twicefor every operating cycle (half the fuel quantity isdelivered at each opening).

Although it is possible to use a signal from theignition system to trigger sequential injection, manysystems use separate sensors to identify one of thecylinders, for example cylinder number 1; the ECU thenuses this signal as a master reference and operates theinjectors in sequence at the appropriate times. Thesensor is referred to as a ‘cylinder identification sensor’or ‘phase sensor’: these are usually either inductive orHall effect sensors.

The camshaft rotates once for every engine cycle(while the crankshaft rotates twice), so the cylinderidentification sensor is usually located adjacent to thecamshaft. Therefore, a single reluctor tooth or triggerlug attached to the camshaft could then cause aninductive sensor to provide a single reference signal(see section 2.2.5). Alternatively a Hall effect rotor (seesection 2.2.6) attached to the camshaft could have asingle ‘cut out’, thus producing a single reference pulse.A crankshaft speed/position sensor provides thenecessary crankshaft angle and speed information.


ECU+5V

Simultaneous injectionpattern

Cylinder 1

Cylinder 2

Cylinder 3

Cylinder 4

0° 180° 360° 540° 720°

Crankshaft angle

Intakestroke

Fuel injection

Ignition

Injection timingFigure 3.10 Injector timing forsimultaneous injection on a four-cylinder engine

Independent injectionpattern

Cylinder 1

Cylinder 2

Cylinder 3

Cylinder 4

0° 180° 360° 540° 720°

Crankshaft angle

Intakestroke

Fuel injectionIgnition

Injection timing

ECU+5V

900° 1080° Figure 3.11 Injector timing forsequential injection on a four-cylinder engine

[647] Chapter 03 25/8/06 15:38 Page 85

Some systems used a trigger signal provided by a sensorattached to a spark plug lead. The sensor generated asmall electrical pulse that was used by the ECU as themaster reference signal.

Injection durationThe ECU uses the information from the sensors tocalculate the operating conditions for the engine andthus enable the correct volume of fuel to be injected. Intheory, the mass of air entering the engine is the mainitem of information required by the ECU to enable thecorrect quantity of fuel to be calculated. The air masscan be measured with an air mass sensor but furtherinformation is used to assist the ECU in the calculationprocess. This additional information, supplied by othersensors, is covered in the following sections.

When the ECU has calculated the quantity of fuel tobe delivered (effectively by noting the information fromthe sensors and then referring to the programmed fuelmap within the memory), it will then switch on theinjectors for an appropriate length of time (the injectionduration). The duration will vary with the system’sdesign and operating conditions, such as engine load,engine speed, temperature, etc., but typical values arebetween 1.5 ms and 15 ms.

Injectors are produced with differing nozzle sizes, sodifferent injectors will allow different quantities of fuelto flow through the nozzle for a given openingduration. The different nozzle sizes are produced to suitlarger and smaller engine cylinders, which will requirecorrespondingly larger or smaller quantities of fuel tobe delivered. A large nozzle injector used in a largecylinder will have a similar opening duration to a smallnozzle injector used in a small cylinder.

Control signalThe ECU functions as the switch in the injector earthcircuit. The power stage within the ECU is the switchingcomponent, and when the ECU calculates that theinjector should be switched on for a specific length oftime, the power stage will complete the injector earthcircuit for the appropriate time period.

As with any switch that is located in the earthcircuit, when the switch is ‘closed’ there is a completecircuit, which means that the earth circuit voltageshould be 0 volts. When the switch on the earth circuitis open, current does not pass to earth and there is an‘open circuit voltage’ available at the negative terminalof the injector and at the input terminal of the earthcircuit switch.

On an injector circuit, therefore, assuming that a12 volt power supply is connected to the injector, theinput voltage to the injector positive terminal will be12 volts. When the earth switch is closed (completingthe circuit through to earth), then the voltage at thenegative terminal of the injector and at the inputterminal of the power stage or switch will be 0 volts.The switching action of the power stage thereforeresults in the voltage on the earth path (negativeterminal of the solenoid) switching between 0 and

12 volts. In effect, this on/off switching action is thecontrol signal produced by the ECU: this control signalis a simple on/off digital signal. The length of the onpulse of the control signal will dictate the duration ofthe opening of the injector.

Two examples of control signals are shown in Figure3.12. These signals are typical of control signals whenobserved with an oscilloscope.

The example in Figure 3.12a shows a signalproduced when the power stage switches on and off theearth circuit. The spike at point C on the signal isproduced when the circuit is switched off, which causesa ‘back EMF’ to be produced in the solenoid winding, i.e.the rapidly collapsing magnetic field causes a voltage tobe induced in the winding. The duration of injection isdictated by the pulse width B. Therefore points A, B andC on the pulse can be described as follows:

A = injector earth circuit switched off. The open circuitvoltage can be measured at the injector negativeterminal.

B = injector earth circuit switched on. The ECUcompletes the earth circuit so the voltage in thiscircuit is 0 volts. The width of this section of thesignal (pulse width) dictates the opening time orduration of opening of the injector.

C = injector earth circuit switched off. At the end ofduration B, the ECU switches off the earth circuitand back EMF is induced within the solenoid. TheEMF can reach figures in the region of 40 to60 volts on some systems, as indicated at point C.

The example in Figure 3.12b is typical of systemswhere current control is used by the ECU in a two-stage control process. Stage 1 is where full current isallowed to flow through the injector earth circuit (B2),which allows rapid opening of the injector; theduration of this full current period can be very short,e.g. 0.5 ms. Stage 2 is where current limiting isimplemented by the ECU so that the current flowthrough the earth circuit is limited (B3); however,sufficient current will still flow to keep the injectoropen for the required period. The injector duration isvaried by altering the width of the injector controlsignal pulse in stage 2 (B3).

A = Injector earth circuit switched off. The opencircuit voltage can be measured at the injectornegative terminal.

B1 = Total duration of control signal (made up of B2 +B3).

B2 = Injector circuit is switched on. At this point, theECU allows full current to flow through the earthcircuit, thus rapidly opening the injector. At theend of period B2, the current is limited by theECU, which would in theory cause the injector toclose.

B3 = When the current has been limited at the end ofB2, the ECU allows a reduced current to flowwhich is sufficient to keep the injector open. The


[647] Chapter 03 25/8/06 15:38 Page 86


width of the pulse at B3 alters, which controls theduration of opening for the injector.

C = Injector earth circuit switched off. At the end ofduration B, the ECU switches off the earth circuitand back EMF is produced within the solenoid.The EMF can reach 40 to 60 volts on somesystems, as indicated at point C.

SensorsThe main objective of any petrol delivery system is toprovide the correct mixture of air and petrol so that thecombustion process is efficient and produces maximumpower from the mixture. To achieve this, the fuelinjection system must provide the correct quantity ofpetrol to suit the mass of air being drawn in by theengine, i.e. the ratio of petrol to air must be correct.

50V

45

40

35

30

25

20

15

10

5

00 1

B

C

2 3 4 5

Injector (multi-point)

6 7 8 9 10ms

(a)

A

50V

45

40

35

30

25

20

15

10

5

00 1 2 3 4 5

Injector (single-point)

6 7 8 9 10ms

(b) B1

B2 B3

C

A

Figure 3.12 Injector control signalsa A signal produced when the power stage switches on and off the earth circuitb A signal typical of systems where current control is used by the ECU in a two-stage control process

[647] Chapter 03 25/8/06 15:38 Page 87

If we therefore assume that, at all times, the air:fuelratio is stoichiometric (as discussed in Book 1), then themixture delivered should be in the ratio ofapproximately 14.7 parts of air to 1 part of petrol byweight. This is the ratio that, in theory, will providecomplete combustion of the mixture.

However, there are numerous situations thatinfluence the air:fuel ratio: this means that the mixturecan vary above or below the stoichiometric value, i.e.the mixture can have a slight excess of fuel or an excessof air (rich or weak) to suit the conditions andrequirements. A simple example is with cold startingand running, where a slightly rich mixture is required.In other situations, a slightly weak mixture might beprovided for better fuel economy.

In reality, with modern engines, emissionsregulations force the use of a stoichiometric air:fuelratio for a high percentage of engine operatingconditions. However, the same rule applies withmodern and older engines: the injection system ECUmust initially calculate the fuel requirement based onthe mass of air flowing into the engine.

The main information required by the ECU istherefore a measurement of the mass of air. Othersensors are, however, required to enable the ECU to‘fine tune’ the air:fuel mixture. As noted previously,there are a number of factors that affect the air:fuelratio, such as:

● temperature● emissions control● engine load● driver intentions● engine design

and other factors also slightly influence the mixturefinally delivered to the engine. Therefore manyadditional sensors are used to give the ECU sufficientinformation to enable it to finely adjust the fuelquantity delivered by the injectors. The following listhighlights only those sensors that might be used on abasic electronic injection system.

Note: Additional sensors are covered in theexamples of injections systems in section 3.2, whilstother sensors used on modern systems are dealt within section 3.5 (emission control); also refer toChapter 1.

Airflow and mass airflow measurementAirflow sensors (discussed later in this chapter) can beused to measure and transmit a signal to the ECUrelating to either the volume or the mass of air that isflowing through the intake system at a given time.However, the ECU must calculate what mass of air isflowing to each cylinder at any given time, so it alsorequires engine speed information. Some examples ofairflow and mass airflow sensors are shown inFigure 3.13.

Air mass and engine speed are the fundamentalitems of information required by the ECU. Although

some systems use an air ‘mass’ sensor, others usedifferent sensors such as manifold absolute pressuresensors (MAP sensors), which measure the manifolddepression; in this case, the ECU uses the MAP sensorinformation along with engine speed, air temperatureand other information to establish the basic fuelrequirements.

Coolant temperature (Figure 3.14a)The engine coolant temperature information enablesthe ECU to alter the fuel quantity (thus altering theair:fuel ratio) so that, when the engine is cold, an excessof fuel (a rich mixture) can be provided. The ECU canslightly alter the fuel quantity and mixture over thewhole range of operating temperatures, thus allowing‘fine tuning’ of the mixture.

Air temperature (Figure 3.14b)The air temperature sensor information helps incalculating the air mass, because air density changeswith temperature. Many air temperature sensors areincorporated within the airflow sensors. Since a highair temperature can cause ‘pinking’ or pre-ignition,the ignition timing might be altered by the ignitionECU to correct this, but it is also possible that theinjection ECU might slightly alter the fuel quantity ifnecessary.

Throttle positionOn early basic electronic injection systems, a simple‘throttle switch’ was used to indicate when the throttlewas in the closed position (idle). The switch alsoindicated when the throttle was around 60% open,which was a sign that the engine was under load. Themost widely used type of throttle switch on earliersystems contained two sets of contacts, which closedand opened at the relevant time as the throttle wasopened and closed. For both idle and load positions,the ECU provided a slight enrichment of the mixture,which helped stabilise the engine at idle and allowedadditional power to be developed under load. Notethat later engines, with improved emission control,operate with weaker mixtures at idle and under loadconditions; the programming of the ECU and theinformation required are altered to suit theserequirements.

Later throttle position sensors use a potentiometeror variable resistor instead of switches. With apotentiometer based sensor, it is possible to send avarying voltage analogue signal to the ECU; the signalindicates all throttle opening angles and the ECU canalso calculate the rate at which the throttle is beingopened and closed. The ECU is therefore able to alterthe fuelling to suit the minor and major changes inthrottle position (load changes). The ECU can alsoassess the driver’s intentions, such as the intention torapidly accelerate, by detecting the speed or rate atwhich the throttle is being opened.

Figure 3.15 shows an example of a throttle switchand a throttle position potentiometer.


[647] Chapter 03 25/8/06 15:38 Page 88


Figure 3.13 Examples of airflow and mass airflow sensorsa Flap type airflow meterb Hot wire air mass meterc MAP sensor

Ignition trigger or crankshaft position triggerAs noted previously, early injection systems usedignition pulses as a means of triggering the injectors.These earlier injection systems were usually fitted toengines that also had earlier designs of electronicignition with inductive or Hall effect triggers located ina distributor, so it was common practice to provide asignal from the ignition system to the injection ECU.This signal was often taken directly from the coilnegative terminal, which would be the same terminalused to provide a signal to a rev. counter. On someignition systems, the signal might have been a digitalsignal provided by the ignition module.

Figure 3.14 Examples of temperature sensorsa Coolant temperature sensorb Air temperature sensor

Intake manifold pressure

Filter

Vacuumchamber

Silicon chip

(c)

[647] Chapter 03 25/8/06 15:38 Page 89


With those later ignition and injection systems, acrankshaft position sensor was often used. The sensortransmitted the crankshaft speed and position data tothe ignition system ECU, and the ignition ECU thentransmitted a digital speed signal to the injection ECU.

Whichever triggering system is fitted, the ECU canuse the information as a triggering signal for openingthe injectors (the injector timing on simultaneousinjection systems). The ignition trigger signal alsoprovides the ECU with engine speed information, whichcan be used to help calculate fuel requirements.

Figure 3.16 shows the process of an injection systemthat is being triggered by an early ignition system.

Figure 3.15 Measurement of throttle positiona Throttle valve switchb Throttle position potentiometerc Output from potentiometer as throttle valve opens

Other sensorsThe sensors described so far provide the essentialinformation to enable a simple injection system tooperate. In fact, many early systems used only thesesensors.

The need for improved emission control andefficiency led to other sensors being added, which arecovered in the following sections of this chapter.

3.1.5 Fuel systemFuel pumpProviding sufficient fuel flowThe fuel pump moves the fuel from the fuel tankthrough to the injectors. The pump must providesufficient fuel for the engine to operate at full load: itdelivers typically between 1 and 2 litres of fuel eachminute (depending on engine size).

The fuel pumps on most systems operate at full rateall the time: there is no variation in the amount of fueldelivered by the pump, irrespective of engine speed andload. However, at low engine speeds and loads only asmall amount of fuel is used, so the excess fuel flowsback to the fuel tank.

Pressurising the systemLiquids cannot be compressed, but they can exist in anenclosed system under pressure. With petrol injectionsystems, a high volume of fuel delivered by the pumpflows to the injectors and, with only a small amount offuel able to escape through the injectors, thiscontinuous flow of fuel causes the system to build uppressure. Although excess fuel is allowed to return tothe tank, a regulator valve is used (see below) to controlthe pressure in the system. In theory therefore, the fuelis always held in the system at a constant pressure (seethe following paragraphs dealing with pressureregulators). This combination of a high volume of fueldelivered by the pump and the pressure regulatorensures that the fuel flowing through the injectors is ata pressure that causes atomisation of the fuel when itexits the injectors.

Construction and locationWith most modern systems the fuel pump is located inthe fuel tank, although on many earlier systems thepump was mounted externally (Figure 3.17). In somecases, two pumps are used: one which initially movesthe fuel from the tank to the main pump (which mightbe located too high in the vehicle for the fuel to flowinto it), the main pump then moves the fuel through thefuel injection system.

The main pump is driven by an electric motor,which turns a pumping element. A number ofdifferent types of pumping element have been used:the examples shown in Figure 3.18 cover bothpositive displacement (a) and flow type pumpingelements (b).

[647] Chapter 03 25/8/06 15:38 Page 90

One disadvantage of positive displacement types is thatpulses are produced by the pumping action, which cancause noise and vibration in the fuel system. Flow typepumps provide much quieter operation and aretherefore more widely used. However, with the latestgeneration of direct injection systems, positivedisplacement pumps are again becoming more widelyused.

Fuel filterThe fuel filter is located in the fuel circuit after the fuelpump. Although different manufacturers use slightlydifferent construction, the example shown in Figure3.19 is typical of most filters.

The filter usually consists of a fine paper elementand a strainer, which can retain any larger particles.Filters are often constructed in such a way that fuel


Figure 3.16 Ignition system used for triggering an earlier fuel injection system

Ignition ECU

Injection ECU

Fuel injectors

Injection ECU is then able to provide timed injection and calculate the

necessary fuel requirement

Sensors

Sensors

Sensors

Crank sensor

Injection ECU receives speed /trigger signals from ignition ECU

[647] Chapter 03 25/8/06 15:38 Page 91


Figure 3.17 Examples of fuel pumps a Pump located in the fuel tankb An external pump

Figure 3.18 Fuel pump with positive displacement and flow typepumping elementsa Roller cell pumpb Flow type pump

Figure 3.19 Fuel filter

(a)

1. Fuel tank2. Electronic fuel pump3. Fuel level sensor4. Float

Relief valve

Pumpspacer

Magnet

Brush

Check valve

SilencerDiaphragmchamber

Armature

Discharge

Rotor

Filterintake

(b)

should flow in one direction only; markings are usuallyprovided to indicate the fuel flow. Incorrect fitment canresult in collapse of the paper element.

Although fuel filters are designed to prevent anyimpurities and dirt from reaching the injectors,additional fine mesh filters are often also fitted in theinjector inlets.

[647] Chapter 03 25/8/06 15:38 Page 92


Fuel pressure regulator and fuel railThe fuel rail is simply a distribution pipe that allowsfuel to flow to all of the injectors. Fuel flows from thetank through the pump and filter to the fuel rail,where the injectors receive their supply of fuel. Untilrelatively recently, on most systems the fuel rail had anexcess fuel pipe connection, which allowed unused orexcess fuel to flow back to the fuel tank. A pressureregulator in the fuel rail maintains the pressure in thefuel rail at the required value: excess fuel flows fromthe regulator back to the fuel tank. An example of afuel system with an excess fuel return is shown inFigure 3.20a.

Relocation of the excess fuel return pipeA disadvantage of returning excess fuel from the fuelrail to the fuel tank is that the fuel carries heat from theengine bay back to the fuel in the tank. The temperatureof the fuel in the tank therefore rises, which causesevaporation, resulting in fumes that can escape to theatmosphere. Emissions regulations require emissionsfrom the fuel tank to be controlled. A recent change infuel system design helps to reduce evaporative tankemissions by eliminating the return pipe from the fuelrail. The return pipe is relocated so that it collects theexcess fuel from a position much closer to the fuel tank:it is connected just after the fuel filter. These systemsare often referred to as ‘returnless systems’, although inreality a return pipe is still used. An example of areturnless system is shown in Figure 3.20b.

Pressure regulatorOne principle of operation of electronic injectionsystems is that if the fuel pressure remains constant, i.e.the fuel exiting from the injector is always at the samepressure, then any variation in fuel quantity delivered tothe cylinders can be controlled by altering the durationof the open time of the injectors. In effect, the injection‘on time’ or injection duration is the only methodthrough which the fuel quantity is regulated, and theduration is controlled by the ECU in response toinformation from the sensors.

A fuel pressure regulator is therefore fitted to thefuel delivery system to ensure that the fuel pressureremains constant.

Fuel pressure and engine intake pressureOne factor that must be considered in maintaining aconstant fuel pressure is the change that occurs in theengine intake system pressure.

While an engine is operating, the pressure in theintake system (manifold and ports) varies with changesin engine load, engine speed and throttle openingangle. At one extreme, when the throttle is initiallyopened, the intake pressure is almost the same asatmospheric pressure (approximately 1 bar). At theother extreme, for example when the engine is at highspeed and the throttle is suddenly closed, the intakepressure will fall (often referred to as the intakevacuum) due to the restriction of the closed throttle andthe strong suction created by the cylinders. The intakepressure can reduce to a typical value of 0.5 bar, so thepressure will have fallen by 0.5 bar.

If the fuel injection pressure remained constant at atypical value of 3 bar, then the difference between theinjection pressure and the intake system pressure wouldvary as the intake system pressure varied. The extremesof effective or true injection pressure would therefore beas follows:

1 intake system pressure = 1 barinjection pressure = 3 barpressure difference = 2 bar.The injection pressure will therefore be 2 bar.

2 intake system pressure = 0.5 barinjection pressure = 3 barpressure difference = 2.5 bar.The true pressure of injection will therefore be2.5 bar.

Therefore, if the true injection pressure were allowed tovary with the intake system pressure, the quantity offuel delivered would also vary: for the same duration ofinjector opening time, if the true injection pressureincreased, the amount of fuel flowing through theinjector would also increase.

To overcome this problem, most pressure regulatorsare fitted with a pressure/vacuum pipe connection tothe intake system. When the intake pressure reduces,this lower pressure acts on the pressure regulator, whichin turn reduces the pressure in the fuel system. In effect,

Figure 3.20 Fuel delivery systemsa Fuel system with excess fuel return from fuel railb Returnless fuel system

[647] Chapter 03 25/8/06 15:38 Page 93

as the pressure in the intake system rises and falls, thepressure in the fuel system also rises and falls by thesame amount.

The relationship between intake system and fuelsystem pressure will therefore be as follows:

1 intake system pressure = 1 barinjection pressure = 3 barpressure difference = 2 bar.The true pressure of injection will therefore be2 bar.

2 intake system pressure = 0.5 barinjection pressure = 2.5 barpressure difference = 2 bar.The true pressure of injection will therefore be2 bar.

The following statement can therefore be made:

‘Fuel system pressure is maintained by the pressureregulator at a constant pressure relative to intakesystem pressure.’

Figure 3.21 shows a typical pressure regulator, whereintake system pressure (manifold pressure) is connectedto the regulator. The operation is as follows.

● When fuel is delivered to the fuel rail and regulator,it will flow to the underside of the regulatordiaphragm. A spring acts on the diaphragm andvalve assembly, which holds the valve closed.However, the fuel entering the system will build uppressure to a level where it will cause the diaphragmto lift against the spring, and therefore the valve willopen.

● As soon as the valve opens, the excess fuel willescape through the return port, which will cause thepressure to reduce; the valve will therefore close. Inreality, the valve is constantly oscillating betweenthe open and closed positions, with the result that

the pressure is maintained at a value that isdependent on the strength of the spring.

● When the manifold pressure reduces, this lowerpressure acts on top of the diaphragm and helps tolift the diaphragm against the spring. Therefore thefuel pressure will not need to be so high before itlifts the diaphragm and opens the valve.

Therefore, when there is a low pressure in the intakemanifold, the fuel pressure beneath the diaphragm willbe lower when the valve opens. When the intakemanifold pressure is higher, then the fuel pressurebeneath the diaphragm will need to be higher to openthe valve.

3.1.6 Idle speed controlStalling at idle speedWhen an engine is operating at idle speed, many of theengine’s processes are relatively inefficient. Forexample, the air flows through the intake system at alow speed, which does not help the air and fuel to mix.For many engines therefore, especially older designs,emission levels at idle speed were relatively high (as apercentage of the total exhaust gas) and the powerdeveloped by the engine was very low.

The regulations are very much focused on idle speedemissions, so for many years it has been necessary tooperate engines on a weak mixture or at stoichiometricair:fuel ratios. However, these air:fuel ratios at idlespeed do not enable the engine to produce good poweroutputs leading to a tendency for the engine to stallwhen any load is applied.

To overcome this problem, a means of controllingthe idle speed is used which relies on regulating theairflow into the engine to ensure that the engine doesnot stall or run too slowly at idle. One option is to relyon the driver to control the idle speed with the throttle;this is, of course a very imprecise and impracticalmethod. Therefore automated systems are used toregulate the airflow. In effect these systems areautomatic throttle controls, and in some cases, they dophysically control the throttle butterfly. However, manysystems regulate the air flowing through a bypass portby using an ECU controlled air valve.

Stepper motor idle valvesSee section 1.8.2 for additional information on steppermotors.

A stepper motor is effectively an electric motor thatcan be stopped or positioned at selected angles ofrotation. Some stepper motors can be controlled so thatthe motor or armature will rotate and stop inincrements of less than 1 degree of rotation.

Stepper motors are generally used in one of twoways to control idle speed: either by acting on an airbypass port or by acting on a linkage that connects tothe throttle butterfly (throttle valve or plate).Figure 3.22a shows a stepper motor used to regulate the


Figure 3.21 Fuel pressure regulator

[647] Chapter 03 25/8/06 15:38 Page 94

air through an air bypass port and Figure 3.22b shows astepper motor acting on a throttle butterfly linkage.

The control signals for stepper motors are generallyconventional digital control signals: they are on/offpulses. A stepper motor has a number of windings, eachof which can be provided with an on/off control signalfrom the ECU, which enables the armature to bepositioned accurately.

Rotary idle valve using partial rotation motorsAlthough this type of valve also uses a type of electricmotor, the rotation of the armature is restricted bymechanical stops or limiters, so that it can only partiallyrotate. Bosch produces the most widely used type ofrotary idle valve: see Figure 3.22c. The armature isconnected to a valve that controls the air flowingthrough a bypass port (Figure 3.22c); the assembly is


Figure 3.22 Electric motors used to control idle speeda Construction of stepper motor idle control valveb Stepper motor acting on throttle linkagec Rotary idle valve

[647] Chapter 03 25/8/06 15:38 Page 95

usually remotely located, so air pipes connect the valveassembly to the intake system throttle body.

When the throttle is in the idle or closed position,air is able to bypass the throttle butterfly by flowingthrough the air pipes and the rotary idle valve; the idlespeed is therefore dictated by the valve position.

Earlier versions had two windings in the motor. Bytransmitting a control signal through each winding, itwas possible to rotate the armature clockwise or anti-clockwise, depending on which control signal providedthe higher current flow through the winding. Latertypes use a spring to rotate the armature in onedirection, with a control signal flowing to a windingthat will then rotate the armature against the spring;this type is shown in Figure 3.22c.

The ECU alters the duty cycle of the digital controlsignal (see section 1.9), which alters the averagecurrent in the circuit and the winding. This then causesa stronger or weaker magnetic field to be produced,which results in a stronger or weaker force to opposethe armature spring.

Solenoid idle valvesSeveral system manufacturers have used solenoidoperated valves to regulate the air flowing through an


Air intakechamber

Throttlevalve

Inputsensors

Microprocessor

ECU

Decreasedair pressure

Solenoid valveIgnition switch

Battery

Increasedair pressure

Figure 3.23 Solenoid type idle speed control valves From ECU

From aircleaner side

Valve

Solenoid

To intakeair chamber

air bypass port. The principle of regulating air in a portis the same as for stepper motor systems (see above),but a linear solenoid is used to control the valve insteadof a rotary motor (Figure 3.23).

Although there are various designs of solenoidsystems, the basic principle relies on a solenoidarmature being connected to an air valve. When thesolenoid armature moves, it increases or decreases theaperture in the air bypass port.

The solenoid is usually spring loaded in onedirection; current flowing through the winding causesthe armature to move against the spring. By increasingor decreasing the current in the circuit and winding, it isthen possible to create a greater or weaker magneticfield, which creates a stronger or weaker force on thearmature. Altering the current in the circuit thereforemoves the armature (against the force of the spring)giving the required valve position to regulate the airflowin the bypass port and the idle speed.

Although there are variations in the control signalsused for solenoid valves, a digital signal can be usedwhere the duty cycle is altered, which in turn alters theaverage current flow (see section 1.9 for additionalinformation on control signals and solenoids).

[647] Chapter 03 25/8/06 15:38 Page 96


3.2 PETROL INJECTION SYSTEM EXAMPLES (MULTI-POINT INJECTION)

This section explains the operation of different types ofelectronic fuel injection systems. The examples are allmulti-point port injection systems: i.e. they haveindividual injectors for each cylinder, with the injectorslocated in the intake ports. Section 3.3 covers single-point systems.

3.2.1 Example 1: Simple multi-pointinjection with airflow sensor

System components and layoutFigure 3.24 shows the main components used in theBosch LE system, which was widely used by manyvehicle manufacturers and was a system from whichmany others evolved. In addition to the main injectionsystem components, an auxiliary air valve is used toprovide a fast idle speed when the engine is cold. Theair valve is not controlled by the ECU and its operationis explained later in this section. The LE2 systemtherefore has several sensors but only one set ofactuators: the injectors. The fuel delivery system on theBosch LE is identical to the example illustrated in Figure3.20, so it is not described again in this section.

Airflow sensor with combined air temperaturesensorThe airflow sensor provides the ECU with an analoguesignal that indicates the volume of air being drawn intothe engine. Air density changes with temperature, so anair temperature sensor is built into the airflow sensorassembly. The ECU therefore receives airflow and airtemperature information.

Airflow sensor operationThe airflow sensor has a flap or vane that is forced tomove when the air flows through the sensor body(Figure 3.25). The flap is connected to a hinge or shaft,so the angle of the flap increases as the airflowincreases.

A potentiometer (variable resistor) is fitted to thesensor assembly, and the potentiometer ‘wiper’ or‘slider’ is connected to the shaft. Therefore, when theflap moves, the potentiometer wiper moves around theresistance track of the potentiometer. This changes thevoltage at the wiper, with the magnitude of this changedepending on the flap position. The voltage reading atthe wiper is sent to the ECU, which provides the ECUwith an indication of airflow.

PotentiometerOne of the disadvantages of the flap system is thechange in angular movement that occurs with increasedairflow. When the airflow is low, the flap is almost atright angles to the airflow and therefore the force actingon the flap is relatively high; any small change inairflow will cause a relatively large change in the flapangle. However, when the airflow is already high, thehigher forces acting on the flap will have pushed it to aposition where it is almost in line with the airflow(almost lying flat in the sensor body), so any furthersmall increases in airflow will not greatly affect theangle of the flap: it will move only a little further.

When the flap is almost at right angles to theairflow, the voltage change at the potentiometer will bequite large for a small change in the airflow. When theflap is almost in line with the airflow, however, smallchanges in airflow will result only in very small changes

Figure 3.24 Bosch LE injection system

[647] Chapter 03 25/8/06 15:38 Page 97


Figure 3.25 Cutaway views of vane type airflow sensora Air sideb Electrical connection side

in voltage, which does not provide sufficientinformation to the ECU. The ECU ideally requires largechanges in voltage to assess the airflow accurately.

The potentiometer used on the airflow meter istherefore more complex than conventionalpotentiometers. A thick film resistance track is used,made of several segments, each with a differentresistance. The resistance of the segments is designed tocompensate for the reducing angular movement of theflap as the airflow increases: as the wiper moves acrossthe track, the output voltage is progressive and linear.

On this type of airflow sensor, 12 volts is applied tothe potentiometer, and, as the wiper moves across theresistance track, the voltage at the wiper changes fromtypically around 5 volts at low air volumes to around9 volts at high air volumes.

Damping chamberA second flap (attached to the first flap) is positioned ina small chamber (referred to as a damping chamber).Air is drawn or induced into the engine in pulses orwaves (each cylinder creates a single strong pulsegiving as many pulses in one engine cycle as there arecylinders), so the first flap will also tend to pulsatewhen the airflow passes through the sensor. The secondflap is also exposed to the pulsing action of the airflow,but the airflow is directed against the second flap insuch a way that the pulsing on the second flap cancelsthe pulsing of the first flap.

If the compensating flap and damping chamber werenot used, the pulsations caused by the airflow wouldalso cause the signal from the sensor to pulsate.

Air temperature (see section 1.5.1)An air temperature sensor is incorporated within theairflow sensor. The temperature sensor is a thermistor,which is a resistor that changes in resistance withchanges in temperature. Because the sensor forms partof a series resistance circuit that has a reference voltageapplied to it, when the temperature changes, theresistance and voltage in the circuit also change. Thechange in voltage is used as a signal to the ECU.

The ECU uses the signal for the change in airtemperature in conjunction with the airflow signal(because air mass for a given airflow changes withtemperature).

Mixture adjustmentA bypass port is provided on older airflow sensors sothat mixture adjustments can be carried out at idlespeed. This facility is no longer required with modernengine management systems, but on older engines itwas needed to ensure that the idle emissions werewithin specified limits and to enable the engine to idlesmoothly.

Figure 3.25 shows the bypass port, which has anadjusting screw at one end. If the adjusting screw isscrewed fully into the port it will block the bypass port,which will force all of the air being drawn into the

[647] Chapter 03 25/8/06 15:38 Page 98

engine to flow through the main intake port andtherefore move the sensing flap. When the adjustingscrew is unscrewed, a small amount of air is able to flowthrough the bypass port, so the flap will not be affectedby all of the airflow: the sensing flap will move backslightly, which will reduce the signal voltagetransmitted to the ECU. Altering the adjusting screwwill therefore affect the amount of air flowing throughthe bypass port and thus the amount of air affecting thesensing flap. The signal voltage will alter, which causesthe ECU to adjust the quantity of fuel being injected.Adjusting the bypass port screw will therefore affect thefuel quantity and the mixture at idle speed.

Coolant temperature sensor(See section 1.5.1.) The coolant temperature sensor ispositioned (usually in the cylinder head) so that it canmeasure the temperature of the engine coolant. A signalfrom the sensor is transmitted to the ECU so that thefuel quantity can be altered for cold running (byenriching the mixture) as well as for other minorvariations in fuelling that are required at differentcoolant temperatures (to provide fine tuning of themixture). Figure 3.26 shows a typical coolanttemperature sensor.

The coolant temperature sensor operates in exactlythe same way as the air temperature sensor describedabove. The sensor resistance changes with coolanttemperature, resulting in a voltage change in the circuit,which the ECU can then use as an indication of coolanttemperature.

The sensor has a negative temperature coefficient(NTC), so its resistance reduces as the temperature

increases. A typical resistance for the sensor is around7000 ohms (7 kΩ) at 0°C, falling to around 250 ohmsat 100°C.

The voltage in the sensor circuit also reduces as thetemperature increases. A reference voltage is applied tothe sensor circuit, which on the Bosch LE2 system isaround 12 volts. However, when the sensor isconnected to the circuit, the resistances in the circuitreduce the voltage to a value that depends on theresistance of the sensor, which changes withtemperature. For normal operation, the voltage in thecircuit is around 9 volts for a very cold engine andaround 5 volts for a hot engine.

Throttle position switchAs previously described, the throttle switch consists oftwo sets of contacts. One set closes when the throttle isclosed (the idle position). The second set of contactscloses when the throttle is approximately 60% open(this value will depend on the application).

With the Bosch LE2 system, 12 volts is applied tothe centre terminal of the switch. When a set of contactscloses, the 12 volt signal is transmitted back to the ECU.The ECU uses this signal as an indication of idleposition or load position. The air:fuel ratio is usuallyenriched slightly to stabilise the idle speed and toenable the engine to produce full power. Figure 3.27shows a throttle switch and its construction.

Timing/trigger referenceThe Bosch LE2 system relies on a signal from theignition system for information on engine speed (toassist in fuelling calculations) and as a reference fortriggering the injectors. A signal is taken from theignition coil or direct from the ignition module. Ineffect, every time the ignition module switches off theignition coil (spark timing) a signal is transmitted to theLE2 ECU.


Figure 3.26 Coolant temperature sensor Figure 3.27 Throttle position switch

[647] Chapter 03 25/8/06 15:38 Page 99

The injectors on an LE2 system operate simultaneously:they all open and close together on a four-cylinderengine. The ECU uses every alternate ignition pulse (ona four-cylinder engine) as a reference to open theinjectors, so the injectors open twice for every enginecycle.

Injectors (actuator)Simultaneous injectionThe only true actuators on an LE2 system are theinjectors, which are exactly as described in section3.1.4. On a four-cylinder engine, all four injectors areopened and closed at the same time (simultaneousinjection). In fact, all four injectors are connected backto the ECU at one terminal, so the power stage withinthe ECU switches all four injectors together (more thanone power transistor might however be used). Withengines of more than four cylinders, the injectors mightbe triggered and connected in groups of three or four.

LE2 injectors operate with a fuel pressure of around2.5 bar above the intake manifold pressure (see section3.1.5).

Figure 3.28 shows the control signal provided by theECU. Note that the EMF created when the injectors areswitched off causes a voltage spike at around 60 volts.

Idle speed control/adjustmentThe standard LE2 system did not provide an automatedidle speed control system. The idle speed was adjustedmanually using a bypass port adjuster on the throttlebody (Figure 3.29).

Manual idle speed adjustmentA bypass port was usually formed as part of the throttlebody (Figure 3.29). The bypass port allows intakeairflow to bypass the closed throttle butterfly (throttlevalve). The bypass port has an adjusting screw that canbe altered to allow more or less air to bypass thethrottle butterfly.

Therefore, if the adjusting screw is unscrewed, more airis allowed to enter the engine, which will increase theidle speed. Screwing in the adjuster will restrict the air,thus reducing the idle speed. This enables the idle speedto be set to the manufacturer’s specifications.

On the LE2 system, when either more or less air isallowed to flow into the engine through the bypass port,the air will still have to flow through the airflow sensor.An increase in airflow will therefore cause the airflowsensing flap to move, altering the sensor signal to theECU; the ECU will therefore increase the fuel quantityto correspond with the increase in airflow, thusmaintaining the air:fuel ratio.

Auxiliary air valve (cold running)When an engine is cold, all moving components havehigher levels of friction, and the cold oil can also causeadditional drag or resistance in the engine. To preventthis additional friction and resistance from stalling theengine when it is cold, the LE system provides additionalair to the engine, which results in a slight increase inengine speed at idle (with the throttle closed).

An additional or auxiliary air valve is used which isconnected by air pipes to the throttle body. As with themanual idle speed adjuster, the air valve is effectively abypass port, as shown in Figure 3.29. However, insteadof a manual adjusting screw, the auxiliary air valve hasa temperature sensitive plate valve which is open whencold and closes when hot (Figure 3.30).

The valve assembly is exposed to two heat sources.The first heat source is an electrical heating elementintegrated into the air valve body. When the engine iscold the valve plate is in the open position, so when theengine is started, additional air flows through to theengine, thus providing a fast idle speed. However, whenthe engine is running, the full battery voltage is appliedto the heating element, which heats up a bimetallic stripattached to the valve plate. As the bimetallic strip heatsup, it bends, which causes the valve plate toprogressively close the air bypass port.

100 Engine management – petrol Fundamentals of Motor Vehicle Technology: Book 2V

olts

Milliseconds

60

50

40

30

20

10

0

–10

–20–15 –5

injector open

injector closed

back ‘EMF’ spike

0 5 10 15–10

Figure 3.28 Injector control signal

Manual idle airbypass adjustment

Auxiliaryair valve

Figure 3.29 Manual and auxiliary air valve bypass ports tocontrol idle speed

[647] Chapter 03 25/8/06 15:38 Page 100

When the engine is switched off, in theory, thebimetallic strip will cool down, allowing the valve plateto reopen the port. However, the air valve is positionedso that it is exposed to engine heat, so the valve bodystays hot until the engine cools down. So the port willnot open again until the engine is quite cool.

It takes typically around 3 minutes (depending onthe vehicle application) for the auxiliary air valve tomove from the fully open to the fully closed position. Inthis time, the engine should have reached an operatingtemperature that allows it to idle at normal speed.

Electrical systems and wiring (LE2)Although the Bosch LE system is now an old one, itsbasic elements are still relevant to today’s injectionsystems. Therefore, in addition to the wiring circuitshown in this section (Figure 3.31), the operation of thecircuit and some of the functions are also explained.

Power supplyEarly electronic injection systems generally used batteryvoltage for all aspects of system operation. On the

Bosch LE system, the injectors used the 12 volt supplyand the sensors were also provided with a 12 voltreference voltage. Although all systems largely still usethe full battery voltage for actuators such as theinjectors, it is now normal practice to use a 5 voltreference voltage for sensors.

A system relay provides the system components withthe 12 volt supply. The relay acts as a safety device andwill switch off the power supplied to the componentsunless certain signals are received from the engine, etc.

The relay consists of contacts which, when closed,connect the battery voltage direct to the systemcomponents. Energising windings within the relay willcause the contacts to close when a voltage is applied tothe windings.

Relay operationThe relay receives the full battery voltage supply directfrom the battery (possibly fused on some applications)to terminal 30.

● When the ignition is initially switched on, thebattery voltage will be applied to terminal 15 of therelay (the voltage will be applied to the energisingwinding). A timer circuit within the relay will causethe relay to apply the battery voltage from relayterminal 87b to the fuel pump for a few seconds(allowing the pump to operate, thus ensuring thatthe fuel system is under pressure). If the engine isnot cranked or started, the relay will switch off thesupply to the pump.

● When the ignition switch is then placed in thecranking position, the voltage will be applied fromthe starter circuit to terminal 50 of the relay; thiswill again cause the energising winding to close thecontacts and the battery voltage will now be appliedto all of the system components (the fuel pump,injectors and sensors). The engine should now start.

● When the engine starts, an ignition speed signal(from the ignition coil or module) is transmitted tothe relay at terminal 1, which indicates that theengine is running. Because the start signal from thestarter circuit will now switch off (the engine is nolonger cranking), the speed signal acts as areplacement so that the relay will continue toprovide battery voltage to the injection system.

● If for any reason the engine were to stop, theignition signal would disappear and the relay wouldswitch off the power supply to the injection system.

InjectorsAll injectors will receive battery voltage from relayterminal 87 during starting and engine running. Thesecond terminal at each injector is then connected toECU terminal 12, which is the earth path for theinjectors. The circuit passes from terminal 12 throughthe power stage of the ECU to earth. Therefore, whenthe ECU switches on the injectors, the power stage willcomplete the earth circuit for the injectors.


Figure 3.30 Auxiliary air valve located in the intake system

[647] Chapter 03 25/8/06 15:38 Page 101

Fuel pumpThe fuel pump receives a power supply from relayterminal 87b while the engine is starting and running.

Throttle switchThe throttle switch receives battery voltage at terminal18 from relay terminal 87. When the idle contacts or theload contacts in the switch are closed, the batteryvoltage will then be applied from the switch terminals 2or 3 to the ECU at terminals 2 or 3. The ECU will thenhave an indication of idle or engine load.

Coolant temperature sensorThe coolant temperature sensor also has a batteryvoltage supply from relay terminal 87. As previouslynoted, the sensor is part of a series resistance circuit, so,as the resistance of the sensor varies with temperature,the signal from the sensor to ECU terminal 10 willchange, indicating the temperature to the ECU.

Airflow sensorThe airflow sensor is supplied with the battery voltageat terminal 9 through relay terminal 87. This voltage isapplied across the air temperature sensor, whichoperates in the same way as the coolant sensordescribed above. The signal from the air temperaturesensor passes from terminal 8 to the ECU terminal. Thesupply voltage is also applied across the potentiometerwithin the airflow sensor; when the wiper on thepotentiometer moves (due to the airflow sensing flapmoving), the voltage on the wiper contact will change,and this changing signal is transmitted from terminal 7of the airflow sensor to terminal 7 of the ECU. Airflowsensor terminal 5 is the earth connection for thepotentiometer.

Auxiliary air valveThe valve is supplied with the battery voltage from relayterminal 87 while the engine is starting and running;this will cause the heating element in the valve


Figure 3.31 Wiring diagram for Bosch LE2 injection system

[647] Chapter 03 25/8/06 15:38 Page 102

assembly to become hot and remain hot while theengine is running, which causes the air valve to close(so there is no fast idle).

ECUThe ECU is also supplied with the battery voltage fromrelay terminal 87. A speed signal is provided at terminal1 (to enable the ECU to calculate the fuellingrequirements). The cranking or start signal is suppliedto terminal 4; the ECU can then provide additionalinjection pulses or lengthen the duration of theinjection control signal (both of these will allowadditional fuel to be injected for starting).

Signals from the temperature and airflow sensorsare transmitted to terminals 7, 8 and 10, with thethrottle switch signals passing to terminals 2 and 3. Theinjector control signal is provided at terminal 12, withterminals 5 and 13 being earth connections for the ECU.

3.2.2 Example 2: Multi-point systemwith added functionalityNote: This section should be studied in conjunctionwith section 3.2.1. Note also that the fuel deliverysystem on the Bosch M1.5 is identical to the examplecovered in section 3.1.5 and illustrated in Figure 3.20. Itis therefore not described again in this section.

System components and layoutThe system featured in this second example is againmade by Bosch, but is a later system than the previouslycovered LE2 system. The system is referred to as M1.5(Figure 3.32), and features a number of improvementsand changes, as well as added functionality andcapability. In Bosch terminology, the ‘M’ tends to refer to‘Motronic’, the Bosch term that is generally applied toan engine management system. The M1.5 systemcombines the ignition and fuel injection functions aswell as some other functions, which include control ofthe idle speed via an ECU controlled air valve.

Although this section does not deal specifically withengine management, the M1.5 system provides aninsight into later fuel system developments as well asinto early engine management systems. Not all of thefunctions and components of the M1.5 system are dealtwith in this section: some are covered in greater detailin the emissions section and in the engine managementsection.

Sensors and sensor reference voltageMany of the sensors used on the M1.5 system aredevelopments of, or the same as, those used on the LE2system (section 3.2.1). There are however someadditional sensors.

One major change to the system is that the referencevoltages used for the sensors are generally stabilised at5 volts (as opposed to battery voltage). This is because


Figure 3.32 Bosch Motronic M1.5 system

[647] Chapter 03 25/8/06 15:38 Page 103

the M1.5 system uses digital electronics to a muchgreater extent than previous systems and a 5 volt circuitis more suitable for use with electronic components.

Standard multipoint gasoline injection systems usesolenoid type injectors mounted in the inlet portspraying directly at the back of the inlet valve

The quantity of fuel delivered is a function ofinjector opening duration as the pressuredifferential across the injector is kept constant.This is achieved via the fuel pressure regulatorwhich takes into account manifold pressure

Key

Poin

ts

Airflow sensor with combined air temperaturesensorThe airflow sensor and the air temperature sensoroperate in much the same way as the sensor on the LE2system (section 3.2.1). However, one major change is inthe idle mixture adjustment or CO (carbon monoxide)adjustment. Although the task remains the same, theadjuster on the M1.5 airflow sensor is a potentiometerinstead of an air bypass adjuster. Since the referencevoltage to the sensor is 5 volts, the output signalvoltages during normal operation will typically bebetween 0.25 and 4.75 volts.


Figure 3.33 Wiring diagram for Bosch M1.5 injection system

[647] Chapter 03 25/8/06 15:38 Page 104

The CO adjusting screw is connected to a smallpotentiometer, so when the screw is adjusted it altersthe voltage at the potentiometer wipe connection. Thevoltage across the potentiometer is applied to the ECU.When the adjustment is made, the voltage changes andthe ECU alters the injected fuel quantity, which in turnalters the mixture/CO setting.

In some vehicle applications, the system uses anoxygen sensor (lambda sensor) to control the mixture,so the CO adjuster is not used. Oxygen/lambda sensorsare covered in the section 3.5.

Coolant temperature sensorThe coolant temperature sensor is a negativetemperature coefficient (NTC) sensor, which operates inexactly the same way as the version used on the LE2system. Note that the reference voltage to the sensor is5 volts.

Figure 3.34 shows the typical resistance values forthe sensor and typical voltages in the sensor circuit atdifferent coolant temperatures. Although the valuesquoted are typical for many systems, some systems mayhave sensor resistances and voltages that differ; alwaysrefer to the appropriate specifications when testing.


Figure 3.34 Temperature against resistance and voltage for acoolant temperature sensor

Temperature, °C Resistance (ohms) Signal voltage

0 6,000–8,000 3.7–4.020 2,000–3,500 3.0–3.240 1,000–1,500 2.0–2.260 500–700 1.2–1.580 275–350 0.6–0.9

100 150–250 0.4–0.5

Figure 3.35 Throttle position potentiometer: Bosch MotronicM1.5 system

See sections 1.5.2 and 2.2.3 for additional informationon variable reluctance sensors used to indicatecrankshaft speed and position.

On the M1.5 system there is a trigger or reluctor discon the crankshaft (different positions on the crankshaftare used for different engine applications). The disc has60 reference points or trigger teeth, although one toothis missing, which functions as the master reference.Figure 3.36 shows the sensor and reluctor disc.

The sensor is constructed with a permanentmagnet and a winding. It is located next to thereluctor disc and as each tooth passes the sensor, itinduces a small current into the winding. So, when thecrankshaft is rotating, the sensor will produce anelectrical pulse or signal as each tooth passes thesensor. The missing tooth will create a slightly

Figure 3.36 Crankshaft speed/position sensor: Bosch MotronicM1.5 system

Throttle position sensorOn the M1.5 system, the throttle position sensor (Figure3.35) is a potentiometer instead of a switch. A 5 voltreference is applied to the potentiometer. When thethrottle is opened and closed, the potentiometer wiper(which is connected to the throttle butterfly shaft),moves across the resistance track, thus providing achange in voltage corresponding to the position of thethrottle. The output signal is transmitted to the ECU toenable it to assess the angle of throttle opening.

When the throttle is closed (at idle) the voltage fromthe sensor potentiometer should be at the specifiedvalue, which is typically between 0.3 and 0.9 volts.When the throttle is opened, the voltage rises and, atfull throttle, the voltage will be in the region of 4 to4.5 volts.

Trigger/timing reference (engine speed sensor)The M1.5 system is a combined injection and ignitionsystem, and a single sensor is used to provide the ECUwith crankshaft speed and angular positioninformation. The crankshaft speed/position sensor is aninductive or variable reluctance sensor.

[647] Chapter 03 25/8/06 15:38 Page 105

different pulse shape, which the ECU will use as themaster reference. Figure 3.37 shows part of the ACanalogue signal that would be seen when the sensor isconnected to an oscilloscope.

The ECU uses the master reference signal toestablish a master position for the crankshaft, e.g. TDCfor cylinders 1 and 4. This can be used as a triggerreference for operating the injectors and as a masterreference for the ignition timing. The M1.5 system is asimultaneous injection system, i.e. all injectors openand close together. Also note that the ignition systemhas a single coil for all cylinders and a rotorarm/distributor cap (connected to the end of thecamshaft) to distribute the HT voltage to theappropriate spark plugs.

The additional reference points on the reluctor discprovide the ECU with angular rotation information forthe crankshaft: the ECU can determine crankshaftspeed as each tooth passes the sensor (each referencetooth represents six degrees of crankshaft rotation).

Injectors (actuator)The injectors operate in exactly the same way as thoseon the LE2 system (section 3.2.7). However, on theM1.5 system, the injectors are connected in groups oftwo on four-cylinder engines, but are still all openedand closed at the same time.

Idle speed control valve (actuator)(Also see section 3.1.6). The air valve used on the M1.5system is referred to as a ‘rotary idle valve’. The valve isoperated by a type of electric motor that has itsrotation limited by mechanical stops; the motor istherefore able to rotate only partially. Connected to themotor is a flap or valve that is placed in a bypass portthrough which air flows around (bypasses) the throttlebutterfly to the intake system.


Figure 3.38 ECU controlled air valve controlling the airflowthrough a bypass port

10V

8

6

4

2

0

–2

–4

–6

–8

–10–90 –80 –70 –60 –50 –40

Crankshaft sensor engine idling

–30 –20 –10 0 10ms

Figure 3.37 Signal produced by crankshaft speed/positionsensor: Bosch Motronic M1.5 system

[647] Chapter 03 25/8/06 15:38 Page 106


Figure 3.39 Control signals for rotary idle valve

15.0V

13.5

12.0

10.5

9.0

7.5

6.0

4.5

3.0

1.5

00 10 20 30 40 50

ch A: Frequency(Hz) 99.78Idle speed control valve (rotary)

60 70 80 90 100ms

The electric motor is spring loaded in one direction ofrotation, and current flowing through the motorwindings will tend to rotate the motor in the oppositedirection. By varying the current, it is possible torotate the motor against the spring to achievedifferent angles of rotation or positioning; this allowsvarying volumes of air to flow through the port(Figure 3.38).

Control signal The ECU provides the earth path for the idle valvecircuit. However, the earth path passes through apower stage in the ECU, which rapidly switches on andoff the earth circuit. The result is that a digital controlsignal is produced with an on/off frequency of around100 Hz (100 times a second). The ECU alters the dutycycle (on/off ratio) of the control signal, which altersthe average current in the circuit; this in turn altersthe position of the motor and valve (see section 1.8 forinformation about altering duty cycles to controlactuators). Figure 3.39 shows the typical controlsignal.

Maintaining and increasing idle speedThe idle control system can control the idle speed intwo ways. First, when the engine is at normaloperating temperature, if certain loads are applied tothe engine, such as an electrical load (headlights,heated rear window, etc.), the additional load wouldnormally cause the idle speed to reduce. The ECU,which is receiving the speed signal from the crankshaftsensor, will immediately detect a minor drop in enginespeed, and will change the control signal so that thevalve opens slightly, thus restoring the idle speed to thespecified value. This process is effectively continuousand ensures that any minor change in engine idlespeed is corrected.

The second process for controlling the idle speed relieson information from other sensors. For example, whenthe engine is cold, the ECU assesses the enginetemperature from the coolant temperature sensorinformation and opens the idle air valve slightly toincrease the engine speed and overcome the additionalfriction and drag that exist in the engine at lowtemperatures.

Other information can also be used by the ECU toalter the idle speed: for example, when the airconditioning system is switched on, the load of the airconditioning compressor would slow the enginedown, but to drive the compressor also requiresconsiderable power that may not be available fromthe engine at the normal idle speed. The ECUtherefore opens the air valve an increased amountwhich increases the idle speed. Note that the airconditioning system is connected to the ECU so, whenthe ECU receives an appropriate signal from the airconditioning system, the ECU can implement a fasteridle.

Ignition coil (actuator)The ignition coil is not part of the fuel system, but thesame ECU controls the fuelling and ignition systems.The ignition module effectively forms part of the ECU,so the ECU can use the same information from thevarious sensors to calculate the ignition timing, andthen switch the ignition module, which in turn switchesthe ignition coil. See section 2.3 for information oncomputer controlled ignition systems.

Electrical systems and wiring (M1.5)Figure 3.33 shows the wiring of the M1.5 system.

Although some functions of the M1.5 system aresimilar to the LE2 system, there are many significantdifferences, as explained below.

[647] Chapter 03 25/8/06 15:38 Page 107

Power supply and relay The power supply to the M1.5 system is split into twocategories: the first category is the supply to theactuators, which operate using full battery voltage; thesecond category is the reference voltage for many of thesensors, which is usually 5 volts.

The actuator supply is therefore via the system relay,but the reference voltage is provided by the ECU, whichhas a voltage stabiliser system to reduce the batteryvoltage down to a stabilised 5 volts for the sensors.

Relay operationThe relay has two sets of contacts: one set switches thepower supply to the fuel pump and Lambda sensorheater (where fitted); the second set switches thepower supply to the system actuators.

Compared with the LE2 system previously covered,the relay on the M1.5 system operates slightlydifferently. Each of the energising windings within therelay is earthed via the ECU; therefore, the ECU controlswhen the energising windings are able to close thecontacts.

● The relay receives full battery voltage direct fromthe battery to relay terminals 30 and 86. When theignition is switched on, the ECU receives batteryvoltage via the ignition switch (to ECU terminal 27),which indicates that the driver intends to start theengine. The ECU then completes the earth path forthe relay energising windings at ECU terminals 3and 36 (connecting to relay terminals 85b and 85).Both of the relay contacts will then close, providinga power supply to the fuel pump (relay terminal87b) and to the rest of the actuators (relay terminal87). The fuel pump will run briefly to ensure thatthere is fuel pressure.

● If the engine is then not started, the ECU will switchoff the earth path to relay terminal 85b, thuscausing the pump contacts to open and switch offthe fuel pump.

● When the engine is cranked over for starting, thecrankshaft position sensor will provide a signal tothe ECU, which will now have an indication that theengine is being started; the ECU will then reconnectthe earth path for the energising winding (at relayterminal 85b), thus causing the fuel pump contactsto close again and provide power to the fuel pump.

● The relay will continue to provide power supplies toall components so long as the ECU is receiving theignition ‘on’ voltage and a signal from the crankshaftposition sensor.

● If the engine were to stop, the signal from thecrankshaft position sensor would disappear and theECU would switch off the earth paths for the relayenergising windings. The relay contacts would thenopen, causing all actuators to switch off.

InjectorsAll injectors will receive battery voltage from relayterminal 87 during starting and engine running. Note

that the injectors are then connected to the ECU atterminals 16 and 17; these are the earth paths for theinjectors. From terminals 17 and 18 the circuit passesthrough the power stages of the ECU to earth.Therefore, when the ECU switches on the injectors, thepower stages will complete the earth circuit for theinjectors. Although there are two groups of injectors, forthis application the injectors are still switched at thesame time.

Idle speed control valve The idle control valve receives power from relayterminal 87. The earth path for the valve is via ECUterminal 4; this is the circuit within the ECU thatconnects to the power stage and therefore provides thecontrol signal.

Ignition coilThe ignition coil receives a power supply from theignition switch, and the coil is switched to earth viaECU terminal 1.

Fuel pumpThe fuel pump receives power from relay terminal 87bwhile the engine is starting and running.

Other actuatorsThere are some other actuators fitted, including aLambda sensor heater and an EVAP canister purgevalve. Although not all applications of M1.5 had thesecomponents, they are emissions control componentsfitted to many vehicles and are therefore covered insection 3.5.

Coolant temperature sensorThe coolant temperature sensor is connected to the ECUat terminals 45 and 26. Terminal 26 is an earth paththat is shared with other components. The referencevoltage (5 volts) is applied to the sensor from terminal45. Because the sensor is part of a series resistancecircuit, the voltage at terminal 45 will then reduce,depending on the temperature, and therefore also theresistance value at the sensor.

Airflow sensorThe airflow sensor receives a 5 volt supply at terminal 3from ECU terminal 12. The voltage is applied across theairflow sensor potentiometer and when the wiper onthe potentiometer moves (as the airflow sensing flapmoves), the voltage on the wiper contact will change.This changing voltage level is transmitted from terminal2 of the airflow sensor to terminal 7 of the ECU.

The supply voltage is also applied to the COadjustment potentiometer (within the airflow sensor).The wiper position on the CO potentiometer resistancetrack depends on the adjuster screw position, andtherefore the voltage at the wiper also depends on theadjuster screw position. However, the voltage at thewiper is applied back to the ECU from airflow sensorterminal 1 to ECU terminal 43. The voltage at theseterminals is used by the ECU to adjust the fuelling atidle speed.


[647] Chapter 03 25/8/06 15:38 Page 108

The air temperature sensor (located within the airflowsensor) operates in the same way as the coolant sensordescribed above. The air temperature sensor has aseparate 5 volt supply (which is the reference voltage) atsensor terminal 5. As with the coolant sensor, when theair temperature sensor is connected in the circuit, theresistance of the sensor alters the voltage in the circuit.Therefore the voltage at airflow sensor terminal 5 or atECU terminal 44 depends on the air temperature.

All sensing elements within the airflow sensorassembly connect through to earth via sensor terminal 4to ECU terminal 26.

Throttle position sensorThe throttle position sensor or potentiometer receives a5 volt supply to the potentiometer resistance at sensorterminal 2 (supplied from ECU terminal 12). The earthfor the potentiometer resistance is via sensor terminal 1to ECU terminal 26. The potentiometer wiperconnection passes to sensor terminal 3 and to the ECUat terminal 53. Therefore, when the throttle is openedand closed, the voltage at sensor terminal 3 and ECUterminal 53 will increase and decrease, thus providingan indication of throttle angle to the ECU.

Crankshaft position sensorThe crankshaft position sensor (or engine speed sensor)is an inductive sensor that produces its own signal. Thetwo main connections from sensor terminals 1 and 2connect to ECU terminals 48 and 49. These twoconnections provide a complete circuit for the sensorwinding. The signal is transmitted via terminal 1 of thesensor to ECU terminal 48; the other connection istherefore the return or earth path.

Note that a third connection to sensor terminal 3connects to ECU terminal 19 and to earth; this circuitforms a screen or shield around the sensor wiring toshield out other electrical interference.

ECUThe following list indicates the function of eachconnection at the ECU terminals. Note that not allterminals are used.

Terminal 1 Switched earth path for the ignition coilTerminal 2 Earth connectionTerminal 3 Switched earth path for fuel pump relay

energising windingTerminal 4 Switched earth path for the idle speed

control valveTerminal 5 Switched earth path for the EVAP canister

purge valve (covered in section 3.5.1)Terminal 6 Connection to automatic transmission

ECUTerminal 7 Airflow sensor signalTerminal 9 Signal from vehicle speed sensorTerminal 10 Earth connectionTerminal 12 5 volt supply to airflow sensor and

throttle position sensorTerminal 13 Connection to diagnostic plug (covered in

section 3.7.3)

Terminal 14 Earth connectionTerminal 16 Switched earth path for a group of

injectorsTerminal 17 Switched earth path for a group of

injectorsTerminal 19 Earth connectionTerminal 20 Earth connection (only connected if the

engine does not have a catalyticconverter, this connection effectively‘programs’ the ECU to control fuellingand ignition applicable to a vehiclewithout a catalytic converter)

Terminal 21 Earth connection (only connected if thevehicle has automatic transmission, thiseffectively ‘programs’ the ECU to performcertain functions differently)

Terminal 22 Connection to dashboard warning light(illuminates the light if there is an enginemanagement system fault)

Terminal 24 Earth connectionTerminal 26 Earth circuit for various sensorsTerminal 27 Ignition on supply from ignition switchTerminal 28 Signal from lambda/oxygen sensor

(covered in section 3.5.7)Terminal 32 Signal to trip computer (to enable the

trip computer to calculate fuelconsumption, etc.)

Terminal 34 Connection to automatic transmissionECU

Terminal 36 Switched earth path for relay energisingwinding (to close main contacts)

Terminal 37 Battery voltage power supply from relaymain contacts

Terminal 40 Connection to air conditioning systemTerminal 41 Connection to air conditioning systemTerminal 43 Signal from CO adjuster (in airflow

sensor)Terminal 44 Air temperature sensor signalTerminal 45 Coolant temperature sensor signalTerminal 46 Connection to octane adjust plug

(connector plugs with different resistancevalues are connected across this circuit;this indicates to the ECU the octane gradeof fuel being used; the ignition timingand fuelling may alter depending onwhich octane plug is used)

Terminal 47 Earth connection (used on specificapplications if the vehicle has four-wheeldrive)

Terminal 48 Connection to crankshaft speed/positionsensor

Terminal 49 Connection to crankshaft speed/positionsensor

Terminal 51 Connection to automatic transmissionsystem

Terminal 53 Signal from throttle position sensorTerminal 55 Connection to diagnostic plug (covered in

section 3.7.3).


[647] Chapter 03 25/8/06 15:38 Page 109


3.2.3 Example 3: Multi-point systemswith hotwire air mass sensorsThis section should be studied in conjunction withExample 2 (section 3.2.2).

System componentsThe major difference between the system detailed inExample 2 and the examples in this section is the use ofa hotwire air mass sensor instead of a vane or flapairflow sensor. The rest of the system is similar to, or thesame as, Example 2 and is therefore not repeated.

Injection system wiring differs betweenmanufacturers and applications in each vehicle, soadditional wiring diagrams are not included in thissection, apart from the wiring for hotwire sensors.However, the wiring of many systems will be similar tothat for the Bosch Motronic M1.5 system shown inFigure 3.33, with the obvious exception that terminalnumbers at the ECU, sensors, actuators and relays, etc.will be different. Most injection systems are designed toperform the same basic tasks, so the components andwiring requirements are generally similar.

Hotwire air mass sensor (see also section 1.5.5)Hotwire air mass sensors perform a similar task to vaneor flap mechanical airflow sensors, except that air masssensors measure the mass of air as opposed to thevolume of air. Additionally, air mass sensors do not usemechanical means of measurement, but rely totally onelectronic measurement of the air mass. The mass of airmeasured changes with density and temperature (bothof which change with altitude). The mass of a givenvolume of air therefore varies, and the volume of fuel

provided should be dependent on the mass of air ratherthan its volume; this means that mechanical airflowsensors (which measure the volume of air) do notprovide sufficient information to the ECU. Any changein altitude, or any other factor that affects the airdensity, is not accounted for by mechanical airflowsensors.

Sensor operationHotwire air mass sensors use the cooling effect of the airflowing through the intake system; the greater the massor density of that air, the greater the cooling effect.Cooling changes the resistance of a heated wireelement; this resistance change is used to produce theoutput signal from the sensor. A temperature sensingelement is also used.

Hotwire air mass sensors generally provide ananalogue signal, where the voltage rises and falls withthe change in air mass (increase and decrease of airdrawn into the engine). Some air mass sensors providea digital signal.

Some designs of hotwire air mass sensor use anadditional heating process to burn off contaminationthat could build up on the sensing wire; this ‘burn off ’function is implemented after the engine is switched offand lasts for a short period of around one second.

Figure 3.40 shows the typical appearance of ahotwire air mass sensor and the sensing element, aswell as a wiring diagram for a sensor.

Typical wiring connections for this sensor would be:

● a battery voltage supply connection via the injectionrelay

● a battery voltage supply connection from the ECUwhen burn off sequence is required

Figure 3.40 Example of hotwire mass airflow sensora Hotwire air mass meter b Wiring for hotwire air mass sensor

[647] Chapter 03 25/8/06 15:38 Page 110

● an output voltage terminal giving mass air flow asan analogue voltage

● earth connection for burn off● sensor earth connection● spare terminal (used for CO adjustment when

available).

Hot film air mass sensorsA development of the hotwire sensor is the ‘hot film’sensor. Instead of using a heated wire, the hot film typehas a thin metallic film sensing element. The operationand general appearance are similar to the hotwiresystem, although the construction eliminates the needfor the ‘burn off ’ process.

3.2.4 Example 4: Multi-point systemswith map sensorsThis section should be studied in conjunction withExample 2 (section 3.2.2).

System components and layoutThe major difference between the system detailed inExample 2 and the examples in this section is the use ofa MAP sensor instead of an airflow sensor. The rest ofthe system is similar to, or the same as, Example 2 andis therefore not repeated.

Map sensorSee also section 1.5.4.

Principle of operationMAP sensors (Figure 3.41) are used as an alternative toairflow sensors to enable the ECU to calculate the massof air entering the engine. The MAP sensor informationis used in conjunction with the engine speed informationto enable the ECU to make the appropriate calculations.MAP sensor systems will therefore usually have acrankshaft speed/position sensor with a large number of


Figure 3.41 Examples of MAP sensorsa Remotely mounted MAP sensorb MAP sensor located on intake manifold

reference teeth on the reluctor disc, to provide veryaccurate engine speed information to the ECU.

The MAP sensor measures the pressure (ordepression) in the intake manifold; the pressuredepends on engine load and throttle position, as well asengine speed. However, intake pressure can be at acertain value for many combinations of engine speedand throttle position. For example, the pressure ordepression at idle speed with a closed throttle can besimilar to the depression when the engine is operatingat light load with a partially open throttle. Therefore theengine speed and throttle position information areprovided separately to the ECU, so that differentoperating conditions can be allowed for.

Modern MAP sensors use a pressure sensitivecomponent such as a piezo crystal that changes itselectrical resistance with presssure. The pressuresensitive components form part of an electronic circuitwhich then produces an electrical signal that will varywith any variation in pressure.

The MAP sensors are connected to the intakemanifold via a pipe, although, for many applications,the sensor is connected directly to the intake manifoldor plenum chamber. When the intake system pressurechanges, the signal produced by the MAP sensor alsochanges. MAP sensors can provide digital or analoguesignals.

Analogue signal MAP sensorAny analogue signal MAP sensor provides a voltage thatrises and falls with changes in intake pressure. ModernMAP sensors are provided with a 5 volt supply orreference voltage, and the output from the sensortherefore generally varies between approximately 0.25to 4.75 volts (depending on the intake pressure). Referto section 1.6.3.

An analogue MAP sensor will typically have a lowoutput at low pressure (i.e. high manifold vacuum), withincreasing voltage output as manifold pressure increases.

[647] Chapter 03 25/8/06 15:38 Page 111

terminal 2 provides the analogue signal. Note thatdifferent sensors will have different terminal numbers.

Digital signal MAP sensorsA digital MAP sensor operates in a similar way to ananalogue sensor, but it produces a digital signal insteadof an analogue signal.

The sensor produces a simple on/off signal.However, when the intake pressure changes, the signalfrequency changes, i.e. the number of pulses producedin a second will increase or decrease. As an example, atidle speed where the manifold pressure is low (highdepression), the frequency could be around 90 Hz, butwhen the throttle is opened and the intake pressurerises, the frequency could rise to approximately 150 Hz.

The sensor power supply (typically 5 volts) is astabilised voltage supplied by the ECU and connectedvia a terminal on the sensor. Additionally an earth (0volt) connection terminal as well as a signal outputterminal will be available.

Air mass or volume flow meters measure the airconsumption of the engine directly as they aremounted in the inlet tract

Manifold pressure sensors are used in conjunctionwith the throttle position signal and engine speedto calculate engine air consumption indirectly

Key

Poin

ts

3.3.1 Simplified injection systemCompromise between a carburettor and multi-point injectionSingle-point injection systems have a cost advantageover multi-point injections systems, but they have manyof the features of multi-point injection systems thatallow them to provide much better fuel delivery andmixture control than carburettors. However, there areseveral limitations with single-point systems, includingcertain limitations on emissions control and in the typesof engine that can efficiently operate with single-pointinjection. These limitations are explained below.

A single-point injection system is in many wayssimilar to a carburettor, because the fuel enters theengine intake system from a single point in the throttlebody (Figure 3.43). However, whereas a carburettorrelies on the creation of a lower pressure area within theventuri to draw in fuel from a reservoir, single-pointinjection makes use of a single injector that injects fuel(under pressure) directly above the throttle butterfly(throttle valve or plate). Although the fuel pressure for


Figure 3.42 Analogue MAP sensor

3.3 SINGLE-POINT (THROTTLE BODY) PETROL INJECTION

a single-point system is not as high as for multi-pointsystems, the pressure is higher than that in the intakesystem. Typical injection pressures for single-pointsystems are around 1 bar or slightly less.

With single-point injection, all cylinders receive fuelfrom the single injector. However, because the injectoris controlled by an ECU in the same way as on a multi-point injection system, it is possible to use sensors toprovide information to the ECU; this therefore providesbetter control of fuel quantity than a carburettor, butwith reduced cost compared with a multi-pointinjection system.

DisadvantagesThe disadvantages of a single-point system are in factnot dissimilar to those of a carburettor; for example,fuel/air separation when the air and fuel mixture flowsaround corners in the intake system. Additionally, fuelcan still condense against the cold manifold wallsduring cold running.

Single-point injection was quite widely used onfour-cylinder engines but these systems were not

In the example shown in Figure 3.42, the 5 volt supplyis provided by the ECU to terminal 1 of the sensor.Terminal 3 is the earth connection (via the ECU) and

[647] Chapter 03 25/8/06 15:38 Page 112

suitable on longer engines, such as straight six-cylinderengines, because the different intake manifold lengthsresult in uneven distribution of fuel. This is the sameproblem that affected many carburettor engines wherethe length of the intake pipe from the carburettor orsingle injector to outer cylinders was much greater thanto the central cylinders; this resulted in the outercylinders running more weakly than the inner cylinders.A rich mixture was therefore provided to ensure that allcylinders developed reasonable power and could runreasonably efficiently. However, the central cylindersthen operated with a slightly rich mixture, which causeshigh emissions. Single-point injection is thereforesuitable for vehicles with smaller engines, althoughsome V8 engines were fitted with a single-point system;this was possible because the location of the injectorwithin the centre of the V resulted in similar intake pipelengths to all cylinders.

One other major disadvantage relates to emissionscontrol and emissions control regulations. It is nownecessary on modern systems to stop delivery of fuelto a cylinder if that cylinder is operating veryinefficiently. If the spark at the plug were veryinefficient or failed completely, unburned fuel wouldflow through the cylinder and into the atmosphere aspollution. Modern multi-point injection systems, can

Single-point (throttle body) petrol injection 113

Figure 3.43 Single-point injection system

detect which cylinder is operating inefficiently andswitch off the injector to that cylinder. This is notpossible on single-point systems where the injectorsupplies fuel to all cylinders.

3.3.2 Operation of a single-pointinjection systemInjector (actuator)The injector (Figure 3.44) operates in much the sameway as an injector for a multi-point system. Theinjector is a solenoid that, when energised, causes theneedle to lift off the seat (the typical needle lift isapproximately 0.06 mm). A control signal from theECU opens and closes the injector for a calculatedperiod of time (typically 1.25 ms to 8 ms, dependingon operating conditions). It is usual for the injector tobe opened at every ignition spark; i.e. on a four-cylinder engine, the injector would be opened eachtime a spark occurred, which equates to four times forevery engine cycle.

Idle speed control (actuator)As with multi-point injection systems, some form ofautomated idle speed control is provided. A common

[647] Chapter 03 25/8/06 15:38 Page 113

method is to use a stepper motor (see section 3.1.6),which acts on the throttle butterfly via some form oflinkage. The ECU controls the stepper motor to eithermaintain or increase the idle speed for cold running orwhen load is applied to the engine at idle.

Alternatively, the stepper motor can control a valve,which alters the aperture in a bypass port. The portallows air to bypass the throttle butterfly; therefore,when the valve allows more air to flow through theport, the idle speed increases. Bypass port systems arecovered in section 3.1.6.

SensorsThe main information required for a single-pointinjection system to calculate the required fuel quantityis engine speed and throttle position (throttle openingangle). These two signals provide sufficient informationfor the ECU to calculate the required quantity of fuel tosuit the engine load. In effect, the ECU has an indicationof ‘air charge’ per cylinder from the engine speed andthrottle opening signals. Some systems have a MAPsensor to provide additional information relating toengine load.

Ignition trigger or speed signalOn earlier systems, a speed signal was received directfrom the ignition system (ignition coil or ignitionmodule). Later when injection and ignition werecombined, a signal was provided by a crankshaftspeed/position sensor.

Throttle positionA throttle position sensor (usually a potentiometer)provides information relating to throttle angle openingand the rate at which the throttle is being opened or


closed. The throttle position sensor operates in thesame way as those previously described for multi-pointinjection systems (sections 3.1 and 3.2). It was,however, common practice to use a set of contacts inthe throttle sensor to indicate the closed or idle throttleposition.

Air temperatureAn air temperature sensor is located in the throttle body(Figure 3.43). Because air density changes withtemperature, the information from the sensor assists incalculating the required fuel quantity to match the airdensity. An air temperature sensor operates in anidentical way to air temperature sensors previouslycovered under multi-point systems (sections 3.1 and3.2).

Coolant temperatureThe operation and function of coolant temperaturesensors is the same as for multi-point injection systems(sections 3.1 and 3.2). As with all fuelling systems,enrichment (excess fuel) is needed during cold running,and minor fuelling adjustments can be made for minorchanges in engine temperature: the coolanttemperature sensor provides the relevant information.

Other sensorsFigure 3.43 shows a lambda (oxygen) sensor and othercomponents that are applicable to emissions control.These components are covered in section 3.5.

Fuel systemThe fuel system of a single-point injection system issimilar to that of a multi-point system (Figure 3.43). Afuel pump filter and regulator assembly are used, whichoperate in much the same way as on a multi-pointsystem (see section 3.1.5). However there are twomajor differences between single-point and multi-pointfuel systems. First, single-point systems operate atlower fuel pressures, typically 1 bar.

The second difference is that, because the fuel isinjected ahead of (or upstream) of the throttle butterfly,the fuel is injected into a pressure zone that does notchange significantly with throttle opening. In section3.1.5 it was explained that, because a multi-pointinjector injects fuel into the intake port, the injectionpressure is regulated so that it is always at a constantpressure ‘above the pressure in the intake port’. Thepressure regulator is therefore connected to the intakesystem pressure so that the regulator can ‘sense’ intakesystem pressure.

On a single-point injector, the fuel is injected into arelatively constant pressure zone above the throttlebutterfly (which is at atmospheric pressure) andtherefore the injection pressure does not need to bealtered when the intake pressure changes. The pressureregulator therefore has no connection to the intakepressure.

The fuel supply system and pressure regulator areshown in Figure 3.45.Figure 3.44 Injector for a single-point system

[647] Chapter 03 25/8/06 15:38 Page 114

Manifold heatingIf atomised petrol condenses on cold surfaces, problemscan occur when the air:fuel mixture flows through theintake manifold when the engine is cold. On someapplications, therefore, an electric heater is located atthe base of the intake manifold (Figure 3.46) to helpprevent the petrol from condensing.

The heater is switched on when the ignition isinitially switched on and during starting; the heater canremain switched on for a number of minutes afterstarting. During cold running, therefore, the air:fuelmixture flowing from the throttle body is heated, whichhelps to ensure that the fuel remains atomised.

Direct petrol injection 115

Figure 3.45 Fuel system for a single-point system

Figure 3.46 Manifold heater on a single-point injection system

3.4 DIRECT PETROL INJECTION

This section covers the basic principles of direct petrolinjection (also called gasoline direct injection or GDI).Direct injection systems help to achieve overallcombustion efficiencies by operating in conjunctionwith special combustion chamber designs and withelectronic throttle control. In addition, emissions

control for direct petrol injection systems is slightlydifferent from that for engines with multi-point portinjection. For these reasons, additional informationabout direct petrol injection is provided in theemissions section (section 3.5).

[647] Chapter 03 25/8/06 15:38 Page 115

3.4.1 Benefits of direct injectionDirect injection into the cylinderSection 3.1.1 mentions older generations of petrolinjection systems which used injection pumps todeliver petrol to injectors that were directly injectinginto the cylinders. This type of injection was usedwith considerable success on aircraft engines duringthe 1930s and 1940s, but the requirements forinjection on aircraft engines were slightly differentfrom those for the modern automobile. However,while diesel engines have relied on direct injectionthrough almost the whole life of this type of engine,petrol delivery systems for automobiles were able tobe much less sophisticated (i.e. to use carburettors)until emissions regulations forced better control offuel delivery.

While cost was inevitably a major factor in usingrelatively inexpensive petrol delivery systems, thetechnologies and materials that were available at thetime also restricted the mass production of what wewould now consider to be the ideal fuel system. Modernelectronic control and materials have enabled designersto develop fuel injection systems that can efficientlydeliver fuel direct to the cylinder, rather than to theintake system.

Figure 3.47 compares multi-point port typeinjection, single-point (throttle body) injection anddirect injection systems.

It is claimed that direct injection, when comparedwith an equivalent engine with port injection, providesa decrease in fuel consumption in the region of 15% to20%, while engine power is slightly improved. Thedetails within this section provide an understanding ofhow these benefits are achieved.

One other benefit is that direct injection systemsrequire very rapid vaporisation of the petrol to enableit to mix quickly with the air. This rapid vaporisation isachieved through the use of high fuel pressures and aspecial injector nozzle design. Importantly, when aliquid vaporises, it has the effect of drawing heat fromthe surrounding air, i.e. it cools the surrounding air.Therefore, when fuel is injected into the cylinder, thevaporisation process reduces the temperature of theair in the cylinder, reducing the potential forcombustion knock (which can occur if temperaturesare too high).

This reduced tendency for combustion knockenables higher compression ratios of around 12:1 to beused (which would otherwise raise cylindertemperatures and cause combustion knock). Thuscombustion efficiency is improved, giving more poweras well as improved fuel consumption and emissions. Inaddition, the cooling effect on the air in the cylindercauses the air to become denser; the greater the airdensity or mass within the cylinder, the greater thepower produced.


Figure 3.47 Comparison of different types of petrol injectiona Multi-point fuel injection (MPI)b Throttle body fuel injection (TBI)c Direct fuel injection (DI)

1 Fuel 4 Intake manifold2 Air 5 Injector3 Throttle valve 6 Engine

(b)

(a)

(c)

[647] Chapter 03 25/8/06 15:38 Page 116

Mixture formationUntil very recently, the vast majority of petrol enginesoperated with the air and petrol mixed outside thecylinder (e.g. port type injection); the air and petrolmixture was then drawn into the cylinder during theintake stroke. Although mixing can continue after theair and petrol are inside the cylinder, the initial mixingprocess starts in the intake manifold (carburettors andsingle-point injection) or in the intake ports (portinjection). With direct injection, the air is still drawninto the cylinder in the conventional manner, but thepetrol is injected directly into the cylinder, so mixingoccurs only within the cylinder.

Gasoline direct injection engines inject fuel directlyinto the combustion chamber, in a similar way to adiesel engine

These engines have two distinct operating modesfor combustion: one is similar to a standardgasoline engine where increased power is needed;the other is similar to diesel engine whereeconomy is most important (e.g. part-load)

One main advantage of mixing the air and petrol in thecylinder is that different mixture formation processescan be achieved using different injection timing.Essentially, there are two types of mixture formationused with direct injection systems: ‘homogenous’ and‘stratified’.

Homogenous mixture formationA homogenous mixture is one where the fuel mixes withthe air in such a way that the mix is uniform orunvarying throughout the whole volume of air/petrolmix (Figure 3.48a). This means that the whole volumeof mixture will have the same air:fuel ratio (no weak orrich pockets of mixture). Therefore, when ignitionoccurs, all of the mixture will ignite and burn (combust)with equal efficiency and the flame created by initialcombustion will therefore spread through the wholemixture (flame prorogation).

In general, a homogenous mixture will operate at oraround the stoichiometric air:fuel ratio of 14.7 parts ofair to 1 part of petrol (by weight). This is the theoreticalideal ratio which will also provide low emissions ofmost pollutants. It is possible to operate with weakmixtures of up to 20:1 (or slightly higher) beforemisfiring occurs. These weaker mixtures provide goodeconomy and low emissions of most pollutants. Inpractice, maximum torque and power are usuallyachieved with slightly richer air:fuel ratios of around12:1 but with higher emissions of some pollutants.

Since the early 1990s in Europe (earlier in the USA),emissions regulations have resulted in enginesoperating with air:fuel ratios that are generally close tothe stoichiometric value for most operating conditions.This allowed catalytic converters to convert most of thepollutants into harmless gases (refer to emissions insection 3.5). Operating at stoichiometric air:fuel ratios

Key

Poin

tsDirect petrol injection 117

throughout the mixture effectively means that themixture should be homogenous under all engineoperating conditions.

Stratified mixture formationWith stratified mixture formation, a small isolatedpocket or cloud of air:fuel mixture is created within thecylinder; the remainder of the air is effectively pure(Figure 3.48b). In reality, it is possible to have a pocketof mixture with a stoichiometric air:fuel ratio (whichtherefore burns normally), while the remaining air iseither completely free of any petrol or has a very smallamount of petrol mixed in, i.e. it is very weak.

The small pocket of mixture is directed by theairflow within the combustion chamber so that it isdirectly exposed to the spark plug. When the sparkoccurs, therefore, it is only this pocket or cloud ofmixture that ignites and combusts. The combustion ofthis isolated cloud of mixture is used to heat up all ofthe remaining air, thus producing expansion of the gaswithin the cylinder. If the remaining ‘fresh air’ does infact contain a small quantity of petrol (forming a veryweak mixture), it will combust slowly, which will in factassist in the expansion of the gases.

It should, however, be noted that a stratifiedmixture formation will not produce as much energy orforce within the cylinder as a fully homogenous mix ofair and petrol, because only a small percentage of thefull charge of air in the cylinder is used to generate theheat. With homogenous mixtures, the full charge of airis mixed with petrol, and therefore all of the mix

(a)

(b)

Figure 3.48 Homogenous and stratified mixture formationa Homogenous mixture formationb Stratified mixture formation

[647] Chapter 03 25/8/06 15:38 Page 117

combusts. It is also possible to alter the air:fuel ratio forthe small pocket of mixture so that this pocket alsooperates on a weaker mixture, but the mixture must berich enough to achieve good combustion.

The important point to remember is that, althoughthe small localised pocket of mixture has an air:fuelratio that is rich enough to achieve combustion, theoverall mixture within the cylinder has an excess of airbecause of the large volume of pure air in the rest of thecylinder. It is in fact possible to achieve a total air:fuelratio of up to 40:1 (the total quantity of air comparedwith the total quantity of petrol).

The obvious advantage of stratified mixtureformation is that the amount of fuel required is muchsmaller than for homogenous mixtures and thereforefuel consumption is much lower. However, a stratifiedmixture formation cannot produce the same power as ahomogenous mixture, which means that stratifiedmixes can be used for light engine load operation (idlespeed and light load cruising).

One disadvantage of stratified mixture formation isthat, at higher engine speeds, excessive turbulence iscreated, which does not allow the formation of thecloud or pocket of gas to localise around the spark plugtip. This results in poor combustion. Additionally, if anincrease in power or torque is required, the air:fuel ratioprovided to the pocket of mixture must be richer. Thiscan lead to very small, but very rich, zones of mixture(within the cloud) which can result in soot beingproduced.

Stratified mixture formation is therefore ideal forlight load conditions and lower engine speeds, but,when engine speeds increase above mid-range(typically around 3000 rev/min) or increased enginetorque and power are required, the engine must operatewith a homogenous mixture.

Injection timingMost direct injection petrol engines operate withstratified and homogenous mixture formationsdepending on operating conditions. This is achieved bycontrolling the injection timing. Direct injection systemsgenerally have two distinct timing periods, whichprovide different characteristics for mixing the air andfuel. One timing period is during the induction (intake)stroke; the other is at the end of the compression stroke.

Intake stroke injection timing When petrol is injected during the intake stroke (whilethe air is being drawn into the cylinder), the fuel willmix with all of the air in the cylinder, resulting incomplete mixing or homogenous mixture formation.Note that the intake ports can be designed to createswirl or controlled turbulence of the air entering thecylinder, which assists in mixing the petrol with the air.The mixture is typically at or close to the stoichiometricair:fuel ratio, thus enabling good power to be producedwith reasonably low emission of pollutants. The highfuel injection pressures used and the design of theinjector nozzle create good atomisation of the petrol,


improving the mixing process, which continues duringthe intake and compression strokes. Figure 3.48 showsthe injection of fuel during the intake stroke.

Compression stroke injection timingA relatively small amount of petrol is injected at the endof the compression stroke, just prior to ignition (Figure3.48). The design of the combustion chamber includesan area (usually in the top of the piston crown) whichpromotes swirl or turbulence in a small, localised region.This allows the injected fuel to mix with a small pocketof air, forming a small pocket or cloud of mixed air andpetrol. The small pocket of mixture is then directed tothe spark plug tip, ensuring ignition of the mixture.

To create the small localised pocket of air:fuelmixture requires special piston and combustion chamberdesign. In addition, the location of the spark plug andinjector in the cylinder are critical. One specific designfeatures an additional flap in the intake tract (known asa charge motion valve). This is used in conjunction witha specially shaped piston crown and inlet manifolddesign to provide the required gas behaviour in stratifiedoperation mode. This characteristic behaviour is knownas ‘tumble’. The flap valve is actuated electronically via astepper motor and is controlled by the ECU. The angle ofthis valve reduces the cross sectional area of the inletmanifold, thus increasing gas velocity and tumbleimparted to the incoming air charge during stratifiedoperation. During homogeneous operation this valve isfully open and has no effect.

Using both timing periodsDirect injection systems in petrol engines generally useboth timing periods (intake stroke and compressionstroke timing) independently, depending on theoperating conditions. For light load driving and at idle,compression stroke injection timing means that verylean mixtures can be used (stratified mixtureformation), which provides low power but goodeconomy. When higher engine power is required orwhen the engine is operating at higher speeds, theinjection timing changes to the intake stroke, providinga full charge of mixed air and petrol to the cylinder(homogenous mixture formation).

Because the injector timing is entirely controlled bythe ECU, it is possible to time the injection to any pointin the engine operating cycle. The exact time ofinjection during the intake stroke period and thecompression stroke period can therefore be adjusted tosuit the exact operating conditions, such as speed,temperature, etc.

There are also certain conditions under whichinjection takes place on both the intake and thecompression strokes. A small quantity of fuel is deliveredon the intake stroke, which produces a homogenous butweak mixture. Injection occurs again on the compressionstroke to produce a normal stratified charge (which willhave an air:fuel ratio that is close to stoichiometric).With this dual injection process, the stratified chargeignites and combusts normally which then creates

[647] Chapter 03 25/8/06 15:38 Page 118

combustion in the rest of the air (which has a weakhomogenous mixture). This process produces morepower or torque than when the system is operating withonly a stratified mixture formation. This is used whenthe system is changing from stratified to homogenousoperation to provide a smooth transition.

Throttle control and mixture control to regulatepower and torqueWhile direct injection has a number of advantages thaton their own help to improve engine efficiency andreduce emissions, it is the fact that direct injectionallows ‘stratified mixture formation’ to be used thatprovides the greatest benefit. With direct injection,there are effectively two types of mixture formation andcombustion process that can be used at different times(see injection timing above). However, getting the fullbenefit of both these processes, especially the stratifiedmixture formation, requires additional changes toengine design and engine control.

Filling the cylinder with airIdeally, a cylinder should be fully charged with air at theend of the intake stroke (the largest possible volume ofair), causing higher pressures at the end of thecompression stroke. When combustion occurs, the heatproduced causes the air to expand, but when a highervolume of air is compressed into the small combustionchamber, the expansion will be greater.

Ideally therefore there should be no restrictions thatcould prevent the cylinder from filling with air duringthe intake stroke. Unfortunately, petrol engines havetraditionally had a throttle butterfly to regulate airflowinto the cylinder as a means of controlling enginetorque and power: when the engine is operating at lightloads the throttle is almost fully closed, restricting theairflow into the cylinder. The cylinder is therefore onlypartially filled with air, resulting in low efficiency (lowvolumetric efficiency). Additionally, power is wasted bythe pumping action of the piston on the intake stroke,which is trying to draw air through the restriction.

To avoid this, the throttle should remain as far openas possible to enable improved volumetric efficiencywith subsequent improvements during the combustionand expansion phases. This is in fact achievable withdirect injection by holding the throttle open during light

load conditions and then using an alternative means ofcontrolling torque and power. The throttle is electricallyoperated using a stepper motor or similar device, whichis in turn controlled by the system ECU (see enginemanagement systems, section 3.1.6).

Controlling power by altering the air:fuel ratioAs explained earlier in this section, when an engine hasdirect fuel injection, the stratified mixture formationprocess is used during light load operation. It is possibleto alter the air:fuel ratio of the stratified charge (thesmall pocket of mixture), which will alter the energyproduced during the combustion and expansion phases.So, if a weak stratified charge mixture is used, lessenergy will be produced compared with when themixture is at the ideal air:fuel ratio (or slightly richer).

Therefore if the throttle is held in the open position(by controlling the stepper motor), a full charge of airwill fill the cylinder on each intake stroke, but theenergy produced on the power stroke will be regulatedby the air:fuel ratio in the small pocket of mixture.

Note that when the engine is required to producemore power because the load is increasing (whenaccelerating), it must operate with a homogenousmixture formation. The stratified mixture formationprocess operates with air:fuel ratios that are much tooweak to enable good torque and power to be produced.Air:fuel ratios must be controlled within a fairly tighttolerance for homogenous operation, so it is notpossible to use changes in air:fuel ratio to controltorque and power: the throttle must be used to controlthe torque and power.

Different processes for different operatingconditionsDepending on the driving conditions, the enginetherefore operates with different processes as shown inFigure 3.49.

3.4.2 Operation and componentsEvolution from port injection systemsDirect injection systems have many similarities to theport injection systems described earlier, so many of thecomponents (sensors and actuators) are identical or


Figure 3.49 Using the different operating processes for different operating conditions

Condition Stratified or Intake or compression Power regulation: homogenous stroke injection throttle or air:fuel ratio

Light load and idle Stratified Compression air:fuel ratio(above mid-range engine speeds, the system reverts to homogenous operation)Load (torque and power) Homogenous Intake ThrottleTransition from stratified to homogenous Stratified and homogenous Intake and compression ThrottleCold running (warm-up phase) Homogenous Intake Throttle

[647] Chapter 03 25/8/06 15:38 Page 119

very similar. In effect, direct injection is an evolutionfrom port injection. The main physical differences arethe fuel system pressures, fuel pumps and the locationof the injectors.

While the injectors are inevitably more robust tocope with the harsh environment within the cylinder(high pressure and temperatures), the principle ofoperation is the same as for port type injectors.

The ECU is in principle the same as those used onport type systems, but inevitably, differentprogramming is used to control the slightly differentoperating processes. In this section, therefore, detaileddescriptions will be provided only for thosecomponents, sub-systems and processes that aresignificantly different from those of port injectionsystems.

InjectorsInjectors used in direct injection systems operate inmuch the same way as for those in port injectionsystems: the injector is constructed with a solenoid thatopens the injector valve by moving the needle off a seat,thus allowing fuel to flow through the valve. Theopening time and opening duration of the solenoid arecontrolled by the ECU so that the required quantity offuel is injected at exactly the correct time. Figure 3.50shows an injector used on a direct injection system.

Vaporising the fuelTo provide a good mixture of air and fuel in bothstratified and homogenous mixture formations requiresfinely vaporised petrol. The high pressures (typically upto 5 bar) used by direct injection systems, in

conjunction with the injector nozzle design, cause thepetrol to be delivered from the injector in very finedroplets that can vaporise rapidly before they contactthe cylinder or piston surfaces, which could then causethe fuel to return to a liquid state. The rapidvaporisation provides much quicker mixing with the air.However, an additional benefit is that, whenvaporisation occurs, it has a cooling effect on the air,which lowers the potential for combustion knock.

High voltage for rapid openingThe injectors have a very limited time in which todeliver the fuel to the cylinders. On port injectionsystems, the whole of one engine cycle is available forinjecting fuel, i.e. two crankshaft revolutions (whereeach cylinder can pass through the four strokes and fuelcan be injected at almost any time or all the time ifnecessary). With direct injection, there is only limitedtime on the induction stroke or on the compressionstroke to inject the fuel; the injectors must thereforeopen as quickly as possible to maximise the timeavailable to be used for injecting fuel.

While the ECU produces the control signal in thesame way as on a port injection system, the controlsignal is then transmitted to a driver module which isusually separate from the ECU. The driver modulecontains capacitors that are charged up while theinjector is switched off (closed). When the ‘on’ sectionof the control signal is received by the driver module(indicating the start of injection), the capacitor rapidlydischarges at between 50 and 100 volts (depending onsystem design); this high voltage is discharged throughthe injector circuit. This short high voltage dischargefrom the capacitor causes a very rapid and strong build-up of the magnetic field in the injector solenoidwinding, which in turn causes the injector needle toquickly lift off the nozzle seating. Once the injector isopen, current flow through the injector solenoidwinding is reduced and a ‘hold on’ voltage of around7 volts is used to hold the injector open until it is timeto close the injector.

Injection pressureIn direct injection systems, fuel can be injected at theend of the compression stroke, when cylinder pressurescan reach 20 bar. To obtain the required atomisation ofthe fuel in this high pressure environment and to deliverthe required quantity of fuel quickly, it is necessary touse a high fuel injection pressure. The pressure in thefuel rail (to which all the injectors are connected), istypically around 120 bar (see fuel delivery system laterin this section).

Throttle controlSection 3.4.1 explained that, when the engine isoperating with the stratified mixture formation process,the throttle is held open and engine torque is controlledusing changes in the air:fuel ratio. The throttle musttherefore not be directly connected to the throttle pedal,and is in fact controlled by the ECU, which sends


Figure 3.50 Direct injection high pressure injector

1 Fuel inlet with finestrainer

2 Electricalconnections

3 Spring4 Solenoid5 Injector housing6 Nozzle needle with

solenoid armature7 Valve seat8 Injector outlet

passage

[647] Chapter 03 25/8/06 15:38 Page 120

control signals to a motor (usually a stepper motor),making the throttle open and close as required.

In effect, the driver selects a desired level ofperformance or engine operation in the usual way bymoving the throttle pedal. The throttle pedal isconnected to a potentiometer and, as the throttle pedalis moved, an analogue signal is sent to the ECU. TheECU can control the opening of the throttle dependingon the driver input via the throttle pedal and on otherfactors such as temperature, engine speed, etc.However, when the system is operating using thestratified mixture formation process, the throttle is heldopen and engine power is controlled by changes in theair:fuel ratio.

Figure 3.51 shows an electronically controlledthrottle assembly.

SensorsThe sensors used for direct injection systems aregenerally the same as those used for conventionalmulti-point port injection systems, but an additionalsensor measures pressure in the fuel delivery rail. Adirect injection system forms part of a complete enginemanagement system, which also controls ignition,emissions systems and other engine related systems, soreference should be made to section 3.6, as well asChapter 1, which covers sensors used in modern fuelinjection systems.

Although the sensors used are generally the same asthose previously described for port injection systems,there are slight differences relating to measurement ofthe mass of air entering the cylinders.

Mass airflow measurementIt is important to note that, when the engine isoperating with an open throttle and engine power iscontrolled by the air:fuel ratio, there is effectively littlerestriction in airflow through the intake system, sothere is little reduction in pressure in the intakemanifold and ports. Remember that the low pressure ordepression in a throttled engine is caused by therestriction of the throttle butterfly. Changes in manifolddepression are therefore not as significant as with athrottled engine, but it is more difficult to calculate themass of airflow.

Many direct injection systems use a hot film airflowsensor (see sections 1.5.5 and 3.2.3), which can be usedin conjunction with an intake manifold pressure sensor(section 1.5.4). An air temperature sensor is integratedwith the hot film sensor assembly. The combinedinformation from the sensors enables the ECU to makeappropriate adjustments for the mass of air entering thesystem.

Other systems use two pressure sensors: onemeasures the atmospheric or ambient air pressure; thesecond measures the intake manifold pressure. An airtemperature sensor is also used. The ECU uses theinformation to make the appropriate air masscalculations.

The ECU uses its calculations about the mass of airinduced into the engine to provide the appropriatesignals to control the amount of fuel to be injected, andalso other functions such as ignition timing.

Fuel delivery systemThe fuel delivery system has to provide fuel at muchhigher pressures than on normal, port injection systems,to enable fuel to be injected into the cylinder whencylinder pressures are high (on the compression stroke).Additionally, higher injection pressures help to createbetter fuel atomisation and vaporisation.

Low pressure pumping system A conventional low pressure electric pump (the same asthat on a port injection system) is used to move thepetrol from the tank to a high pressure pump. The lowpressure system operates at around 3 bar to 5 bar,depending on the system design. A high pressure pumpis driven by the engine (usually from the camshaft) anddelivers fuel to the fuel rail at pressures up to 120 bar.

The low pressure system has a pressure regulator,which is usually located in the fuel tank with the fuelpump. When the fuel pressure exceeds the requiredvalue, excess fuel is released from the regulator and isallowed to flow back into the tank. A normal fuel filteris also usually located in the tank. Figure 3.52 shows thebasic layout of a low pressure system which feeds fuel tothe high pressure pump.


Figure 3.51 Electronically controlled throttleFigure 3.52 Low and high pressure fuel pumping system fordirect petrol injection

[647] Chapter 03 25/8/06 15:38 Page 121

High pressure pumping systemThere are two main types of high pressure pumpingsystem, shown in Figures 3.53 and 3.54.

The example in Figure 3.53 is referred to as a‘continuous delivery’ system. The engine driven pumpprovides an excess of fuel to the fuel rail; i.e. it suppliesa greater quantity of fuel than is consumed by theengine. The pump volume will increase with enginespeed, which will result in excessively high fuelpressures. However, a pressure control valve (orregulator valve) is fitted at the end of the fuel rail,which allows excess fuel to flow back to the lowpressure system. The pressure control valve iselectrically operated (usually a solenoid type) andcontrolled by the ECU, which provides a control signalto the valve.

The ECU receives information from a pressuresensor, which is also located on the fuel rail. In responseto the signal from the pressure sensor, the ECU controlsthe regulator valve, adjusting the pressure to therequired value. The control valve also acts as amechanical safety valve in case fuel pressure exceedssafe limits. The excess fuel flowing from the controlvalve then flows back to the low pressure side of thesystem.

The example shown in Figure 3.54 is referred to as a‘demand controlled’ system. The fuel pump contains afuel quantity control valve which regulates the flow offuel from the pumping element. In effect, the controlvalve performs a similar function to the pressure controlvalve described in the previous section, except that, inthis example, when the quantity control valve opens itallows excess fuel to flow directly from the pump backto the return line. In this system, it is therefore possible

to control the quantity of fuel delivered by the pump tomatch engine requirements.

The ECU uses the pressure sensor signal to identifyfuel pressure and then control the ‘quantity controlvalve’ to regulate pressure to the required value.

Three barrel pumpThe three barrel pump shown in Figure 3.55 has threepumping plungers which are forced to move along thebarrel through the rotation of a cam ring located on themain pump shaft (an eccentric element). The pump isusually mounted on the engine, with the pump shaftdriven by the engine camshaft. While the engine isturning, the three plungers will move up the barrels dueto the action of the cam ring, and a return spring forcesthe plungers to return down the barrels. This movementof the plungers forces fuel (delivered by the lowpressure pump to the high pressure pump) to bepumped out to the fuel rail at high volume andpressure. A separate pressure control valve thenregulates fuel pressure.

The use of three plungers and the shape of the camring ensure that the pumping action from the threeplungers overlaps; this reduces the pressure pulsationsand fluctuations produced by the individual plungers.

Single barrel pumpThe single barrel pump is mounted so that a cam lobeon the engine camshaft can act against a single plunger.Fuel flows from the low pressure system into the highpressure plunger, and, in the example shown in Figure3.56, the pump uses the integral ECU controlledquantity control valve to control the quantity of fuelthat is able to flow back to the return line. When thevalve is fully open, all of the fuel pumped by the


Figure 3.53 Continuous fueldelivery system

Figure 3.54 Demand controlledfuel delivery system

1 High pressure pumpHDP1

2 High pressure sensor3 Fuel rail4 Pressure control valve5 High pressure fuel

injectors6 Fuel tank with pump

module, including pre-supply pump

1 High pressure pumpHDP2

2 High pressure sensor3 Fuel rail4 Pressure limiter5 High pressure fuel

injectors6 Fuel tank with pump

module, including pre-supply pump

[647] Chapter 03 25/8/06 15:38 Page 122


Figure 3.55 Three barrel high pressure pumpa Longitudinal sectionb Cross-section

Figure 3.56 Single barrel high pressure pump

plunger flows back to the return line, but when thevalve is closed, all the fuel flows to the fuel rail. Bycontinuously moving the control valve between theopen and closed positions, the amount of fuel flowingto the fuel rail can be controlled as required by theinjectors.

A pulsation damper or pressure attenuator is used todampen the pressure pulses produced by the singleplunger. The damper has a diaphragm that is forced

against the fuel pressure by a spring. When pressurepulses occur, the diaphragm moves against the spring,creating a larger volume above the plunger whichslightly reduces the pressure. As the plunger rises andfalls, creating pressure pulses, the diaphragm alsomoves to create larger and smaller volumes above theplunger, thus damping the pressure pulses. In addition,the fuel rail contains a large volume of fuel, which alsohelps to reduce pressure fluctuations.

1 Eccentric element2 Slipper3 Pump barrel4 Pump plunger (hollow

piston, fuel inlet)5 Sealing ball6 Outlet valve7 Inlet valve

8 High pressureconnection to rail

9 Fuel inlet (low pressure)10 Cam ring11 Axial seal (sleeve seal)12 Static seal13 Input shaft

1 Fuel inlet (low pressure)2 High pressure connection

to rail3 Leakage return4 Outlet valve5 Inlet valve6 Pump plunger7 Piston seal8 Pump barrel9 Fuel quantity control valve

10 Fuel pressure attenuator

[647] Chapter 03 25/8/06 15:38 Page 123

Injection system pressure during startingWhen the engine is being cranked for starting, there islittle or no residual pressure in the high pressure system.Full pressure is not produced by the high pressure pumpuntil the engine is turning over at higher speeds. Duringstarting, the injectors are timed to inject during theintake stroke when the cylinder pressure is low; it istherefore possible to use low pressure fuel (that isflowing out of the high pressure pump) for starting.When the engine is turning at a sufficient speed, so thepump is delivering high pressures, the pressure sensorsignal indicates the higher pressure value to the ECU,which can then alter the injection timing to the

compression stroke, provided, of course, that this isappropriate for the operating conditions.

Injection pressure in a gasoline direct injectionengine is much higher than a port injectedgasoline engine. This pressure is monitored andcontrolled by the ECU

The fuel supply system consists of a low pressurepump to lift the fuel from the tank as well as a highpressure pump to raise the pressure to injectionpressure levels. The high pressure pump is enginedriven; the low pressure pump is electrical

Key

Poin

ts


3.5 EMISSIONS AND EMISSION CONTROL (PETROL ENGINES)

Air:fuel ratio

Lambda scale

Rich(lack of oxygen)

Weak(excess oxygen)

12:1 13:1

0.9 0.97 1.03 1.11.0

14.3:1 15.14:1 16:1 17:1 18:114.7:1

Lambda window

Figure 3.57 Lambda (excess air factor) compared with the air:fuel ratio scale

3.5.1 Lambda and the stoichiometricair:fuel ratio

The excess air factorReferences have been made here and in Hillier’sFundamentals of Motor Vehicle Technology: Book 1 to thestoichiometric air:fuel ratio, i.e. the ratio of air and fuelthat would in theory provide complete combustion. It isusually quoted as 14.7 parts of air to 1 part of fuel byweight (e.g. 14.7 grams of air for 1 gram of petrol). Inreality, it is oxygen contained within the air which isrequired for the combustion process. To obtain theappropriate quantity of oxygen, it is necessary to mixthe air and petrol in the stoichiometric ratio. It isinteresting to note that the quoted ratio is 14.7:1 byweight, but if the volumes of the two elements arecompared, the volume of air is approximately 9500times larger than the volume of petrol. Therefore everylitre of petrol burned in an engine requires 9500 litres ofair to be drawn in (assuming the stoichiometric ratio isused).

There has been a tendency in recent years to refer tothe correct ratio of petrol and air as the ‘air factor’ ormore precisely the ‘excess air factor’. The stoichiometricair:fuel ratio should provide the correct amount ofoxygen, which can be regarded as 1. The Greek symbollambda (λ) is used to indicate the excess air factor asshown below:

1 If the mixture is correct, then the excess air factor iscorrect, which is expressed as λ = 1 (lambda = 1).

2 If there is too much air (a weak mixture) then theexcess air factor is greater than 1, which is expressedas λ > 1 (lambda is greater than 1).

3 If there is too little air (a rich mixture) then theexcess air factor is less than 1, which is expressed asλ < 1 (lambda is less than 1).

Emissions control systems often function efficientlyonly when there is the appropriate amount of oxygen inthe exhaust gas (see section 3.5.6 and other parts ofsection 3.5).

Comparison of lambda and air:fuel ratiosFigure 3.57 shows a comparison between the air:fuelratio scale and the lambda scale. Note that thestoichiometric value of 14.7:1 relates to lambda = 1 (λ=1). When the air/fuel mixture is weaker, the lambdavalue increases, i.e. lambda is greater than 1 (λ > 1).When the air/fuel mixture is rich, lambda is less than 1(λ < 1).

Lambda windowAlthough the ideal air:fuel ratio is 14.7:1 (λ = 1), thereis a small tolerance or window for an air:fuel ratio thatresults in low emissions and good combustion. Theamount of oxygen contained within the exhaust gas iscritical to the operation of catalytic converters (section3.5.6) and some other emission reducing devices. Toensure the correct amount of oxygen is containedwithin the exhaust (for efficient catalytic converteroperation) it is necessary to operate the engine at thestoichiometric air:fuel ratio (i.e. λ = 1). At this ratio,

[647] Chapter 03 25/8/06 15:38 Page 124

the CO, HC and O2 are in a balance so the catalyticconverter can function at its most efficient and reducepollutant levels. However, as mentioned above, there isa window of tolerance for the air:fuel ratio, withinwhich the catalytic converter can still functionefficiently; this tolerance is referred to as the lambdawindow.

The lambda window range is generally quoted at alambda value of 0.97 to 1.03. Figure 3.57 provides anindication of the air:fuel ratio corresponding to thelambda window.

If the air:fuel ratio is controlled accurately so thatthe excess air factor is always within the lambdawindow, the catalytic converter will operate atoptimum efficiency at all times. However, section3.5.4 explains that under certain engine operatingconditions, it is necessary to operate with air:fuelratios that are outside the lambda window.

3.5.2 Pollutants and theenvironment

Worldwide problemEnvironmental considerations have forced manycountries to introduce regulations to limit the pollutioncaused by motor vehicles. Emissions from motorvehicles damage human health, plant life and theenvironment. Problems are particularly severe in areaswhere the geographic and climatic conditions create anatmospheric envelope which traps the pollutants.

The first countries of the world to introducestringent emission controls were the USA, Australia,Japan and Sweden. The EU has for many yearsenforced strict emission regulations, especially sincethe early 1990s. This has resulted in very significantchanges in automotive technology and design. Eachtime new emission standards are introduced, the limitsare reduced. Manufacturers have to continually updatetheir vehicles to meet requirements adopted in thecountry in which they are to be sold.

Although most industrial nations now recognise theproblems of pollution from vehicles, and thereforeimpose legislation, not all countries or regions imposethe same standards. However, the main vehicleproducing nations have to manufacture vehicles thatcomply with the tougher regulations imposed in thecountries where their vehicles are sold in volume, soemissions control systems are still reasonably effectiveeven for those vehicles sold into countries wherelegislation is weaker.

Pollutants and the petrol engineWith vehicles powered by petrol engines, there arespecific areas of the vehicle’s operation that causepollution:

● exhaust gas – can contain unburned fuel (HC),partially burned fuel (CO), dangerous nitrogenoxides (NOx) from combustion, and lead (Pb) from

petrol additives (leaded fuel is no longer widelyused on most modern vehicles)

● crankcase – during engine operation, emissions arepassed into the crankcase, including somecombustion gases which pass the piston, vaporisedlubrication oil (HC) and corrosive acid compounds

● fuel system including the fuel tank – the petrol thatis stored in the fuel tank gives off a vapour (HC).

Devices used to ‘clean up’ vehicle pollutants are costly,and their use has often resulted in a lower power outputand higher fuel consumption. Consequently, the trendhas been for manufacturers to fit emission controldevices only if they are required to meet local countryregulations. However, modern technologies arechanging this situation, because most emission controltechnologies are not as restrictive to engine power or as‘fuel thirsty’ as older emission control systems. Emissioncontrol is therefore becoming more consistent acrossmost regions of the world.

Conflicting requirementsEmission control and emissions reduction systems areoften complex in their operation, and frequently rely onchemical and thermal reactions to achieve the desiredresults. However, there have been, and still are, manysystems or designs which reduce one or more pollutantsbut which increase the emission of other pollutants.When this happens, it is necessary to treat the increasedpollutant, so, emission control systems must worktogether to reduce the levels of all the variouspollutants.

It is also true that some emission reduction systemsreduce power or increase fuel consumption, but this isbecoming less of a problem owing to the technologiesavailable, and the philosophy of ‘reducing pollution atsource’; i.e. engines are now designed to be moreefficient or ‘pollution conscious’. This is a reversal of thetrend from the early days of emission control whenexisting engine designs produced high emissions levels,and treatment of the pollution occurred after thepollutants had been created by the engine. However,even the latest engines with exceptional combustionefficiencies still produce pollutants that must bereduced by subsequent means. This method of pollutantreduction is often referred to as ‘after treatment’.

One conflicting area of emission control that isincreasingly becoming a focus of attention is theimproved efficiency of converting the pollutants intothe so-called harmless gases. Within the exhaust gas,the three largest components are in fact not normallyregarded as pollutants:

● nitrogen (N2 ) – approximately 71.5% of the exhaustgas (but note that nitrogen forms almost 78% of theatmosphere, and is not regarded as a pollutant)

● water (H2O) – approximately 13.1% of the exhaustgas (again, not regarded as a pollutant)

● carbon dioxide (CO2 ) – a product of complete orefficient combustion, representing approximately

Emissions and emission control (petrol engines) 125

[647] Chapter 03 25/8/06 15:38 Page 125

13.7% of the exhaust gas (not regarded as a harmfulpollutant but is now of concern, as explainedbelow).

The primary concern with regard to the harmless gasesis carbon dioxide (CO2), which is not a toxic or directlyharmful pollutant but which does have an influence onglobal warming. Unfortunately, the more complete thecombustion of fuel, the greater the level of CO2.

Note that CO2 naturally exists in the atmosphere;animals breathe out CO2, and it is absorbed by plants.The problem is that the level of CO2 in the atmospherehas increased (due to the burning and combustion offossil fuels), which is upsetting the natural balance ofatmospheric gases. This is added to by the destructionof plant life (especially the rainforests). The overallincrease in CO2 is a significant factor in the greenhouseeffect which leads to global warming.

The problem is that increased efficiency ofcombustion leads to an increase of CO2. Furthermore,when some of the true pollutants are treated, they arealso converted to CO2. In effect, during combustion, thehydrocarbons in the petrol mix combine with oxygen inthe air to form CO2, and any unburned or partiallyburned fuel emitted from the combustion chamber isthen treated in a catalytic converter and also turnedinto CO2.

The only effective way of reducing production ofCO2 is to reduce the consumption of fuel. This can beachieved by operating on leaner (weaker) air:fuelratios, such as are achieved with direct injectionsystems.

Composition of the exhaust gasFigure 3.58 shows the composition of exhaust gas whenthe combustion process takes place with astoichiometric air:fuel ratio (λ = 1).

Emission regulations for petrol enginesTest programmeBefore any new vehicle can be sold in the EU, and manyother countries, a number of examples of the vehicle

must be submitted to the regulating body of a memberstate for tests to be carried out. This is to test whetherthe vehicle type meets current legislated standards.

Each vehicle has to complete a test cycle whichreflects the vehicle being driven in an urbanenvironment. The process is similar in most countrieswhere emissions legislation is in force.

The vehicle is started from cold and subjected tovarious speeds, including motorway type driving. Tomonitor the emissions accurately the tests are carriedout with a vehicle dynamometer. The tests involve astandardised driving procedure, including engine atidle, as well as gear changing and braking, to simulatedriving conditions in a reasonably sized town.

Previous emission regulations limits weredetermined on the cubic capacity of the engine.However in 1992 the EU introduced EU Stage Iemission regulations with which all new vehicles had tocomply (Figure 3.59). In January 1996, the EU imposedstricter emission limits with EU Stage II. In 2000,vehicle manufacturers had to ensure that vehicles metEU Stage III emission standards before they could besold. In 2005 all new vehicles had to comply with EUStage IV.

So far we have described engine combustion relatedemissions. Vehicles are also subject to evaporativeemission testing, i.e. emissions that are emitted throughthe vaporisation of fuel stored in the fuel tank andcontained in the fuel pipes. These tests are carried outin a gas-tight chamber at various ambient temperatures,with the engine stationary and running.

Engines submitted for test must also be designed torun on unleaded petrol, to reduce lead based additivesin fuels. In addition the petrol tank filler pipe must bedesigned to prevent the tank from being filled from apetrol pump delivery nozzle with an external diameterof 23.6 mm or greater. This regulation means thatnozzles of pumps supplying lead free petrol have to besmaller than those used with pumps dispensing leadedfuel.


0.7% miscellaneous (noble gases, oxygen, hydrogen

13.1% water (H2O)

13.7% carbon dioxide (CO2)

71.5% nitrogen (N2)

0.1% nitrous oxides (NOx)

0.7% carbon monoxide (CO)

0.2% hydrocarbons (HC)

0.005% particulates

Figure 3.58 Composition of exhaust gas

[647] Chapter 03 25/8/06 15:38 Page 126

Different emissions reduction technologiesThe method adopted by an engine designer to meet theemission limits depends on the technology available toa manufacturer. When it appeared likely that one optionto reduce emissions would involve the fitting of acatalytic converter and its accompanying fuel mixturecontrol system, the likely cost for use on small andmedium sized cars would be relatively high comparedwith the cost of the vehicle. This encouraged manymanufacturers to develop lean burn engines. Testsshow that this type of engine gives a good fuel economyand a much lower emission level than conventionalengines. It was expected that the lean burn enginewould meet the expected future emission requirements,but the introduction in 1989 of a more stringentstandard that was based on a new European extra-urban driving cycle (EUDC) meant that to attain thenew limits, even the lean burn engine would need to befitted with an exhaust catalyst.

It is only recently that developments in direct petrolinjection systems met with the requirements to use lessfuel (for CO2 reductions), so lean burn processes areagain becoming viable. The use of stratified mixtureformation (see section 3.4) and different types ofcatalytic converter design now enable very leanmixtures to be used under light load and at lowerengine speeds.

3.5.3 The pollutantsCreation of pollutantsAs noted earlier, when complete combustion occurs,the fuel and oxygen combine to form carbon dioxide(CO2). Put simply, one carbon atom combines withtwo oxygen atoms. The thermal reaction within thecylinder (the combustion process) is most efficientwhen the air:fuel ratio is stoichiometric. In theory, ifthis ratio is used all the time, all of the fuel and oxygencombust and produce carbon dioxide. Therefore, intheory there should be no unburned fuel or unburnedoxygen. Carbon dioxide is a product of completecombustion. Water (H2O) and nitrogen (N2) are alsoemitted. With the carbon dioxide they form just over98% of the exhaust gas. However the remaining gasesinclude oxygen and hydrogen but also includepollutants, which represent around 1% of the totalexhaust gas.

The 1% of exhaust gas that is regarded as pollutioncan be broken down into three main pollutants (seeFigure 3.60): carbon monoxide (CO), hydrocarbons(HC), and oxides of nitrogen (NOx). In addition, verysmall percentages of ‘particulates’ exist which areeffectively soot, but for petrol engines the percentage isexceptionally low and generally not of concern. Dieselengines produce much more soot, which is regarded asa diesel engine pollutant that must be treated.


Figure 3.59 EU Emissions regulations

Standard Year of introduction CO g/km HC g/km NOx g/km NOx + HC g/km

EU Stage I July 1992 2.72 N/A N/A 0.97EU Stage II January 1996 2.2 N/A N/A 0.5EU Stage III January 2000 2.3 0.2 0.15 N/AEU Stage IV January 2005 1.0 0.1 0.08 N/A

Figure 3.60 Main pollutants from internal combustion engines

Pollutant Origin Effects

Carbon monoxide Incomplete combustion or partially Poisonous to human beings when inhaled, CO adheres (CO) burned fuel to haemoglobin in the blood and prevents oxygen being

carried to body cellsHydrocarbons Unburned fuel, vaporised fuel escaping Irritates eyes and nose. Cancer risk. Odour(HC) from fuel systemCarbon (C) Partially burned fuel Smoke – restriction in visibility. Can carry carcinogens

(cancer causing agents). OdourNOx (oxides of Very high combustion temperatures Toxic to humans. NO2 combines with water to form nitrous nitrogen – NO cause nitrogen to combine with oxygen. acid, which causes lung disorders. It combines with and NO2) Highest when air:fuel ratio is just slightly other exhaust products to give eye and nose irritants;

weaker than stoichiometric ratio it also affects the nervous system. Component of smogLead (Pb) Added to petrol to raise octane rating. Toxic to humans, causing blood poisoning and nervous

Lead is no longer allowed as an additive disordersin fuel in most regions; it also damages catalytic converters

[647] Chapter 03 25/8/06 15:38 Page 127

The exhaust gas also includes sulphur compounds invery small quantities and these are not regulated bylegislation. However, the compounds (primarilysulphur dioxide (SO2)) are produced because ofsulphates in the fuel. The amount of sulphate in the fuelis subject to legislation which has progressivelyrestricted the level from 1000 ppm (parts per million)down to 50 ppm.

Figure 3.60 shows the main pollutants that aresubject to emissions legislation and therefore subject toreduction processes. Those pollutants that occur duringcombustion are referred to as byproducts ofcombustion.

Treating or reducing the pollutants There are effectively two main routes to reducingpollutants.

1 The first route is to design the engine, fuel systemand ignition system so that lower levels of

pollutants are produced during combustion. Muchprogress has been made in these areas in recentyears and modern designs of combustion chambers,fuel systems and ignition systems have allcontributed to substantial reductions in pollutants.

Importantly, providing the correct air:fuel ratiohas a major influence on the levels of pollutants.Figure 3.61 shows how CO, HC and NOx emissionsare affected when the air:fuel ratio changes. Otherfactors affect the levels of pollutants produced (suchas high temperatures) but the chart indicates thechanges in pollutants assuming other factors remainthe same. Note that the chart indicates the trend inthe gas values and not the actual values.

2 The second route to reducing pollutants is to changethe pollutants chemically once they have left thecombustion chamber (after treatment). There are anumber of ways in which this is achieved, describedin section 3.5.5 through to section 3.5.10.


Air:fuel ratio

Lambda scale


Weak(excess oxygen)

12:1

Scale for CO

2 (%)

Scale for CO

(%)

10%

9%

8%

7%

6%

5%

4%

3%

2%

1%

CO2

HC

CO

Scale for NO

x (parts

per million)

NOx

3000

2500

2000

1500

1000

500

5%

10%

15%

Scale for CO

(%)

1%

2%

3%

13:1

0.9 0.97 1.03 1.1

1.0

14.3:1 15.14:1 16:1 17:1 18:1 19:114.7:1

Lambda window

Lambda window

Scale for HC (partsper million)

1000

900

800

700

600

500

400

300

200

100

Air:fuel ratio

Lambda scale


Weak(excess oxygen)

12:1 13:1

0.9 0.97 1.03 1.1

1.0

14.3:1 15.14:1 16:1 17:1 18:1 19:114.7:1

1. It can be seen from the graphs that when all the relevant gases are taken into consideration, the best compromise is when the air:fuel ratio is at or around 14.7:1 (lambda =1).

2. When the excess air factor is around lambda = 1 (within the lambda window) the CO2 is at its highest, indicating relatively complete combustion.

3. It is only the NOx emissions that are high when the excess air factor is in the region of lambda = 1 (theoretically correct air:fuel ratio). Note that the NOx peaks when the mixture is slightly weak of 14.7:1 air:fuel ratio (lambda = 1.05 approximately).

4. When the excess air factor is less than lambda = 1, there is virtually no O2. When the air:fuel ratio is weaker (lambda greater than 1), the O2 level rises significantly.

Note. The values shown for all of the gases are typical for an engine operating at medium load conditions. The values will change with load, engine design, fuel and ignition system designs.

Figure 3.61 The influence of air:fuel ratio (lambda/value) on pollutant levels

[647] Chapter 03 25/8/06 15:38 Page 128

Carbon monoxide (CO)Carbon monoxide (CO) is formed when fuel is onlypartially burned (incomplete combustion). A carbonatom from hydrocarbons (fuel) combines with a singleoxygen atom (from the inducted air) in the cylinder toform CO. Compared with CO2, CO lacks one oxygenatom, due to a deficiency of oxygen in the combustionmixture (or a pocket of mixture). Carbon monoxide istherefore formed when a cylinder receives either a richor poor mixture of fuel and air, which leads to isolatedpockets of rich mixture. Carbon monoxide can also formwhen the mixture is excessively weak, when fueldroplets do not vaporise; however, formation of CO in aweak mixture is not at the same level as in a richmixture. Carbon monoxide measurement is therefore agood indicator of a rich mixture, but not of a weakmixture.

Methods of reducing CO include:

● Control of mixture strength – control of the air:fuelratio by the fuel system, especially under slowrunning and cold starting conditions. Enginemanagement control has improved this under allconditions.

● Improved fuel distribution – multi-point fuelinjection has largely overcome this problem, withdirect injection providing further improvement.Good distribution was difficult to achieve with acarburettor or single-point injection system.

● More precise engine tuning – engine managementsystems ensure that the correct air/fuel mixture issupplied to the engine during all operatingconditions; therefore no manual adjustment ispossible to alter the mixture strength.

● Compact combustion chamber – modern enginedesigns incorporate very compact combustionchambers. Long narrow chambers associated withan ‘over-square’ engine often gave a high COcontent.

● Improved mixing of air and fuel – intake port andcombustion chamber design (including pistoncrown shape) can help to promote good mixing; thisis especially true for engines using direct injection.

● Leaner air/fuel mixtures – the recent trend towardslean mixture operation (stratified mixtureformation) has helped reduce CO levels.

● Throttle positioner system – these open the throttleslightly when the engine is at idle or whendecelerating.

● Precise ignition timing – ensures that the sparkoccurs at the correct time and remains constantbetween servicing intervals. Computer controlledignition and engine management systems achievethis.

● After treatment – catalytic converters and othersystems help to convert CO into CO2 (covered laterin this section).

Hydrocarbons (HC)Hydrocarbons in the exhaust gas represent unburnedfuel from incomplete combustion. A rich mixture (lackof oxygen or excess fuel), results in high levels ofhydrocarbons, because there is insufficient oxygen tocombine with the fuel during the combustion process.Any reduction in combustion efficiency will result inhigh levels of hydrocarbons, e.g. a cylinder misfirecaused through an ignition fault or reducedcompression (a mechanical fault). Excessively weakmixtures can also result in high levels of hydrocarbons,because excessively weak mixtures cannot supportcomplete combustion within the combustion chamber.However, careful design of the engine and fuel systemsreduces this problem to a level where weak mixturesproduce very low levels of hydrocarbons.

Methods of reducing hydrocarbon emissionsinclude:

● Control of mixture strength – control of the air:fuelratio by the fuel system, especially under slowrunning and cold starting conditions. Enginemanagement control has improved this under allconditions.

● Improved distribution of the fuel – multi-point fuelinjection has largely overcome this problem, withdirect injection providing further improvement.Good distribution was difficult to achieve with acarburettor or single-point injection fuel system.

● More precise engine tuning – engine managementsystems ensure that the correct air/fuel mixture issupplied to the engine during all operatingconditions; therefore no manual adjustment ispossible to alter the mixture strength.

● Improved mixing of air and fuel – intake port andcombustion chamber design (including pistoncrown shape) can help to promote good mixing; thisis especially true for engines using direct injectionsystems.

● Leaner air/fuel mixtures – the recent trend towardslean mixture operation (stratified mixtureformation) helps reduce HC levels.

● Mixture adjustment during deceleration – fuelinjection systems provide precise metering of thefuel during deceleration (decel fuel cut off orreduction).

● Precise ignition timing – ensures that the sparkoccurs at the correct time and remains constantbetween servicing intervals. The ignition timing isretarded when the engine is slow running ordecelerating. Computer controlled ignition andengine management systems achieve this.

● After treatment – catalytic converters and othersystems help to convert HC into CO2 (this is coveredlater in this section).


[647] Chapter 03 25/8/06 15:38 Page 129

Hydrocarbons are also formed when fuel vaporises andescapes into the atmosphere from the fuel system, andwhen unburned fuel passes the pistons into thecrankcase. These problems can be reduced in thefollowing ways:

● Closed crankcase ventilation system – unburnedfuel passing the pistons and entering the crankcaseis prevented from escaping to the atmosphere by apositive crankcase ventilation (PCV) system. Theunburned fuel is returned to the induction system.

● Sealed fuel system – a fuel evaporative emissioncontrol (EVAP) system seals the fuel tank, collectsthe vaporised fuel, passes it through a charcoal filledcanister and delivers it to the induction manifold forcombustion in the engine cylinders (this is coveredlater in this section).

Oxides of nitrogen (NOx)Nitric oxide (NO) and nitrogen dioxide (NO2) aregrouped together under the term oxides of nitrogen(NOx). The atmosphere consists of approximately78% nitrogen and 21% oxygen. The air drawn intothe combustion chamber is heated duringcombustion; under certain conditions the oxygen andnitrogen can combine to form harmful NOx. Theformation of NOx occurs with combustiontemperatures above approximately 1300°C. However,combustion temperatures can easily exceed 2500°Cduring full load conditions when the production ofNOx reaches a critical limit. Formation of NOx is alsoaccelerated when the air:fuel ratio is slightly weakerthan stoichiometric.

Methods of reducing NOx include:

● Combustion chamber shape – the shape of thecombustion chamber can be designed to increaseflame speed. Used in conjunction with lowercompression ratios, such chambers reduce NOx

formation, but normally only at the expense of fueleconomy and engine power.

● Increase in air:fuel ratio – the highest flame speedand NOx content occurs when the mixture is about12% richer than the stoichiometric value. An enginedesigned to operate on a weak mixture has reducedemissions, but vehicle driveability suffers unless theignition timing and air/fuel mixture are set correctly.However, direct injection systems with stratifiedmixture formation help to overcome this problem.

● Ignition timing – computer controlled ignitionsystems can control the ignition timing to preventsudden advance in ignition timing (for a given brieftime period) when the throttle is snapped open.

● Valve timing – by changing the inlet and exhaustvalve timing (i.e. the overlap period), the combustiontemperature can be lowered by inducing exhaust gasinto the intake port (the process of using exhaustgases to reduce combustion temperatures is coveredlater in this section). Variable valve timing canoptimise the overlap period during engine operation.

● Intake air temperature – reducing the intake airtemperature can lower the combustion temperatureand therefore lower NOx production. If the engine isfitted with a turbocharger, the fitting of anintercooler can reduce NOx emissions significantly.

● Decrease in flame speed – exhaust gas recirculation(EGR) systems direct some exhaust gas back to theinduction manifold to slow down the combustionwhen the engine is under certain load conditions.The EGR system does however reduce the maximumpower of the engine.

● After treatments – the fitting of a three-way catalystin the exhaust system reduces the level of NOx. Morerecent designs of catalytic converter include NOx

storage catalysts, covered later in this section.

‘Catalyst’ is the name given to a material that producesor hastens a chemical action without undergoing anychange itself.

Carbon dioxide (CO2)In theory, if the correct amount of air combines with thecorrect amount of fuel during perfect combustion, thisresults in carbon dioxide (CO2), water (H2O) andnitrogen (N2). Carbon dioxide is therefore a product ofcomplete combustion. Although it is not possible forperfect combustion to occur in the ‘real world’, the moreefficient the combustion process, the higher the CO2

content in the exhaust. It is therefore necessary toprovide the correct air:fuel ratio to produce as perfectcombustion as possible. Any fault in the ignition system,fuel system or combustion efficiency will lower the CO2

content in the exhaust gas.Carbon dioxide is not directly harmful to humans

and is not regarded as a pollutant, but it is harmful inthe long term to the environment, and contributes toglobal warming. Therefore, if the combustion process isefficient and can be operated with weak mixtures, lessfuel should be used, reducing CO2 emissions.

Oxygen (O2)Oxygen is an essential element to the combustionprocess. During the combustion process the oxygenshould combine with hydrocarbons to form carbondioxide and water, leaving no oxygen or hydrocarbonsin the exhaust gas. Although not a harmful emission,the exhaust gas contains a very small percentage ofoxygen (which is effectively unburned during thecombustion process). A rich mixture will result in nooxygen in the exhaust gas whilst a weak mixture(whether intended or caused through insufficient fuelor an air leak in the inlet manifold) will result in highoxygen content in the exhaust gas.

A reduction in combustion efficiency (e.g. a misfire)will result in some of the fuel and oxygen not beingburned, which will increase the oxygen content in theexhaust. Excess oxygen can combine with additionalgases to form other pollutants such as NO. If the air:fuelratio is chemically correct and the combustion process isefficient, oxygen emissions should be almost zero.


[647] Chapter 03 25/8/06 15:38 Page 130

3.5.4 Influences of engine operationon the pollutants

In theory, a petrol engine should operate at thestoichiometric air:fuel ratio at all times, because thisratio should result in complete combustion of theoxygen and petrol: harmful emissions should be lowand power should be at its optimum level. However, anengine is not a perfect machine. Other requirements,such as reduced fuel consumption, dictate a need to useair:fuel ratios that are either slightly richer or slightlyweaker than the stoichiometric value.

Although there are many factors dictating the exactair:fuel ratio used in an engine, the following listprovides an understanding of the differentrequirements. Note that inefficiencies in older enginedesigns and older type fuel ignition systemsexaggerated many of the problems, such as poor mixingof the air and fuel. Therefore some of the problems arenot as severe as in the past, although in most cases theystill occur to a limited extent.

Light load/part loadIn general, under these conditions, an engine canoperate on mixtures where there is excess air, i.e. withweak mixtures. With the modern generation of directfuel injection and stratified mixture formation, mixturescan now be used with excess air ratios as weak as 40:1during light load conditions. Note however that if anengine is not operating with stratified mixtureformation and the emissions control includes a standardthree-way catalytic converter, it is necessary to operateat around the stoichiometric ratio to enable theconverter to convert the pollutants into harmless gases.

Full loadTo achieve maximum power, the stoichiometric ratioshould in theory be suitable; however, the air and fueldo not completely or perfectly mix (especially oncarburettor systems) which results in some fuel notmixing with the air; i.e. a weak mixture can exist in thecombustion chamber with any unmixed fuel unable toburn efficiently. Therefore, a slightly richer mixturemust initially be provided so that enough fuel is able tomix with the air and form a good combustible mixture.Any excess fuel will cause some rich pockets of mixtureto exist. In addition, there are conditions where fuel willcondense on the intake manifold and cylinder walls,which again will require additional fuel to ensure thatsufficient is available to form a good mixture.

For older engines with a less efficient design ofintake manifold and intake port, poor mixing of air andfuel was a major problem. Excessively rich mixtureswere provided by the fuel system to ensure thatsufficient correctly mixed fuel was available duringcombustion. Modern intake and fuel systems haveimproved the mixing process. It is therefore possible toreduce the amount of excess fuel for full load operation.

Cold starting/cold runningThe problems associated with cold engine operation aresimilar to the problems of operating at full loadconditions, i.e. poor mixing and fuel condensing on coldsurfaces. When the fuel is cold, it is more difficult tovaporise and therefore does not mix as easily with thecold air. In addition, the intake manifold and intake portwalls are cold, which causes atomised fuel to condense.This problem is repeated when the fuel contacts the coldcylinder walls. It is therefore necessary to provide anexcessively rich mixture to ensure that sufficient fuel canmix with the air and achieve combustion.

Using port type fuel injection (as opposed to acarburettor or single-point injection system) reduces theproblem of fuel condensing on the intake system walls.A further improvement is gained by injecting directlyinto the cylinder. However, cold cylinder walls stillcause condensation of fuel; therefore a rich mixture isstill required during cold starting.

When the engine has started, heat is passed to thecylinder walls so it is possible to reduce the amount ofenrichment. On older engines, it was necessary tomaintain a rich mixture immediately after starting, butthe use of fuel injection into the ports or into thecylinder reduces condensation problems so it is nowpossible to operate using air:fuel ratios that are at orclose to stoichiometric immediately after cold starting.

It is also general practice to provide a slightly fastidle speed after cold starting; this enables the engine toovercome power losses caused by increased friction andoil drag on cold engines. A fast idle also enables someloads to be applied to the engine (e.g. electrical or auto-transmission loads) which could otherwise cause theengine to stall. During the warm-up period, theincreasing temperature of the engine allowsprogressively weaker mixtures to be used and idlespeeds to be reduced.

Idle speed (normal operating temperatures)Modern engines generally operate using air:fuel ratiosaround the stoichiometric value or slightly weaker.Most engines are relatively inefficient at idle speed,especially engines with carburettors, where low airspeeds contribute to poor mixing of air and fuel. It istherefore necessary to provide relatively rich mixturesto ensure that there is sufficient fuel available to mixwith the air.

Modern engines are very efficient. Injection systemsand other design features improve the mixing of air andfuel at idle speed. However, valve timing and otherdesign features are generally compromised so that anengine is more efficient at normal operating speeds, i.e.at mid-engine speeds where most driving takes place.There is still a theoretical requirement for a slightlyricher mixture to be provided at idle speed. In reality,the use of idle speed control systems and improvedefficiencies enable engines to operate at stoichiometric(or close to stoichiometric), which is essential whereconventional three-way catalytic converters are used.


[647] Chapter 03 25/8/06 15:38 Page 131

TransitionWhen a driver changes the operating conditions of theengine, e.g. from light load to full load, this is referredto as a transition. One major problem with olderengines that used a carburettor was that air wouldrespond to a transition much more rapidly than petrol.Because air is less dense than petrol, when the throttleis suddenly opened, the volume of air flowing to theengine increases rapidly but the petrol takes longer torespond. It should also be remembered that theincreased flow of air created an increased depression inthe carburettor venturi which in turn caused additionalfuel to be drawn from the float chamber (fuelreservoir). There was, however, a small but importanttime delay in this process. Therefore, with mostcarburettor systems, when the throttle is initiallyopened, a momentarily weak mixture results.Carburettors were fitted with different devices thatcaused additional fuel to be delivered when the throttlewas opened, e.g. small pumps, directly operated by thethrottle linkage.

With injection systems, it is much easier to ensurethat the appropriate amount of fuel is injected at exactlythe same time the throttle is opened and the air volumeincreases. It is possible, so sensitive is the controlsystem, to detect the rate at which the throttle is beingopened, so that the fuel can be increased in anticipationof the air volume increasing.

During deceleration, when the load on the engine isreduced by the driver closing the throttle, it is possibleto provide a weak mixture or to cut off fuel deliverycompletely. Although this function was difficult toachieve with carburettors, injection systems can controlthe amount of fuel injected so that a progressivereduction (or cessation) in fuel supply is achievedwhich results in lower fuel consumption and reducedemissions.

3.5.5 Processes and devices foremissions control andreduction

Engine and engine control systemsAs noted earlier, modern engine design and controlsystems have been continuously improved with theresult that the whole fuel delivery, ignition andcombustion process is considerably more efficient thanin older engines. The overall effect is that engines nowproduce far lower levels of pollutants. The use ofcomputer controlled fuel injection and ignition systemshas had a dramatic effect on overall efficiencies, butfurther improvements have been achieved throughimproved engine design and electronic control ofmechanical systems. Examples include changes tocombustion chamber and intake port design as well aselectronic control of valve timing (variable valvetiming). Higher compression ratios and four valve per

cylinder designs also help to reduce the levels of mostpollutants, although in some cases there is a risk ofincreasing some pollutants, which then have to betreated separately.

Most of the significant changes and improvements,such as fuel injection and ignition systems, are dealtwith individually within this book, however, manychanges go almost unnoticed and simply form anevolutionary part of general engine development. Infact, many small changes or features might only beapplicable to one particular engine design and do notnecessarily justify individual explanation or coverage.However, technicians will encounter many individualdesign features when working on particular vehicles, soreference should always be made to specific vehicleinformation wherever possible.

Although engines now produce far less pollutionthan in the past, even the low levels now produced areregarded as excessive. Additional means are needed tofurther reduce pollutant levels. Sections 3.5.6 to 3.5.10cover the main pollutant reduction and control systemswhich are referred to as ‘after treatment’ systems.

Lean burn technologyLean burn technologies have been developed andapplied for many years, although there have beencertain limitations in the past, due to the control of NOx

emissions.Lean burn engines use very weak mixtures (excess

air) which is viable for light load engine operation. Foran engine to produce power or torque, a more enrichedmixture is required, closer to the stoichiometric air:fuelratio. Lean burn engines do, however, produce lowemissions of CO and HC, and use less fuel than enginesthat have to operate around the stoichiometric ratio formost of the time.

Lean burn engines generally use stratified mixtureformation (see section 3.4), whereby a small pocketof rich mixture is created adjacent to the spark plug,but the rest of the cylinder is filled with a weakmixture or with air containing no petrol. Thestratified mixture principle results in the rich mixturepocket igniting easily; the combustion of this pocketthen spreads through to the rest of the weak mixtureor simply heats the remaining air, thus causing gasexpansion. The overall mixture is weak which resultsin low consumption of petrol and low CO and HCemissions.

The use of mixtures slightly weaker thanstoichiometric can cause high NOx emissions; however,when the mixture is further weakened (by a largeexcess of oxygen) such as on lean burn engines, NOx

levels reduce owing to lower combustion temperatures.The problem is that the NOx does still exist in theexhaust gas of a lean burn engine, so a catalyticconverter is needed for further reduction. However,since there is an excess of oxygen in the exhaust gaswhen the CO and HC are catalytically converted to CO2,the oxygen molecules are taken from the excess oxygen


[647] Chapter 03 25/8/06 15:38 Page 132

and not from the NOx. This means that the NOx does notlose its oxygen and is therefore not reduced to nitrogen.More recent developments have enabled direct injectionsystems to provide a high level of control over mixtureformation (see section 3.4), because direct injection canswitch from stratified mixture operation (an overallweak mixture with a rich pocket) to homogeneousmixture operation (a normal air:fuel ratio, close tostoichiometric throughout). Additionally, recentlydeveloped NOx accumulator catalytic convertersovercome the problem of untreated NOx escaping to theatmosphere (see section 3.5.6).

After treatment systemsIn many after treatment systems, a process of chemicalor thermal reaction is used to convert harmful gasesinto less harmful or harmless gases. Whilst an in-depthunderstanding of chemistry would be an advantage,you need only to appreciate the basic process by whichchemical conversions take place.

For those systems that do not directly use chemicalor thermal reactions, individual explanations areprovided in the following sections.

Chemical change and thermal reactionsIn very simple terms, a thermal reaction uses heat tocreate a change in a substance, e.g. the change of waterinto steam. However, thermal reactions can promote amore fundamental change in substances and can causeor assist different substances to combine (chemicalchange); this phenomenon is used to reduce harmfulemissions.

With exhaust gases, it is of interest to note thatmuch of the chemical change occurs by movingoxygen atoms from one gas to another. Carbonmonoxide (CO) has one oxygen atom (hence‘monoxide’); by adding another oxygen atom themonoxide changes to dioxide. A similar process occurswith hydrocarbons (HC). Two oxygen atoms cancombine with each carbon atom to form CO2.Although oxygen atoms can be extracted from thesmall amount of unburned oxygen in the exhaust gas,there are also oxygen atoms contained within theoxides of nitrogen (NOx). The chemical changes thatoccur include removing oxygen from NOx; the oxygencan then combine with CO and carbon from the HC toform CO2. The advantage is that the NOx now losesoxygen molecules to leave just nitrogen which is not apollutant. In reality, NOx in exhaust gas is generallymade up of nitric oxide (NO) and nitrogen dioxide(NO2), of which the latter is the most harmful. Whenoxygen atoms are removed some NO can remain,which is relatively harmless unless it later meets withmore oxygen (usually when the gases leave theexhaust pipe and enter the atmosphere).

The above brief explanation provides a very basicunderstanding of the process of chemical change, whichin reality is much more complex. It is sufficient toappreciate that such chemical changes occur, which

result in harmful gases being converted into relativelyharmless gases.

Oxidation process and catalystsIt has been mentioned previously that heat helps topromote chemical change; this includes oxidationprocesses (the reaction of oxygen with the gases). Aninitial oxidation process occurs within the engine itself –the combustion process in the cylinder. During thisprocess, oxidation of the fuel occurs and CO2 is formed.In reality, perfect oxidation or combustion does notoccur, so partially or completely unburned fuel (CO andHC) is produced.

By creating a secondary combustion or oxidationprocess, it is possible to convert most of the remainingCO and HC into CO2. Although various devices cancreate a secondary oxidation, the most commonly usedtype on modern vehicles is the three-way catalyticconverter; this is discussed in section 3.5.6. It should beremembered that a catalyst is a substance that helps topromote chemical change without actually changingitself. So, making use of heat and a catalyst, a secondaryoxidation process converts the harmful gases intoharmless ones.

The important point to remember is that for anoxidation process to be effective, spare oxygen must beavailable.

Other processes for treating pollutantsA number of other devices are used to convert or reduceemissions of harmful gases.

● Thermal afterburning – this process relies oninjection of air into the exhaust ports, which resultsin continued combustion when the gases have leftthe cylinder. It was used in the past to convert COand HC, and has been used in conjunction withcatalytic converters, the thermal afterburning beingeffective during the warm-up phase, before thecatalytic converter has reached operatingtemperature.

● Other thermal devices – some engines have in thepast been fitted with ‘hot spots’ in the exhaustsystem. In effect a glow plug device is located in theexhaust manifold or down pipe. The device can beelectrically heated but can also rely on exhaust heat.When the device is at high temperature, it causessecondary or continued combustion of the exhaustgases, thus converting CO and HC to CO2. This typeof thermal device is no longer widely used.

● Exhaust gas recirculation (EGR) – this is a processwhereby a controlled amount of the exhaust gas isfed back into the intake system. The air drawn intothe cylinder therefore contains a percentage ofexhaust gas, which has the effect of reducing thecombustion temperature. Because the levels of NOx

increase with high combustion temperatures, theuse of EGR to lower the combustion temperaturescauses a reduction in the amount of NOx produced.EGR is covered in detail in section 3.5.8.


[647] Chapter 03 25/8/06 15:38 Page 133

● Evaporative emission control (EVAP control) –evaporative emissions can be released into theatmosphere from the fuel tank and fuel system.Petrol vapour (HC) occurs naturally, so methods areused to prevent its escape. The main system in userelies on creating a sealed fuel system. Fuel vapoursare fed through to the engine intake system.However, when the engine is not running, vapoursare collected in a charcoal canister, which releasesthe vapour to the engine when it is running. EVAPsystems are covered in section 3.5.10.

● Valve and ignition timing – although not strictlyregarded as after treatment, control of valve andignition timing can help to reduce pollutants. Valvetiming can be arranged so that the intake valveopens before the exhaust valve closes (valve timingoverlap). In this way, during the induction strokesome of the exhaust gas that has not been expelledfrom the cylinder will mix with the fresh intake air.As with controlled exhaust gas recirculation, theexhaust gas helps to reduce the combustiontemperature and the formation of NOx. Variablevalve timing allows different overlap periods to beused at different engine speeds. Although thisfacility is aimed at producing improved torque andpower throughout the engine speed range, it alsoassists in maintaining lower emissions.

Ignition timing can be altered to help improvesome emission levels. An appropriate example iswhen ignition timing is retarded, so that thecombustion process is retarded, which means thatcombustion continues when the exhaust valveopens. The result is that excess oxygen in the exhaustsystem can now combine with the CO and HC duringa combustion process that continues within theexhaust ports. This process is very effective for coldengines where a rich mixture is required (an excessof CO and HC). The additional heat created by thelate combustion process also helps to heat up thecatalytic converter. The disadvantage is that the hightemperatures produced when the gases enter theexhaust ports (where there is oxygen) increase theformation of NOx.

3.5.6 Catalytic convertersPrinciple of operationCatalytic converters are the most commonly useddevices for after treatment for CO, HC and NOx

reduction. As the name implies, catalytic convertersconvert some of the exhaust gases, primarily thepollutants, into harmless gases. Converters use heat tochange the gases, but a catalyst is used to accelerate theprocess.

During the 1970s, oxidation catalysts were fitted(primarily for the US market) as a means of convertingCO and HC into CO2 (see the previous section). Thistype of catalyst was also referred to as a ‘single-bed’

converter. The catalytic converter was located in theexhaust system; the exhaust gases flowing through heatthe converter, enabling it to carry out the oxidationprocess. However, the subsequent requirement to alsoreduce levels of NOx resulted in an additional catalystbeing used, ahead (upstream) of the oxidation catalyst.The two catalysts were often combined into a singleassembly, referred to as a ‘dual-bed’ catalyst because theconversion processes still remained separate.

One problem with the oxidation catalytic converterwas that, if the system were fitted to an older enginethat operated with a relatively rich mixture, there wasinsufficient oxygen in the exhaust gas to allowoxidation to take place. This was also true if the NOx

reduction catalyst were fitted ahead of the oxidationcatalyst, because the NOx catalyst operated with arelatively rich mixture. It was therefore necessary to usean air injection system to ensure that the oxidationcatalyst had sufficient oxygen (Figure 3.62).

A significant development was the introduction ofthe three-way catalytic converter which was able toreduce the three main pollutants (CO, HC and NOx)within one converter. The three-way converter becamethe most widely used across the US, Europe andelsewhere.

Correct air:fuel ratios for different converterprocessesAs discussed in the previous section, there is a need forfree oxygen to enable CO and HC to be converted toCO2. However, oxidation catalysts (and dual-bedcatalysts) were fitted to earlier vehicles with enginesthat operated with relatively rich mixtures (a lack ofoxygen). These vehicles were often fitted withcarburettors as opposed to the more efficient fuelinjection systems, resulting in relatively poor mixtureformation. There was therefore insufficient spareoxygen to allow the oxidation process to take place. Toprovide the oxygen required by the catalytic converteran air injection pump was usually fitted to pump airinto the exhaust system just ahead of the oxidationcatalyst.

However, NOx reduction operates more efficientlywith a lack of oxygen (a slightly rich mixture or λ < 1).When the NOx flows through the NOx reduction catalyst,the oxygen will readily separate from the NOx, owing tothe lack of free oxygen in the exhaust gas, thus leavingjust nitrogen (N2). However, if there is already an excessof oxygen in the exhaust gas, there is a reducedtendency for oxygen to separate from the NOx. The NOx

reduction process is therefore more suited to enginesoperating on relatively rich mixtures in which oxygen islacking in the exhaust gas.

With the three-way catalytic converter, the processof NOx reduction and CO/HC conversion takes place inthe combined assembly. Although some spare oxygen isavailable in the exhaust gas, the oxygen required for COand HC conversion is also taken from NOx. Theimportant factor for the efficiency of a three-way


[647] Chapter 03 25/8/06 15:38 Page 134

converter is the balance of the gases; i.e. the oxygencontent must not be excessive, otherwise NOx reductionwill not take place. At the other extreme, too littleoxygen in the exhaust gas will prevent the oxidation ofCO and HC (the CO and HC will not combust andconvert into CO2). Additionally, too much CO and HC inthe exhaust gas will mean that there will be insufficientoxygen to facilitate full conversion to CO2. It is essentialthat the balance of the gases is correct, which meansthat the air:fuel ratio must also be correct. The correctoxygen level in the exhaust gas is achieved by operatingthe engine at the stoichiometric air:fuel ratio (λ = 1).

Figure 3.62 shows the three most commonly usedtypes of catalytic converter. A fourth type of converter isnow gaining popularity – an NOx accumulatorconverter. This type is now being used on vehiclesoperating with stratified mixture formation and directfuel injection. NOx accumulator converters are coveredlater in this section.

Operating temperaturesTo enable this chemical change in the gases, thecatalytic converter must operate at relatively hightemperatures. Because the conversion of CO and HC isbased on an oxidation process (in effect a second

combustion process), the temperature must besufficient to allow ignition of the gases. A typical idealtemperature is between 400°C and 800°C for most typesof oxidation converter.

Heat is provided by the flow of exhaust gas throughthe converter, but the oxidation process createsadditional heat. So long as the air:fuel ratio is correct,the oxidation process maintains the heat to continue theprocess.

For most types of catalytic converter, excessive heatwill cause either loss of efficiency or permanentdamage. Consistent temperatures of 1000°C or morewill permanently damage the converter, and longperiods of operation at between 800 and 1000°C willaccelerate the ageing process of the converter. Thelocation of the converter is therefore critical to preventexcessive heat being passed from the exhaust gases asthey exit the exhaust ports. Exhaust gases lose heat asthey flow through the exhaust system, so a locationfurther from the exhaust port is more suitable.

However, there is a need to heat up the converterimmediately after starting the engine. The closer theconverter is to the exhaust ports, the quicker thishappens. There has therefore been a tendency to use asmall ‘pre-cat’ located very close to the exhaust ports,


Figure 3.62 Catalytic converter systemsa Single-bed oxidation catalystb Dual-bed catalystc Single-bed three-way catalyst

[647] Chapter 03 25/8/06 15:38 Page 135

and a larger conventional converter in the exhaustsystem under the vehicle. The pre-cat is designed forhigher temperature operation and stability, whilst thelarger converter is designed for lower temperatureoperation. Note, however, that many vehicles are fittedwith a single high temperature converter located closeto the exhaust ports (usually just after the exhaustmanifold), as shown in Figure 3.63.

Construction of catalytic convertersAlthough there are a few variations in the constructionof catalytic converters, the majority of the three-waytypes are constructed as detailed below and shown inFigure 3.64. Note that the illustration also shows alambda sensor, which is discussed in section 3.5.7.

Maximum surface areaThe objective is to expose the exhaust gases to thecatalyst material. To enable as much of the exhaust gasas possible to be exposed to the catalyst, it is necessaryto use a series of small tubes that, together, have a largesurface area. If there is not enough surface area, theconversion process will not treat sufficient amounts ofgas. It must be remembered that the exhaust gases flowthrough the converter at high speed, so a large numberof small tubes ensures that the gas flow is not restricted,but at the same time the large number of tubes allowsthe maximum amount of surface area.

It is common practice to construct the converterusing a monolith or substrate that is a honeycomb madeof ceramic material (typically a magnesium aluminiumsilicate). Note that some monoliths are produced usinga finely corrugated, thin metal foil.

The surfaces of the tubes formed by the honeycombare thinly coated with aluminium oxide (referred to as awashcoat). The coating provides a rougher surface,with a much larger surface area than a smooth surface(imagine a smooth surface, such as a mirror, covered invery small bumps or hills). The aluminium oxide canincrease the surface area by as much as 7000 times.

Coating of precious or noble metalsThe active material, the catalyst material, is then addedto the washcoat. Active materials vary, but for theoxidation catalysts (CO and HC conversion), platinumand/or palladium are used. These two materialsaccelerate the oxidation of CO and HC. In a three-wayconverter, rhodium is also used to accelerate thereduction of NOx.

These active materials are expensive and arereferred to as noble or precious metals. However, onlyaround 2 to 3 grams of these materials are used to coatthe surface areas of the converter, as the coating isexceptionally thin.

CasingThe honeycomb or monolith is contained within a steelcasing. To protect the honeycomb from damage causedby vibration, etc., it is mounted in matting that swellswhen initially heated. The matting therefore forms aprotective layer around the honeycomb and also formsa gas seal.

Potential faults and problems with catalyticconvertersHigh temperatures and misfiresAs was mentioned, catalytic converters operate ideallybetween 400°C and 800°C, with temperatures muchhigher than this causing accelerated ageing orpermanent damage.

One major factor that can increase the temperaturewithin the catalytic converter is an engine misfire.When a misfire occurs, a quantity of petrol and airremain unburned, and these flow out of the cylinderand through the exhaust system into the catalyticconverter. The heat in the exhaust system and existingcombustion process within the converter will cause theunburned air and fuel to ignite and combust and thisadditional fuelling of the combustion process within theconverter creates excessive heat that can lead totemperatures in excess of 1400°C. Such temperatures


Figure 3.63 Location of catalytic converters

1 Engine2 Lambda oxygen sensor

upstream of the catalyticconverter (two-step sensoror broadband sensor,depending on system)

3 Three-way catalyticconverter

4 Two-step lambda oxygensensor downstream of thecatalytic converter (only onsystems with lambda dual-sensor control)

[647] Chapter 03 25/8/06 15:38 Page 136

will quickly melt the substrate or monolith in theconverter, causing permanent damage. Note, however,that even brief misfires will accelerate the ageingprocess of the converter.

Modern engine management systems can detect amisfire using the various sensors. One indicator of amisfire is a high oxygen level in the exhaust gas, whichis detected by the oxygen sensor (section 3.5.7). Inaddition, the engine management ECU can check theacceleration rate of the crankshaft using informationprovided by the crankshaft speed/position sensor.When the power stroke occurs within each cylinder, itcauses an acceleration of the crankshaft, but a misfireby one cylinder will cause reduced accelerationcompared with the ‘good’ cylinders. The ECU can assesswhich cylinder is misfiring and cut off the fuel supplyto the affected cylinder. Without delivery of fuel, therewill be no unburned fuel passing through to thecatalytic converter, although unburned oxygen willremain.

Leaded fuel and other contaminantsLead compounds restrict the ability of human cells toabsorb oxygen. Leaded petrol has been banned in theEU since 2000. The amount of lead allowed in petrolwas previously subject to progressive reductions. Manyother countries have also banned leaded petrol.

With regard to leaded fuel and catalytic converters,lead compounds clog the pores of the active materials,reducing their efficiency. Excessive build-up of leadstops the conversion processes from taking place.Therefore, vehicles fitted with catalytic converters mustnot be operated with leaded fuel.

It is also possible for the active materials in catalyticconverters to be affected by other contaminants, such asdeposits from the engine oil that pass by the pistonrings into the combustion chambers. A worn engine,producing high emissions, can also contribute to therapid deterioration of its catalytic converter.

NOx accumulator convertersThe increased use of lean burn technologies (primarilystratified mixture formation with direct fuel injection),results in an excess of oxygen in the exhaust gases whenvery weak air/fuel mixtures are used. The excessoxygen prevents effective reduction of NOx by the three-way catalytic converter (see lean burn technologies insection 3.5.5). It is therefore necessary to findalternative methods of reducing the amount of NOx.

NOx accumulator catalytic converters are now beingfitted to many vehicles with lean burn engines. Theaccumulator type converter is not dissimilar to thestandard three-way converter in construction, but otheractive materials are included, such as oxides ofpotassium, calcium and barium. The converter canoperate in the same way as a three-way converter whenthe air:fuel ratio is around the stoichiometric value.

However, when the engine is operating with veryweak mixtures (i.e. with excess oxygen in the exhaustgas), the accumulator converter operates in a differentmanner. The oxidation process causes the nitrogen andNO to attract excess oxygen thus producing NO2.However, the additional active materials cause a furtherchange which results in nitrates being formed. Thenitrates are then stored (accumulated).

Accumulation and conversionAs described above, the accumulator converter storesNOx as nitrates, but at some stage the stored NOx mustbe released; otherwise the converter will eventuallybecome over-saturated (its maximum storage capacitywill be reached). As the amount of NOx stored in theaccumulator converter increases, it impairs the device’sability to store more NOx so a means of assessing whenthe stored NOx must be released is required. Twomethods can be used.

1 An NOx sensor downstream of the converterindicates to the ECU when higher levels of NOx are


Figure 3.64 Catalytic converter construction (three-way type)

1 Lambda oxygen sensor2 Swell matting3 Thermally insulated double shell4 Washcoat (Al2O3 substrate

coating) with noble metal coating5 Substrate (monolith)6 Housing

[647] Chapter 03 25/8/06 15:38 Page 137

flowing out of the converter because the converter isno longer storing the NOx in sufficient quantities asit is almost full to capacity. At this stage the NOx isreleased from the converter.

2 A mathematical process is used (ECUprogramming), which calculates the likely amountof NOx that is stored (based on operating conditionsand temperature). At the calculated time the processfor NOx release is implemented.

When the accumulator converter is assessed to be full,the engine is briefly operated using a rich mixture(typically with an air:fuel ratio of 12:1 or richer),resulting in an excess of CO, HC and H2 in the exhaustgas. In simple terms, the oxygen in the NOx is lost,leaving nitrogen. The oxygen combines with CO and HCto give CO2 and H2O (water).

The end of the conversion process can be assessedeither using another ECU calculation (again usingconditions and time), or with an oxygen sensor locatedafter the converter: the end of the process will be whenthe oxygen in the exhaust gas (after the converter) hasfallen to a predefined level.

An NOx accumulator converter can be used inconjunction with a three-way converter. Figure 3.65shows an arrangement where both converters are fittedinto an exhaust system. Note that the three-wayconverter is fitted close to the exhaust manifold (in thepre-cat location) with the NOx accumulator furtherdownstream. The NOx accumulator converter operatesat lower temperatures than the three-way converter(ideally around 300–400°C, compared with 400–800°Cfor the three-way type). This means that the twoconverters must be separate.

3.5.7 Oxygen/lambda sensing(controlling the air:fuel ratio)

Oxygen content: the critical factorIt has been highlighted a number of times within thischapter that the oxygen content of the exhaust gas iscritical to the operation of the catalytic converter (as

well as to other emission control devices). Achievingthe correct oxygen content very much depends on theair:fuel ratio, which should be stoichiometric (λ = 1).

Modern engine management systems are able toprovide very accurate control of the air:fuel ratio.However, to attain the highest possible accuracy, it isnecessary to monitor the oxygen levels in the exhaustgas.

An oxygen sensor (as shown in Figure 3.66) is usedto measure the oxygen content; this device thentransmits a signal to the ECU. If the oxygen content inthe exhaust gas is too high or low, the ECU is able tochange rapidly the fuelling as necessary, ensuring thatthe oxygen level is restored to the correct value. Whenthe oxygen level is correct, λ = 1. The sensor is oftenreferred to as a ‘lambda sensor’.

When the oxygen content is at λ = 1 (or within thelambda window described in 3.5.1), the catalyticconverter is able to convert CO and HC efficiently intoCO2 and also reduce NOx levels.


Figure 3.65 Location for an NOx accumulator converter and three-way converter

Figure 3.66 Oxygen/lambda sensor

Closed and open loop operationSection 3.5.3 provides details of the air:fuel ratios usedfor different engine operating and driving conditions.Older engines operated using a wide range of air:fuelratios often far beyond the lambda window. For lightload conditions an air:fuel ratio of 17:1 or weaker wasnot uncommon, and under full load 13:1 or richerwould have been used. However, such extreme ratiosare not suited to vehicles fitted with catalyticconverters, because they would dramatically reduce theconverter efficiency (which relies on the excess airfactor being λ = 1).

1 Engine with EGR system2 Lambda oxygen sensor

upstream of the catalyticconverter

3 Three-way catalytic converter(pre-cat)

4 Temperature sensor5 NOx accumulator type catalytic

converter (main cat)6 Two-step lambda oxygen sensor,

optionally available with integralNOx sensor

[647] Chapter 03 25/8/06 15:38 Page 138

For most engines since the early 1990s air:fuel ratiosare controlled as closely as possible within the lambdawindow (ideally at λ = 1). It is not always possible toavoid slight enrichment for full load conditions, and forcold running it is also often necessary to briefly providesome enrichment. However, light load conditions aregenerally achieved using air:fuel ratios within thelambda window. In general, every new or modifiedversion of an engine or an engine control system allowsincreased operation within the lambda window,enabling more efficient catalytic converter operationand a greater reduction in pollutants.

Engines with stratified charge mixture formationand direct fuel injection operate with exceptionallyweak mixtures under light load conditions; for furtherinformation, see section 3.4 (direct fuel injection) andalso lean burn technology in section 3.5.5.

Closed loopWhen an engine is operating under conditions wherethe mixture is at or close to λ = 1, the oxygen or lambdasensor monitors the oxygen content in the exhaust gas,and the ECU responds to the sensor signal as necessary.The process is referred to as ‘closed loop’ because thereis a continuous loop or repeat of the necessary actions,i.e. monitoring and correction, as shown below and inFigure 3.67.

Action 1 Oxygen/lambda sensor measures theoxygen.

Action 2 Signal sent by sensor to ECU.Action 3 ECU alters fuel quantity if necessary to

change the oxygen level. Action 1 Oxygen/lambda sensor measures the

oxygen.

In a closed loop operation, the ECU calculates therequired fuel quantity based on information from thevarious sensors (air mass, throttle position,temperature, etc.) but the oxygen/lambda sensorchecks the actual oxygen content and provides the

appropriate signal to the ECU. If the oxygen content isincorrect, the ECU makes a fine adjustment to the fuelsupply to enable the correct oxygen content to beachieved.

Open loopWhen engine operating conditions or driving conditionsdictate that the air:fuel ratio should be outside thelambda window (usually a rich mixture for full load orcold running conditions), the lambda sensor signal iseffectively ignored. The fuel quantity will thereforedepend entirely on the ECU calculations from othersensor information. The ECU will, however, revert backto closed loop operation as soon as the operating anddriving conditions dictate.

Lambda/oxygen sensor operation (step type)General principlesAlthough there are a few variants of lambda sensor, thegeneral principle of operation relies on comparing theoxygen content in the exhaust gas to the oxygen contentin the air. In effect the oxygen content in the air acts asa reference level against which the oxygen content inthe exhaust gas is compared.

In all types of lambda sensor in common use, anelectrical signal is produced by the sensor, dependingon the amount of oxygen within the exhaust gas. Thesignal voltage changes with changes in oxygen level. Asignal is therefore transmitted to the fuel injection orengine management ECU, which alters the fuel quantityas necessary, until the oxygen content of the exhaustgas is correct for efficient catalytic converter operation.

Step type sensor operationThe more commonly used early generation of lambdasensor is referred to as the step type or ‘narrow band’sensor. The name originates from the characteristics ofthe signal voltage produced by the sensor as explainedin the following paragraphs.

Figure 3.68 shows a schematic view of a simple steptype lambda sensor located in the exhaust pipe. In the


Figure 3.67 Closed loop operation Figure 3.68 Schematic view of a step type lambda sensor

[647] Chapter 03 25/8/06 15:38 Page 139

illustration, the sensor ceramic (item 1) is coated onboth sides with a layer of platinum that acts aselectrodes (items 2). The inner electrode is exposed tothe atmosphere, whilst the outer electrode is exposed tothe exhaust gas. The outer electrode has an additionallayer of porous ceramic (item 6) to protect it against theexhaust gases that would eventually erode the platinumlayer.

At higher temperatures, the main ceramic (item 1)allows oxygen ions through if there is a difference in theoxygen content either side of the ceramic. When thisoccurs it causes a small voltage to be produced acrossthe two platinum electrodes (shown as the small meterconnected across the electrodes in the illustration). Thevoltage produced depends on the difference in oxygencontent on either side of the ceramic. Because one sideof the ceramic is exposed to the atmosphere (with anoxygen content of around 20.8%) and the other to theexhaust gas (with an oxygen content of around 0.2% to0.3% in the region of λ = 1), there is a large differencein oxygen content either side of the ceramic. The largedifference causes ions to be conducted through theceramic, which in turn results in a small voltage beingproduced across the electrodes.

Figure 3.69 shows the voltage output from a steptype lambda sensor. An important point to note is thatwhen the air:fuel ratio is weaker than stoichiometric(λ > 1), the voltage produced is typically lower than100 mV (0.1 volts); in fact, the voltage may be as low as50 mV and increase only very slightly as the mixturebecomes a little richer. When the air:fuel ratio is richerthan stoichiometric (λ < 1), the voltage produced isaround 950 mV (0.95 volts). The voltage only decreasesslightly as the mixture becomes weaker. However, whenthe mixture changes to just slightly weaker or richerthan stoichiometric (just either side of λ = 1) thevoltage suddenly jumps from the weak mixture value ofapproximately 100 mV to the rich mixture value ofapproximately 950 mV.

In effect, the voltage produced by the sensorchanges little as a weak mixture becomes richer (i.e.lambda is still greater than approximately 1.03).However, when lambda reaches approximately 1.03,the voltage suddenly jumps in a large step to a lambdaof approximately 0.97. If the mixture continues tobecome richer (i.e. λ < 0.97), the voltage change is onceagain very small.

The large step in voltage that occurs between aboutλ = 0.97 and λ = 1.03 provides a very distinct signalthat enables the ECU to detect the change in oxygenwhen the air:fuel ratio alters slightly away from λ = 1.Conveniently, the lambda values of 0.97 and 1.03effectively define the lambda window (this is covered insection 3.5.1 and shown in Figure 3.57).

The signal voltage shown in Figure 3.69 is obtainedat a temperature of approximately 600°C. The voltageproduced also varies with temperature, so it is essentialthat the ECU responds only to the sensor signal whenthe sensor is at a defined operating temperature. The

sensor starts to function reliably when the temperaturereaches approximately 300–350°C. An idealtemperature is around 600°C. Because the sensorresponse to changes in oxygen content is very slow atlower temperatures, the ECU is able to ignore theseslow responses and operate in ‘open loop’ mode. Whenthe sensor is at the required temperature and theresponse time is quicker, the ECU then operates in‘closed loop’ mode. Response times to changes inoxygen level are around 50 ms for a fully hot sensor,whilst for a cold sensor the response can take muchlonger than 1 second, which is not suitable foroperating in closed loop because of the time delay.

The standard step sensor relies on heat from theexhaust gas to reach operating temperature.Immediately after starting and during warm-up fromcold the system will operate in open loop until thesensor is at operating temperature. The sensor shouldideally be located close to the exhaust manifold toobtain as much heat as possible after starting; however,because of the potential for overheating and prematuredeterioration, sensors are generally located somedistance from the manifold.

Heated step type lambda sensorTo overcome the problems of temperature variation,which can change the accuracy and reliability of thelambda sensor signal, heated sensors were developed.There are two main types of heated sensor (step type),as shown in Figure 3.70.

Figure 3.70a shows a direct development of the stepsensor described above. This type has an electricalheater located in the sensing element of the sensorassembly. When the engine is started and is running,battery power is supplied via a relay to the heater. Theheater rapidly raises the temperature of the sensor sothat closed loop operation can start as little as 20 or30 seconds after a cold start.


Figure 3.69 Voltage output from a step type lambda sensor

[647] Chapter 03 25/8/06 15:38 Page 140

Figure 3.70b shows a later type of heated step typelambda sensor. Although this type operates in a similarmanner to the first type, the sensing element and heaterassembly are layered. The illustration shows thislayered construction.

Limitations of step type sensorsStep type sensors (heated and unheated) provide asignal change that can only be effectively detected bythe ECU when the air:fuel ratio alters at the rich orweak extremes of the lambda window. The sensor canonly clearly indicate when the lambda factor is at thesepoints, i.e. when there is a stepped change in thevoltage (as illustrated in Figure 3.69). These easilyidentifiable stepped changes are easily detected by theECU, but it is not easy to measure the exact signalcorresponding to λ = l (which occurs at around450 mV) because the voltage rise or fall is too rapidwhen the value is close to λ = l.

Whilst the step type sensor signal is very effectivefor vehicles fitted with three-way catalytic converters,this narrow band of operation does not allow thesensors to be effective when engines operate with awider range of air:fuel ratios. This is especially relevantto the modern generation of direct injection enginesthat operate with a stratified charge and air:fuel ratiosthat may be as weak as 30:1 or 40:1. A sensor istherefore required that can measure oxygen levels overa broader range of air:fuel ratios; such sensors arecovered later in this section.

Operating signals for step type lambdasensorsIt was highlighted earlier in this section that asignificant change in voltage is produced by the sensorby changes around λ = 1. The approximate lambdavalue range that causes the jump in voltage is λ = 0.97to λ = 1.03 (see Figure 3.69).

If we assume that the engine and lambda sensor areat full operating temperatures, the ECU will provide theappropriate fuelling that should in theory result in anair:fuel ratio of λ = 1. In reality, minor variations willalways occur; and the lambda sensor signal providesthe reference to the ECU regarding such deviations.

The measuring and correction process then passesthrough the following phases (this is closed loopoperation, also referred to as ‘feedback control’).

1 If we assume that a very slightly rich mixture exists,the lambda factor will be slightly low (λ < 1). Thesensor signal voltage will therefore rise toapproximately 950 mV, and the ECU will reduce thefuel quantity slightly, which will cause a weakeningof the mixture and a reduction in the sensor voltage.

2 The sensor will now detect the weakening of themixture, which will produce a slightly increasedoxygen level. The lambda factor will increase(λ > 1), which will cause the sensor voltage to fallto approximately 100 mV. The ECU will detect thereduced voltage and again alter the fuel quantity,this time to provide a slight enrichment.


Figure 3.70 Heated lambda sensorsa An example of a heated lambda sensorb Alternative type of sensor element with combined heater

[647] Chapter 03 25/8/06 15:38 Page 141

3 The process will now start again, resulting in acontinuous increase and decrease (oscillation) inthe sensor voltage, as shown in Figure 3.71.

The frequency of the oscillations of the lambda sensorsignal depends on the speed of response for the sensor,i.e. how quickly the sensor registers the change inoxygen, which can be as rapidly as within 50 ms.However, the fuelling change leads to an inevitabledelay in the alteration of the oxygen content, as thischange passes through the exhaust system to the sensor.It is common to find that the frequency of change forthe sensor signal is around 1–3 Hz, depending onengine speed, temperature and other factors.

The sensor signal in Figure 3.71 shows a veryconsistent and regular oscillation which is seldom seenin practice owing to minor variations that occur in theoxygen content of the exhaust gas sample. The hashthat exists around the signal is typical of a lambdasensor signal.

Item 7 in the illustration is effectively a step type sensorelement or cell; however there is an additional cell,referred to as the ‘oxygen pump cell’ (item 8). Anextremely small diffusion gap (item 6) separates thepump cell from the step type sensing cell. There ishowever a barrier (a porous diffusion barrier, item 11)through which the oxygen from the exhaust gas mustflow before it reaches the diffusion gap. Once in thediffusion gap the oxygen affects one electrode on thestep type sensing cell. The other electrode of the steptype cell is exposed to the reference air (oxygen in theatmosphere), just as with a standard type step sensor.The step type cell is measuring the oxygen content inthe diffusion gap, which would be lower than theoxygen content in the air, thus causing a voltage to beproduced by the step type cell; this voltage istransmitted to the control unit.

However, exhaust gas must flow through the pumpcell before it can reach the diffusion gap. The pump cellalso has the capacity to pump oxygen through thediffusion barrier in either direction, so it can increase ordecrease the oxygen content in the diffusion gap. Thepumping action is created by providing a smallcontrolled voltage to the pump cell’s platinumelectrodes; the way the current flows dictates whetheroxygen is pumped in or out of the diffusion gap. Theobjective is to use the pump cell to maintain the oxygencontent in the diffusion gap at a value of λ = 1. If thediffusion gap contains too much or too little oxygen(weaker or richer than λ = 1) because of high or lowoxygen levels in the exhaust gas, the pump cell thendecreases or increases the oxygen in the diffusion gap toachieve λ = 1.

The process starts with the step type cell measuringthe oxygen content in the diffusion gap, creating avoltage signal that is assessed by the electronic controlunit. Note that the initial oxygen level and oxygen levelfor normal operation will depend on the oxygen contentin the exhaust gas, which will have flowed into thediffusion gap. If the oxygen level in the diffusion gap istoo high or too low (higher or lower than λ = 1), theelectronic control applies the appropriate current at thepump cell electrodes to pump oxygen in or out of thediffusion gap as required to achieve λ = 1. The level ofcurrent required is an indicator of the oxygen beingpumped in or out of the diffusion gap, and is thereforealso an indication of the oxygen content of the exhaustgas.

If the oxygen content in the diffusion gap is at λ = 1,there is no requirement for pumping oxygen, and thecurrent will be zero. However, if the oxygen content ishigh (λ > 1) a negative current is used to pump oxygenout of the diffusion gap; if the oxygen content is low(λ < 1) a positive current is used to pump oxygen intothe diffusion gap. The electronic control unit produces asensor signal (voltage change) that is dependent on thecurrent level required to maintain the oxygen content inthe diffusion chamber at λ = 1.


Figure 3.71 Continuous signal produced by a step type lambdasensor

Broadband sensorAs mentioned earlier, the broadband sensor can be usedto measure the oxygen content over a very broad rangeof air:fuel ratios, i.e. across a wide range of lambdavalues from approximately 0.7 to 4. The broadbandsensor is therefore suitable for lean burn engines, suchas those using a stratified mixture formation (typicallyon modern direct injection engines). In addition, thebroadband sensor can be used on diesel engines andengines operating on gaseous fuels.

As well as offering a capability of measuring over awider lambda range, the broadband sensor provides aprogressive voltage change across the range ofoperation, as opposed to the stepped voltage jump thatoccurs with a narrow band step sensor.

Operating principlesFigure 3.72 shows the construction of the sensingelements in a broadband sensor. The construction issimilar to the later type step sensors, with the differentmaterials arranged in layers.

[647] Chapter 03 25/8/06 15:38 Page 142

Sensor heatingWithin the sensing unit assembly, a heater element isconstructed to heat the sensor rapidly and maintain arelatively constant temperature that is not greatlyinfluenced by exhaust gas temperature. The functionand operation of the heater are the same as describedfor the step type sensor. Broadband sensors generallyoperate in the range 650–900°C.

Sensor output signalSensors produced by different manufacturers mightprovide slightly different output signals to be monitoredby the engine management ECU. An example of abroadband sensor signal is shown in Figure 3.73. Notethat the voltage change is progressive across a widelambda range; the illustration shows the voltage change

within the range of λ = 0.7–1.5, although higherlambda values are measured and provide slightly highervoltage values.

Other types of lambda sensorSome lambda sensors rely on a special semiconductordevice that responds to changes in oxygen content byaltering its resistance. When a voltage is applied acrossthis device, the resistance change affects the current inthe circuit (a series resistance circuit). This change is anindicator of oxygen content. Thus it is possible to applya 5 volt reference voltage to the sensor, obtaining asignal voltage that ranges from 1 to 4 volts, dependingon the oxygen content in the exhaust gas.

3.5.8 Exhaust gas recirculationA method of reducing NOxNOx formation Within section 3.5, and in other sections of Chapter 3,many references were made to oxides of nitrogen(NOx). These are regarded as a pollutant and are formedby the combination of oxygen molecules and nitrogenmolecules (both of which exist naturally in the air.

Whilst NOx formation increases when the air:fuelratio is very slightly weaker than the stoichiometricvalue (i.e. λ = 1.05–1.1; see Figure 3.61), it is also truethat NOx formation increases significantly whencombustion temperatures rise. NOx formation occursbecause of the heat of the combustion process, whichenables a chemical change to take place. For NOx this


Figure 3.72 Broadband sensor (sensing element construction)

Figure 3.73 Output signal voltage for one type of broadbandsensor

[647] Chapter 03 25/8/06 15:38 Page 143

can start when combustion temperatures are around1300°C, but the formation of NOx increases at around1800°C and accelerates if temperatures exceed around2500°C.

When an engine is operating under load conditions,especially when the air:fuel ratio is close toλ = 1.05–1.1 (which can exist during part loadconditions), combustion temperatures will rise.However, high load conditions, where a full charge ofair enters the cylinder (with the throttle fully open),will inevitably cause higher combustion temperatures,but under high load conditions the mixture is usuallyslightly richer, which helps to reduce NOx formation.

When an engine is operating under part or mediumload conditions with an air:fuel ratio slightly weakerthan λ = 1, combustion temperatures are high and theoxygen content of the exhaust gas enables high levels ofNOx to be formed. Although catalytic converters can beused to reduce NOx levels after it has been formedduring combustion, it is often also necessary to useother means to reduce the production of NOx during thecombustion process. In general, these devices orprocesses are designed to reduce combustiontemperatures.

Using the exhaust gasA very effective and well established method ofreducing combustion temperatures is to pass apercentage of the exhaust gas back into the intakesystem where the exhaust gas mixes with the newcharge of air entering the cylinder.

Exhaust gas is made up of a high percentage of inertgases, such as water vapour (H2O) and carbon dioxide(CO2). An inert gas does not combust, so when thesegases mix with the air flowing into the cylinder, theycause a lowering of the combustion temperature and areduction in the formation of NOx.

A widely used method of enabling exhaust gas to mixwith the fresh intake air is to recirculate some of theexhaust gas back into the intake manifold (Figure 3.74).This process is referred to as exhaust gas recirculation(EGR).

Note that a similar but less effective result can beachieved when the valve timing is arranged so thatthere is valve overlap, i.e. the intake valve opens beforethe exhaust valve closes (at the end of the exhauststroke). The result is that some exhaust gas mixes withthe incoming fresh charge of air. The amount of valveoverlap on most engines was fixed, until recently.However, variable valve timing mechanisms are nowused for many engines so that different overlap periodscan be used at different engine speeds. In reality,variable valve timing systems are used to enable goodpower or torque to be achieved over the whole enginespeed range, but the added benefit of valve overlap is tofacilitate the mixing of some exhaust gas with the freshcharge of air, thus reducing combustion temperaturesand NOx.

Controlling the quantity of recirculated exhaustgasFor those engines operating with homogeneous mixtureformation (all of the air drawn into the engine is mixedwith fuel), it is possible to introduce around 10–15% ofthe exhaust gas back into the fresh charge of air that isdrawn into the cylinder. This percentage does notdramatically affect fuel consumption and power, but isusually sufficient to reduce the NOx by a significantamount. Slightly higher levels of exhaust gasrecirculation can be used on engines operating with astratified mixture formation (direct injection engines).

Importantly, the amount of exhaust passing throughto the intake system will depend on the exhaust gaspressure and the intake manifold pressure (or


Figure 3.74 Exhaust gas recirculation system (EGR)

[647] Chapter 03 25/8/06 15:38 Page 144

depression). If the exhaust gas pressure is high, such aswhen the engine is under high load (when the throttleis fully open, high volumes of air are drawn into thecylinder which results in high volumes of exhaust gas,increasing the pressure in the exhaust system), a higherflow of exhaust gas into the intake system will result. Itis therefore necessary to control the flow of exhaust gasinto the intake system to ensure that excessivequantities of exhaust gas do not enter the intake systemand combustion chamber, which would cause poorcombustion.

In general, the exhaust gas is recirculated duringpart/medium load conditions but not when the engineis at idle or at full throttle (when a richer mixture mightbe used which would cause a reduction in NOx).

EGR valveAn EGR valve (Figure 3.75) is used to control the flowof recirculated gas. Figure 3.74 shows an EGR valvelocated in the pipe that feeds the exhaust gas to theintake system. In this example, the valve is directlyopened and closed by a solenoid which is controlled bythe ECU. The ECU, which is receiving information fromvarious sensors, (e.g. engine speed, throttle position,temperature, etc.) opens the valve the appropriateamount to allow the required amount of exhaust gas tobe recirculated to suit engine operating conditions.

The EGR solenoid may be supplied with a digitalcontrol signal that allows the gas valve to be accuratelypositioned. The control signal duty cycle is altered toprovide an increased or decreased current/voltage tothe solenoid, which causes the valve to open or close toa greater or lesser extent (see section 1.9).

Different types of EGR control systemThe simplest type of system to control the flow of gasesin an EGR system is shown in Figure 3.74. This relies ona single valve, directly controlled by the ECU. However,other systems have been used, requiring additionalvalves and sensors, that ensure that recirculatedexhaust gas is passed to the intake system.

Figure 3.76 shows a system where the main EGRvalve is opened and closed by ‘vacuum’ (low pressure ordepression) which is taken from the intake system. Thevacuum level applied to the EGR valve is in turnregulated by an ECU controlled vacuum valve. On thistype of system, it is the vacuum control valve thatreceives the digital control signal from the ECU.


Figure 3.75 EGR valve with integrated solenoid

Figure 3.76 EGR system with vacuumvalve control

[647] Chapter 03 25/8/06 15:38 Page 145

In some systems, the EGR valve is fitted with a sensorthat tells the ECU the amount the valve is open. Othersystems use a pressure sensor to detect the pressure inthe exhaust recirculation pipe. With these sensors, theECU is able to assess the amount of exhaust gas flowingin the EGR system.

3.5.9 Secondary air injectionPrevious parts of section 3.5 have highlighted theproblems of high emission levels during cold running.Even modern engines require some enrichment duringthe early phases of the warm-up period. Enrichment isrequired to overcome condensing of the fuel on cylinderwalls and intake system walls, and to ensure thatsufficient fuel is able to vaporise in what is initially arelatively cold environment.

However, a rich mixture will have insufficientoxygen (λ of around 0.9 or less) which will result inhigh levels of CO and HC. In many cases, these CO andHC levels exceed what is permitted and must thereforebe reduced. Because the levels are so high, there is toomuch CO and HC, and insufficient oxygen for thecatalytic converter to change the CO and HC into CO2.Furthermore, in the early phases of engine start-up, thecatalytic converter is not at working temperature.

It is possible to inject air into the exhaust ports orexhaust manifold, which enables oxygen to combinewith the CO and HC, owing to the temperature of theexhaust gas. In effect, a combustion or oxidationprocess occurs in the exhaust manifold. The secondaryair injection process is used only for short periods afterthe engine is started from cold, but is sufficient toreduce the CO and HC levels during this period. Inaddition, the combustion of gases in the exhaustmanifold or ports adds additional heat to the exhaustgas, which assists in quickly raising the temperature ofthe catalytic converter, so it is able to function sooner.

Air pump injection Figure 3.77 shows a layout for an air pump based airinjection system. The electrically driven air pump drawsair from the atmosphere (via the air filter) and pumps itinto the exhaust manifold through a control valvewhich regulates the amount of air depending onoperating conditions. The control valve is connected tothe intake manifold (position A on the diagram) andintake vacuum (depression) can therefore be passed tothe secondary air valve via a control valve. The vacuumpassing to the secondary air valve is regulated by acontrol valve, which is controlled by the ECU.

When the engine is able to operate without mixtureenrichment and the catalytic converter is at workingtemperature, the air pump is switched off.

Pulse air sysdemsOperating in much the same way as air pump injectionsystems, the pulse air system relies on pressure pulses inthe exhaust pipe to draw in air, rather than using apump (Figure 3.78). The exhaust system is subject topositive and negative pressure pulses when the engineis running. Positive pulses occur when an exhaust valveopens; negative pulses occur when the valve closes butexhaust gases continue to move through the exhaustsystem.

A valve (one-way valve in the diagram) is used to letthe air into the exhaust manifold. It opens when thepressure pulse in the exhaust system is low, and closeswhen pulse pressure is high. The pulse pressure in eachexhaust port changes when the exhaust valve opens andcloses, so the valve opens and closes continuously andrapidly during the warm period.

Vacuum from the intake manifold is fed to the pulseair solenoid valve, which is controlled by the ECU.When the ECU control signal opens the valve, vacuum ispassed to the ‘vacuum controlled valve’ which in turnalso opens, thus allowing air to flow to the one-wayvalves and the exhaust manifold. The ECU can control


Figure 3.77 Secondary air injection system (with air pump)

[647] Chapter 03 25/8/06 15:38 Page 146

the pulse air solenoid valve so that the vacuum affectingthe control valve is regulated, thus enabling regulationof the air flowing into the exhaust manifold.

3.5.10 Evaporative emission controlEvaporative emission control systems (EVAP) are usedto prevent vapour from the fuel tank and fuel systemfrom escaping into the atmosphere. Figure 3.79 shows abasic EVAP system.

A charcoal (carbon) canister is connected to the fueltank. The canister collects the fuel vapour which isstored by the charcoal. When the engine is running,vacuum (depression) in the intake system draws thevapour from the canister and into the engine where itmixes with the intake air. A ‘canister purge valve’ iscontrolled by the ECU using a digital control signal(section 1.9), to ensure that the flow of vapour to theintake system is regulated. This allows a controlledamount of vapour to enter the intake system fordifferent operating conditions. Typically, the vapour willbe drawn from the canister during light/medium loadengine conditions.

When the vapour mixes with the intake air, therewill be a slight enrichment of the air/fuel mixture. TheECU will therefore adjust the injected fuel quantity tocompensate. However, because the process of drawingvapour from the canister usually occurs only when thelambda control system is operating in a ‘closed loop’,the lambda sensor transmits an appropriate signal tothe ECU if the mixture is too rich, thus enabling the ECUto make any corrections.


Figure 3.78 Pulse air injection system

Figure 3.79 Evaporative emission control system (EVAP)

EGR reduces cylinder combustion temperature,which in turn reduces NOx emission

The main way that emissions are reduced is bymaintaining accurate ignition and mixture strength

Evaporative emissions are reduced by storing themin charcoal canisters and then burning them in theengine

Key

Poin

ts

[647] Chapter 03 25/8/06 15:38 Page 147

3.6.1 Combining the various systemsThe requirement for integrated controlEach of the individual engine systems so far coveredwithin Chapter 3 (ignition, injection and emission)have been subject to continuous development andimprovement. In isolation, there have beenimprovements in engine power, fuel consumption andlowering emission levels. However, the control of eachsystem needs to be integrated with the control of theother systems to achieve the optimum results.

An example of integrated or combined control is theneed sometimes to alter ignition timing and fuelquantity at exactly the same time to suit a change inengine operating conditions. To achieve the best results,communication between the ignition ECU and the fuelsystem ECU is essential. Another example is the need toalter timing and fuelling when certain emission controlfunctions are implemented: again, communicationbetween the different engine systems is essential toachieve optimum system and engine efficiencies.


3.6 ENGINE MANAGEMENT (THE CONCLUSION)

Figure 3.80 Bosch ME Motronic engine management system

Key

1 Activated charcoal canister2 Hot film air mass sensor with integrated temperature

sensor3 Throttle device (electronic throttle control)4 Regeneration valve5 Intake manifold pressure sensor6 Fuel rail7 Fuel injector8 Actuators and sensors for variable valve timing9 Ignition coil and spark plug

10 Camshaft phase sensor11 Lambda sensor upstream of primary catalytic converter12 Engine ECU13 Exhaust gas recirculation valve14 Speed sensor

15 Knock sensor16 Engine temperature sensor17 Primary catalytic converter (three-way catalytic converter)18 Lambda sensor downstream of primary catalytic converter19 CAN interface20 Fault indicator lamp21 Diagnostics interface22 Interface with immobiliser ECU23 Accelerator pedal module with pedal travel sensor24 Fuel tank25 In-tank unit comprising electric fuel pump, fuel filter and

fuel pressure regulator26 Main catalytic converter (three-way)

The on-board diagnostics system configuration illustrated bythe diagram reflects the requirements of EOBD

[647] Chapter 03 25/8/06 15:38 Page 148

On early generations of engine management system,ignition and fuel injection ECUs were separate, butthey communicated so that control functions wereharmonised. In addition, it was possible to share theinformation from the various sensors. Onetemperature sensor could be used to transmit a signalto one of the ECUs, which in turn would transmit asignal to another ECU. This approach reduced theneed to duplicate sensors and wiring, which inevitablyreduced cost.

Single ECUDuring the late 1970s, some engine managementsystems were produced with a single ECU. This type ofsystem became much more widely produced during the1980s. Currently, nearly all light vehicle petrol enginesare fitted with a single ECU that controls most of thefunctions. In effect, all of the engine related systems aremanaged by a single ECU.

The next stage of integrated control systems wouldbe to control all vehicle functions from a single ECU: forexample, the engine and chassis systems (includingfeatures such as anti-lock brakes), could all becontrolled from a single ECU. Although some attemptshave been made to produce a single ‘vehicle’ controlunit, the trend is to use high speed communicationnetworks that enable all of the vehicle system ECUs tocommunicate and share information.

3.6.2 Modern engine managementsystems

Apart from the diagnostic functions, almost all of theindividual systems that make up a modern enginemanagement system have been covered in the previoussections of Chapter 3. The following two examples ofmodern engine management systems illustrate the wayin which all of the previously discussed systems areintegrated using a single ECU.

Figure 3.80 shows a Bosch ME Motronic enginemanagement system which uses port type fuel injectionalong with direct ignition and various emission controldevices.

Figure 3.81 shows a modern direct injection enginemanagement system (Bosch MED Motronic). Thissystem is similar to the example shown in Figure 3.80,but note the differences in the fuel injection andemission control systems.

Both systems feature European on-board diagnostics(EOBD) connections which are covered in section 3.7.

Full engine management systems combine ignitionand fuel control as well as controlling many otheraspects such as EGR

Most modern engine management ECUs arelinked by CAN to other systems such astransmission management

Key

Poin

ts

Engine management (the conclusion) 149

Figure 3.81 Bosch MED (direct fuel injection) Motronic enginemanagement system

[647] Chapter 03 25/8/06 15:38 Page 149

3.7.1 Self-diagnosis of system faultsReasons for self-diagnosis (on-board diagnostics)Since the mid-1980s, an increasing number of Europeanmarket vehicles have been equipped with some form ofself-diagnosis. Self-diagnosis systems were in useearlier, but mainly on American market products.

Self-diagnosis is the capacity of a vehicle system (ormore precisely the system ECU) to detect systemoperating faults, and provide an indication that aspecific fault exists.

Self-diagnosis was initially fitted to fuel and ignitionsystems, but it has become increasingly fitted to mostvehicle systems with electronic control. Self-diagnosis isnow found on vehicle systems such as ABS (and othervehicle stability control systems), automatic gearboxes,air conditioning, airbags and other safety systems. Self-diagnosis is a function of the system computer or ECU,which monitors and assesses all the input signals (fromsensors) as well as providing the output control signals.The system ECU is able to distinguish correct fromincorrect signals (and operation) for much of thesystem that is being controlled.

Self-diagnosis was introduced to overcome anumber of potential problems.

1 One aim of self-diagnosis is to overcome theproblems of working and diagnosing faultsassociated with ‘new technology’. This is especiallyimportant as each new vehicle system is introduced.Repair technicians need to be able to identify faultsrelatively quickly and easily.

2 Fast and accurate diagnosis should help to reducewarranty costs, while retaining customer loyalty andsatisfaction.

3 The self-diagnosis system can be used during vehicleassembly to ensure that each vehicle leaves theassembly plant without detectable faults.

4 Importantly, modern engine management systemsand emission control systems are fitted to vehicles toensure that emission levels remain low and withinlegislated limits. Many faults that would result inunacceptable emissions can remain undetected bythe driver. With a self-diagnostic system, it ispossible to provide a warning to the driver or theworkshop technician, along with an indication ofwhat the fault is.

5 Modern vehicle systems rely on complex softwareprograms within the ECU as well as electrical andelectronic systems. In a high percentage of cases,technicians are not familiar with these ‘new’technologies. Self-diagnostic systems are designedto support technicians on unfamiliar aspects of anew motor vehicle.

Self-diagnostic facilities should help to reduce thelength of time that faults remain undetected.

Self-diagnosis systems form only a small part of thewhole facility. The next section highlights otherfunctions that can be implemented by the systemcomputer to further assist in reducing emissions whenfaults occur, and to assist a technician to test or checkvehicle system operation. The self-diagnosis facility isoften referred to as OBD (on-board diagnostics).

Vehicle manufacturers appoint authorised dealers orrepair workshops to sell and maintain vehicles. Theseusually have very sophisticated, dedicated equipmentcontaining code reading or scan tools as part of theoverall equipment package. However, independentrepair workshops usually purchase general purposecode reading equipment, designed to function on manymakes and models of vehicle. The general purposeequipment is usually designed so that cartridges or podscan be inserted into the code reader; the differentcartridges contain different software programs thatenable the code reader to communicate and operatewith different vehicles.

Fault codes, blink codes and fault relatedmessagesOnce the ECU has recognised that a fault exists on avehicle system, it is able to implement other functions.

The following list covers the common functionsperformed by the ECU when a fault is recognised.

1 The ECU can illuminate a dashboard warning lightto indicate to the driver that a fault exists.

2 A system failure, such as a sensor fault, could causethe engine to run poorly and generate highemissions. Under these conditions, the ECU mightbe able to control the engine management system sothat it operates in a ‘limp home’ or ‘fail safe’ mode.When the ECU implements limp home operation, itnormally substitutes a preset value for the failedsensor, which should ensure reasonable engineoperation.

3 The ECU can provide some form of coded output(fault code) which is accessed or retrieved by a ‘faultcode reader’ or through other means such as a ‘blinkcode’ transmitted via the dashboard warning light orvia an LED based test tool. A fault code usuallyconsists of a number (or series of numbers),although some systems provide a series of letters oreven a short message. As explained later in thissection, some standardisation has been adoptedacross all vehicles for fault code systems, wherebyspecific faults are allocated a dedicated codenumber. This standardisation generally only appliesto emissions related faults, but inevitably embracesthe engine management system.


3.7 ENGINE SYSTEM SELF-DIAGNOSIS (ON-BOARD DIAGNOSTICS) AND EOBD

[647] Chapter 03 25/8/06 15:38 Page 150

Methods of retrieving fault codesAs mentioned above, fault codes can be retrieved fromthe vehicle ECU using a fault code reader which, whenconnected to the ECU, displays a code number relatedto the fault.

Many systems, however (especially older systemsand non-engine management systems), provide what isoften referred to as a ‘blink code’. A blink code can beoutput by the ECU to the dashboard warning light, andthe number of flashes on the warning light correspondsto the relevant code number. Some systems use an LEDlight (often located on the ECU) or codes might beaccessed by connecting a separate LED tester (or a voltor dwell meter). When a fault code number has beenretrieved, the technician refers to the appropriateinformation source to look up the code number andestablish the nature of the fault. Most code readersindicate the messages within the code reader software.

To access a blink code, it is normally necessary toperform a set procedure which effectively ‘instructs’ theECU to output the code. The procedure may involvelinking two terminals on a connector plug or othersimilar procedures which will then cause the ECU tooutput the code.

When a code reader is connected to the appropriateconnector plug, it communicates with the vehicle ECU,and by following the instructions supplied with thecode reader (or the instructions on the code readerscreen display), the technician can make the codereader transmit an instruction to the ECU which causesthe ECU to output the codes. The instruction from thecode reader may take the form of a password; when the

instruction or password is sent to the ECU, the ECUprovides the coded output, thus allowing the codereader to display the code number.

Although some inexpensive code readers effectivelyact as a means of accessing the simple blink code (i.e.they simply provide a code number), most elaboratereaders also display a message which details the natureof the fault associated with the displayed code.

When a blink code is read via a warning light orLED, it is normal practice to count the number of timesthe light flashes. When more than one code exists,different systems will use slightly different methods ofseparating the different codes. There are differentmethods of displaying a code number. For instance,code 12 could simply be displayed by 12 flashes, butalternatively could be displayed by providing one flash,a pause, then two flashes.

The example in Figure 3.82 shows how two codesmight be displayed (i.e. code 12 and then code 23). Abrief pause is provided between the ‘tens and units’ as isthe case between the 1 and the 2 of code 12, andbetween the 2 and the 1 of code 23. However, a longerpause (in this case 2 seconds) is provided between thetwo different codes, as is the case between codes for 12and 23.

Some inexpensive code readers simply provide aflashing LED; the flashes are counted in the same wayas a normal blink code. Most code readers provide adisplay screen that indicates the code as a number. Acode reader might provide a relevant message inaddition to the code number.

Engine system self-diagnosis (on-board diagnostics) and EOBD 151

Two-digit code One-digit code

Normal

T or TE1 terminal ON

Abnormal(Trouble codes 12 and 23 shown)

4.0 or 4.5 s 1.5 s 2.5 s 1.5 s

0.5 2 s 0.5 2 s


Abnormal(Trouble codes 2 and 3 shown)

2.5 s

0.5 2 s 0.5 2 s

4.0 or 4.5 s


Normal0.5 2 s

4.5 s 4.5 s


12 23 2 3

Figure 3.82 Example of a blink code (code 12 followed by code 21)

[647] Chapter 03 25/8/06 15:38 Page 151

3.7.2 How the ECU recognises andovercomes system faultsExpected operating values Virtually all the circuits connected to the systemcomputer (ECU) carry a voltage or signal that should liewithin certain operating values. These values areprogrammed into the ECU, which can compare actualsignal values with expected or programmed values.

For example, the power supply for the ECU might beexpected to fall between lower and upper limits of9 volts and 15 volts. If the voltage is above or below theexpected limits, this would be recorded as a fault by theECU – the starting point for self-diagnosis.

Many sensors on computer controlled vehiclesystems operate by providing a voltage or signal voltagethat should normally lie within certain easily definedlimits. The ECU can thus recognise that a fault exists,assuming that the fault causes the signal voltage to falloutside the expected limits or tolerance.

Example of fault detection using a simpletemperature sensor circuitOne example of a sensor circuit that should usuallyprovide a voltage within defined limits is the enginecoolant temperature sensor. It functions by changing itsresistance as the coolant temperature changes. Thesensor forms part of a series resistance circuit, whichmeans that when the resistance of the sensor changes, itaffects the voltage in the circuit (Figure 3.83).

The ECU provides a reference voltage to thetemperature sensor circuit which, on modern systems, isnormally 5 volts. When the circuit is complete (sensorplug connected) the voltage on the section of the circuitbetween the ECU and the sensor is reduced (by theaction of the resistances). The exact voltage depends onthe value of the resistances and, although the ECUresistance remains the same, the sensor resistancechanges, thus affecting the voltage in the circuit.

On nearly all temperature sensor circuits, a typicaloperating value is around 3.0–3.5 volts when thecoolant temperature is low (a cold engine). The voltagethen falls to around 0.3–0.7 volts when the engine ishot. It is possible for the voltage to reach higher or

lower values, but this would mean an extremely cold orhot engine.

Assuming that the sensor and associated wiring arein good condition, and that extreme cold and hottemperatures are reached, it is possible (although veryunlikely) for the voltage in the circuit to reach as high as4.5 volts or as low as 0.2 volts. These values can beused by the ECU as maximum and minimum values,and the only likely situation that would cause thevoltage to lie outside this range would be a fault in thewiring, sensor or ECU itself. In reality, a tolerance mustbe allowed slightly outside of the expected maximumand minimum values: in our example we will use0.1 volts as the minimum and 4.8 volts as themaximum. Figure 3.84 shows typical values for normaloperation and for detecting a fault.

There are two main faults that the ECU will easilyidentify: a break in the circuit and a short in the circuit.Figures 3.85 and 3.86 show a circuit break (opencircuit) and a short circuit. In both cases the fault isshown as a wiring problem. However, any part of thecircuit, including the sensor resistance itself, couldsuffer a short circuit or a circuit break, which wouldprovide the same results.


Voltage signalwater temperature

5V/12Vstabilisedsupply

ECU

Fixedresistor

ECUearth

Sensorearth

Water temperature sensor

NTCresistor

Figure 3.83 Schematic layout of temperature sensor circuit

ECU

Voltage dropsto zero indicatingshort to ECU

ECUearth

Short circuitbypassessensor

Figure 3.84 Example of temperature sensor circuit operatingvoltages and voltages that would be regarded as a fault condition

Condition Sensor signal voltage

Extremely hot minimum value may (engine running at very high be as low as 0.2 voltstemperature but this condition must be allowed for)Extremely cold maximum value may (engine running at very low be as high as 4.5 voltstemperature but this condition must be allowed for)Fault condition less than 0.1 volt(ECU detects voltage values orthat are outside normal greater than 4.8 voltsexpected values)

Figure 3.85 Temperature sensor circuit with a short to earth

[647] Chapter 03 25/8/06 15:38 Page 152

Short circuitWhen there is a short circuit (Figure 3.85) that results inthe signal wire shorting through to earth, the 5 voltreference supplied by the ECU is connected directly toearth. The current has already flowed through theresistance in the ECU, so, when the circuit is connectedto earth, the voltage falls to zero.

The circuit is now the same as a light circuit inwhich a voltage is applied to a bulb and then from thebulb to earth. The voltage on the earth side of the bulbis zero. The resistance in the ECU acts in the same wayas the light bulb.

The ECU monitors the voltage in the signal circuit,which would normally sit within acceptable operatinglimits of 0.2–4.5 volts. Because the voltage detected isnow zero, the ECU recognises this as a fault. The ECU isable to allocate a fault code to this particular fault.

Note that the ECU might provide a fault code for afault on the sensor circuit. However, many systems willprovide a code that states: ‘the voltage level in thetemperature sensor circuit is too LOW’.

The ECU cannot determine the exact nature of thefault; it can only establish that a fault exists causing alow voltage. Therefore the technician must still carryout a detailed investigation of the circuit.

Open or broken circuitA broken or open circuit (Figure 3.86) prevents the flowof current through the circuit. Without a current flow, aresistance does not have any effect on the voltage levelin the circuit. Therefore, the 5 volts, applied as areference voltage by the ECU, remains at 5 volts.

The ECU is again monitoring the voltage at point A(this is the voltage in the signal circuit which wouldnormally be within the acceptable operating limits of0.2–4.5 volts). Because 5 volts is now detected at pointA, the ECU recognises this as a fault, and allocates anappropriate fault code.

Note that the ECU might provide a fault code whichindicates that a fault exists on the sensor circuit.However, many systems will provide a code that states:‘the voltage level in the temperature sensor circuit is tooHIGH’.

The ECU cannot determine the exact nature of the fault:it can only establish that a fault exists, causing a highvoltage. Therefore the technician must still carry out adetailed investigation of the circuit.

Self-diagnosis on other types of sensor or circuitThe system ECU is able to monitor any of the circuitsconnected to it. The ECU is effectively pre-programmedwith the acceptable values for the various circuits, andis therefore able to identify a fault when values lieoutside acceptable limits.

Many sensors on a system provide a digital signal,i.e. a signal that consists of on/off pulses, such as thesignal from a Hall effect trigger used for vehicle speedsensors, ignition triggers, etc. The ECU can monitor thepulses and detect the correct operating voltage for thesignal, or whether the pulse is acceptable or unavailable.

Other sensors provide analogue signals, which againmight consist of a series of pulses. A crankshaftspeed/position sensor usually provides a series of pulseswhich the ECU is able to detect as being at a certainvoltage. Again, the ECU can detect whether the signal isacceptable or unavailable.

The ECU also provides operating signals to actuatorssuch as injectors or idle speed control valves. If theactuator and associated circuits are good, an electriccurrent will flow through the ECU in order to controlthe actuator. The ECU is therefore able to recognisemany of the faults owing to the fact that, if there is afault, the current flow might be incorrect, or theremight be no current flow at all if the circuit is broken.

In all cases where a fault is identified by the ECU,the technician should attempt to gain as muchinformation as possible about the operation of thesystem. Knowledge of the way in which the ECU detectsparticular faults and provides substitute values canassist in accurate diagnosis.

Limp home/emergency operation In a large number of cases when a fault is detected bythe ECU in a sensor circuit the ECU might be able tosubstitute a value for the failed circuit. If thetemperature sensor signal voltage is outside theexpected values, the ECU will detect the fault. The ECUhas programmed substitution values for certain faults,which can be used to ensure that the engine can stillrun. With a temperature sensor fault, the ECU couldsubstitute a temperature value, such as a warm runningvalue, that enables the car to be driven to a repairworkshop. However, the engine would be difficult tostart from cold because the substituted value is for awarm engine.

Modern systems can implement a more complexsubstitution process. Again using the temperaturesensor fault as the example, the ECU timer facilityenables it to calculate the length of time the engine hasbeen running, and then internally substitute atemperature value depending on that time period (e.g.if the engine has been running for an hour, thesubstitution value would be equivalent to normal


ECU

Voltage rises toequal supply voltageindicating opencircuit to ECU

ECUearth

Break in wiring givingan open circuit

Figure 3.86 Temperature sensor circuit with a break in the wiring(open circuit)

[647] Chapter 03 25/8/06 15:38 Page 153

engine temperature). When the engine is switched offfor a long period, the ECU would calculate that theengine was cold, and provide a cold runningtemperature value. At various time intervals, the ECUcan substitute progressively warmer temperature valuesuntil the engine is assumed to be at normal operatingtemperature.

It is possible for the ECU to use substitution valuesfor many of the sensors. It is often the case that thesubstitution process allows the engine to operatewithout any indications that there is a fault. The driverwould often be unaware that a fault exists, except thatthe warning light on the dashboard illuminates.

There are however a number of faults that will notallow a substitute value to be used. An obvious exampleis a power supply or earth circuit fault. Another sensorthat cannot be easily substituted is the main crankshaftposition/speed sensor. Without the signal from thissensor, the ECU has no reference to engine speed orexact crankshaft position, but if additional positionsensors are fitted to the engine (such as a camshaft orcylinder identification sensor) these can allow the ECUto provide a limited means of operating the enginesystems.

There are a number of terms used to describe theprocess whereby an ECU substitutes operating values;examples include:

● limp home● limited operating strategy (LOS)● emergency operating strategy.

Faulty signal values that are within expectedlimitsBasic self-diagnosis and fault code systems (includingmany of the systems produced before the end of the1990s) have certain limitations, because they onlydefine a fault when the signal is outside expected oracceptable limits. This method of diagnosis can lead tosituations where a fault does exist, but the ECU cannotrecognise it.

Again, using the temperature sensor circuit as anexample, if the sensor itself failed in such a way that thesensor resistance remained at a value corresponding toa normal engine temperature, the voltage in the sensorcircuit would also be at an acceptable value. The ECUwould not then recognise this failure.

If, for example, the sensor failed, so that theresistance of the sensor (and therefore the voltage in thecircuit) corresponded to 40°C (a warm engine) the ECUwould only detect this value, and provide a slightly richfuel mixture which is applicable to the warm enginetemperature. That the voltage in the circuit is ‘stuck’ atthis value will not be registered as a fault by these oldersystems.

However, as described in the previous paragraphsdealing with limp home/emergency operation, mostmodern systems now have a facility to recognise thatthe value is unchanging: the ECU can then provide asubstitute value.

Many systems up until the late 1990s (and possiblylater) did not have the facility to recognise faults forwhich the value remained within acceptable limits.Therefore technicians should always be prepared forself-diagnosis and fault code systems that do notindicate a fault, even if a fault exists in the system.

Overcoming the limitations of less sophisticatedself-diagnosis systemsThe previous paragraphs highlighted the fact that lesssophisticated systems have certain limitations that mustbe accounted for when carrying out diagnosis.However, most of these limitations have now beenovercome by the system manufacturers, by increasingthe capability of the self-diagnostic section of thecomputer.

More capable systems operate on what is sometimesreferred to as ‘improbable or implausible’. In effect, theECU is programmed with a degree of ‘intelligence’,which enables it to judge whether a signal or value on acircuit is probable or plausible.

For example, a temperature sensor value notaltering from a cold value when the engine has beenrunning for a considerable amount of time would beregarded as improbable or implausible. The ECU canuse more than one item of information from the othersensors or from programmed logic to make a judgment.An example of programmed logic would be: ‘a coldtemperature value cannot be correct when the enginehas been running for a long time’.

A signal would not be plausible if:

1 the throttle position sensor indicates that thethrottle has just opened fully

2 the engine speed has increased3 the airflow sensor indicates that there has been no

change in the air being drawn into the engine.

The above situation would indicate a likely airflowsensor fault, which could in fact be overridden by theECU; in this case, it could ignore the airflow sensor faultand rely on throttle position and engine speed as ameans of calculating the fuelling and ignitionrequirements. Therefore, the ECU on a modern systemhas far better ability to diagnose a fault and, wherenecessary, override the faulty inputs.

Adaptive strategyAdaptive strategy provides the ECU with the ability torelearn the values provided by the sensor circuits.

A simple example is the signal voltage provided by athrottle position sensor when the throttle is in the idleor closed position. The ECU might initially beprogrammed to expect a certain voltage, e.g. 0.5 volts(with a small tolerance or range), when the throttle isclosed or in the idle position.

However, wear in the linkage and other changesthat can occur over a period of time, can result in achange in the voltage when the throttle is closed. If,over a period of time or at a set time during the systemoperation, the ECU is able to detect that this voltage is


[647] Chapter 03 25/8/06 15:38 Page 154

now different, it can replace the original expected valuewith the new value.

Although not directly related to fault code reading,it is worth noting that an adaptive strategy can have amajor effect on the way the system identifies or dealswith a fault. Additionally, it will have a direct effect onthe success of rectification work. The particularexample of the throttle position sensor voltage (andmany other examples of the ECU being able to relearnvalues) may not have a direct effect on the self-diagnostic process but if a sensor or a circuit does fail onthe vehicle and a new sensor or component is fitted, theECU must go through a period of ‘relearning’ the newvalues in order to be able to provide the correct controlover the engine.

It is worth noting that, after rectification work iscarried out, or even if a battery has been disconnected,the ECU might need to relearn the values of the sensors,etc. For example, some engine management systemscan take more than 16 km (10 miles) to relearn sensorvalues. It is also a common requirement that an enginemust go through a full cycle of operation, such as coldstart, load and cruise conditions, etc., before it can fullyrelearn the system signals. The engine may thereforenot perform properly until the ECU has relearnt andadapted to the new values.

When a vehicle battery is disconnected the ECUmight suffer memory loss, so it is advisable to test drivethe vehicle after the battery has been reconnected toensure that the engine is performing normally.

3.7.3 Live data/serial data and otheradditional functions of self-diagnostic systems

Live data/serial dataIn addition to providing fault codes, modern systemscan give other information to the technician about theoperation of the vehicle system. Engine managementsystems, for example, are able to provide live data viathe fault code reading equipment.

The ECU is already monitoring the circuits of theengine management system in order to control thesystem. The ECU relies on signals from the sensors, etc.to control fuelling, ignition timing, emissions and otherfunctions. It is relatively easy for the ECU to output ortransmit those same signal values to a code reader orsimilar item of test equipment.

If a capable code reader is connected to the enginemanagement system diagnostic plug, the code reader isable to display a considerable range of measurementsrelating to the operation of the engine managementsystem. Different systems provide different items ofdata, but the following list indicates just a few of thesensor measurements that can be accessed via the livedata systems, with the use of a suitable code reader orsimilar equipment:

● battery voltage● engine speed● airflow sensor signal voltage● coolant temperature sensor signal voltage● throttle position sensor voltage● lambda (oxygen) sensor signal voltage● air temperature sensor signal voltage● MAP (vacuum) sensor signal.

Additionally, the ECU can also provide control signalvalues covering:

● ignition timing● injection control● EVAP system solenoid control ● idle speed valve control duty cycle (on/off ratio).

In fact, a wide range of control signal values and sensorsignal values can be accessed on modern systems.Additionally, operational information can be accessed,such as whether the system is operating in closed oropen loop for emission control (see section 3.5.7).

With the live data information, it is possible for thetechnician to check on ECU operation and ensure thatthe system is operating as intended.

Note that some of the values may be converted bythe code reader into a more acceptable form. Thecoolant temperature sensor voltage could be displayedin the normal way, as a voltage, but can be converted togive the coolant temperature in degrees, e.g.0.9 volts = 85°C.

Live data allows the technician to view aconsiderable range of readings that would otherwiseneed to be accessed by more traditional measurementtools, such as multimeters or oscilloscopes.

In addition to the examples of live data shownabove, the ECU can also output information aboutcomputer or ECU calculations. This can include thecalculated load and other values which are usuallydisplayed in a format that does not assist techniciansunless they have access to dedicated data.

The live data output and other informationdelivered by the ECU is often referred to as ‘serial data’or ‘data stream information’. These terms simply refer tothe way in which a code reader or other test equipmentcommunicates with the ECU. As with personalcomputers, data communication (such as between thePC and a printer) is achieved by transferringinformation in series along a wire, i.e. with one piece ofinformation following another.

Using live data for diagnosisThere is no doubt that live data is an aid to diagnosis. Inprevious sections we have noted that fault codes areprimarily of use if the fault causes the signal values tolie outside acceptable limits. However, if a fault resultsin signal voltages or values that are within acceptablelimits, it may not be recognised by the ECU. Therefore,if an engine management system has an undetected orunrecognised fault, but the engine is obviously not


[647] Chapter 03 25/8/06 15:38 Page 155

running properly, it is possible to examine the live datato see if the signal values are correct for the engineoperating conditions.

Again referring to the coolant temperature sensor asan example, the sensor signal voltage might indicate‘0.9 volts = 85°C’, but this could not be correct if theengine had not been run for some hours. In otherwords, the engine is cold but the temperature sensor isindicating that the coolant is almost up to the hotoperating temperature. The values would not berecognised as a fault on many systems, but the live datadisplay allows the technician to compare values againstwhat should be expected for the conditions.

As another example, if the ECU detects a signal thatis not plausible or probable, such as the airflow sensorsignal discussed previously (see section 3.7.2), the ECUmight not provide a fault code that indicates the exactfault. However, reference to the live data should enablean experienced technician to identify that the airflowsensor signal is incorrect under certain operatingconditions.

Snapshot or playback modeVarious terms are used to describe the snapshot orplayback facility. This facility enables the fault codereader to display fault codes, or in some cases live datathat were recorded earlier.

The exact capability varies depending on thesystem; however, it is possible for the code reader (inconjunction with the ECU) to capture codes or data at atime when a fault occurred. This facility is extremelyuseful when trying to trace intermittent faults that donot necessarily show up during workshop tests. The testequipment can be connected to the vehicle, which isdriven until the intermittent fault occurs.

It is often the case that the code reader/ECUcombination can record a range of measurements or livedata at the time a fault occurs. The measurements anddata are monitored while the vehicle is being driven orthe engine is running. It may be necessary for someoneto press a start button on the code reader when the faultoccurs (such as a noticeable misfire). Themeasurements received from just prior to the fault untiljust after it are stored.

Some vehicle systems might have already registereda fault code relating to an intermittent fault. The codereader/ECU combination can be programmed tocapture and store the data next time the particularintermittent fault occurs. In some cases, the ECU willautomatically store applicable data when a fault occurs,and it is then simply a matter of connecting the testequipment at some later stage to obtain the data thatwas stored at the time the fault occurred.

The final result is that the data can be analysed afteran intermittent fault has been noticed, thus enablingthe technician to perform an accurate diagnosis.

Not all vehicle systems and not all code readersallow the snapshot or playback facility to function.

Service adjust modeSome vehicle systems, notably engine managementsystems, have a service adjust mode which can be usedfor some basic service set procedures. On some oldersystems, idle speed and other operating values could beset or adjusted, but with most modern systems thisfacility is not required.

There are occasions, however, when some basicsettings can be altered, such as in a country with lowquality fuel. It is possible on some vehicles to effectivelyreprogram the ECU so that it alters the timing andfuelling characteristics to suit different fuel qualities. Inreality, the ECU has a number of programmedcharacteristics already located in its memory: it issimply a matter of selecting the appropriate program byusing the code reader or other suitable equipment.

Actuator simulation testsIt is becoming increasingly common for systems toincorporate an actuator test facility. As well asmonitoring the electrical circuits for an actuator such asan injector, modern ECUs are able to provide a controlsignal to the actuator (in effect a simulation of theactuator signal). This control signal has a known value,so that the operation of the actuator can be checked. Itis usual to implement this facility using the code reader:the technician selects the appropriate function, whichcauses the actuator signal to be transmitted from theECU to the actuator, which then operates.

With modern systems, the ECU can provide acontrol signal to a number of components or actuatorsin the system. In engine management systems, forexample, the ECU can control the injectors, the idlevalve, the EVAP canister purge valve and possibly otheritems (such as the exhaust gas recirculation systemvalve). During an actuator test, the ECU is instructed tooperate these components independently and normally,when the engine is not running.

In this way it is possible for the technician to hear orobserve operation of the various actuators, which helpsto establish whether these components are functioningcorrectly, and that the associated wiring is good.

When an actuator test is carried out on the injectors,it is likely that fuel will be injected into the intakemanifold or into some cylinders. It is advisable that thefuel pump is disconnected to prevent excessive fuelentering the engine. As a minimum precaution, thespark plugs should be removed and the engine crankedover to evacuate excess fuel.

‘Set procedure’ testingSeveral engine management systems rely on thetechnician following a set procedure or test routine thatrequires a number of actions to be performed by thetechnician. These set procedures are usually dictated tothe technician via the code reader. In effect, instructionsare provided for the actions that the technician mustperform.


[647] Chapter 03 25/8/06 15:38 Page 156

When the code reader is connected to a diagnostic plug,various messages displayed on the code reader providethe technician with relevant instruction to perform acertain task, before going on to the next phase oftesting. On some systems, the whole self-diagnosticroutine is based around the need for the technician toperform certain procedures all the way through the testsequence.

Typical procedures that need to be carried outinclude altering the engine speed, turning the steering(to check power steering pressure switch operation)and depressing the brake pedal (to check brake switchoperation).

If the procedures are not carried out as instructed,the test results will be incorrect, and the ECU mightdiagnose a fault that does not in fact exist. In othercases, the ECU will not proceed to the next set of testsuntil the correct responses or readings have occurred orbeen recorded.

Clearing codesIt is important that, when a fault has been rectified, theoriginal fault codes are removed from the ECU memory.The process for clearing codes varies with each makeand model of vehicle, and will also depend on whetherblink codes are being accessed or a code reader is beingused.

Reference should always be made to the specificinformation on each vehicle to ensure that the correctprocess is used to clear old fault codes.

Common problems with the use of code readingequipmentAlthough self-diagnostic systems, code reading andretrieval of live data are generally reliable facilities, anumber of problems can arise which are often blamedon the code reader, but may in fact be due to commonproblems such as operator error or system faults thatprevent good test results being obtained.

The following lists those common faults that oftenresult from what are termed ‘communication errors’between the code reader and the system ECU.

1 All instructions provided by the equipmentmanufacturer might not have been followed exactly.When certain procedures are not performed in thecorrect order or as instructed, the ECU might notallow further tests to continue or might provideincorrect results.

2 Instructions provided on the code reader displaymight not be followed accurately. If any procedure iscarried out incorrectly or not at the correct time, theECU might not allow tests to continue or incorrectresults might be obtained.

Minor changes are often made to systemsbecause of modifications that occur throughout theproduction life of a vehicle. This can result in smallchanges to the procedures that have to be carriedout by the technician when accessing fault codes,etc. Although the technician might be conversant

with a particular system, a variant of that systemmight require slightly different procedures or evenuse different fault codes.

3 All connections between the code reader and thediagnostic plug must be checked. Some systems andcode readers require a separate power supply orearth connection to be made (usually an adapterlead or harness is provided to connect theequipment to the battery). Check whether thesystem being tested requires the use of a separateharness.

4 The code reader or the application cartridge/cardmust be correct for the system being tested. It isunfortunately not uncommon for a code reader (orthe application cartridge) to be designed to operatewith a particular range of systems, and for thevehicle manufacturer to then modify the system inproduction, which can prevent communication withthe original code reader or cartridge.

5 All wiring between the code reading equipment andthe diagnostic plug, and between the plug and theECU must be checked. It is also necessary to checkfor poor earth connections to the ECU and check thepower supply to the ECU and to the code reader.

A reduced power supply (flat battery) can causecommunication to cease if the voltage level falls belowa certain limit.

Different terminology for code readingequipmentA scanner or scan tool is another term (often used todescribe American equipment) applied to code readingequipment. In general, the term code reader is appliedto something that simply extracts codes from thevehicle system. A scanner is an item of equipmentwhich effectively scans the ECU memory forinformation.

However, the terms code reader and scanner areloosely used by the industry as product names, as wellas being general terms to identify equipment thatcommunicates with a vehicle system ECU.

3.7.4 Use of test equipment toaccess fault codes

The test equipment (code reader or scan tool) providesspecific instructions about equipment operation and theconnection processes. It is useful to understand sometypical processes for equipment use and for accessingblink codes, fault codes and live data.

Blink codes via the dashboard warning lightsA blink code is the simplest form of fault code reading,especially when displayed via the dashboard warninglight. Dashboard warning lights are now fitted tovirtually all passenger vehicles, with a typicalappearance as shown in Figure 3.87.


[647] Chapter 03 25/8/06 15:38 Page 157

It is necessary to refer to the specific vehicleinformation to find out how its blink codes can beretrieved, but the following example shows how blinkcodes are retrievable on one particular range ofvehicles.

Example of blink code retrievalIn this example, it is necessary to instruct the ECU toprovide a blink code output to the dashboard warninglight. Here the instruction to the ECU is provided bylinking two terminals on the diagnostic harness plug(Figure 3.88). When the two terminals are linked andthe ignition is switched on, the ECU causes the warninglight to flash an appropriate number of times,depending on whether there is a fault code stored in theECU memory.

As discussed in section 3.7.1, the number of flashesof the warning light represents the fault code: countingthe flashes enables the technician to read the code.

There are many other methods used by vehiclemanufacturers for instructing the ECU to output faultcodes via the warning light, so refer to specificinstructions for each vehicle.

Retrieving blink codes via a vehicle system LEDSome vehicles use an LED located on the ECU (or inother locations on the vehicle) to enable a blink codeto be read by the technician. As with the warning lightblink codes, a specific procedure needs to beperformed to instruct the ECU to output the code tothe LED light.

Accessing blink codes using an LED probe On some vehicles (mainly older models), it is possibleto access the blink code using an LED probe (Figure3.89). The process usually involves connecting theprobe tip to a terminal or connection of the ECU, or to aterminal of a wiring harness plug, while the otherconnection for the probe (the crocodile clip in Figure3.89) remains connected to earth.

As with blink code retrieval using the warning light,it is usually necessary to instruct the ECU to output thecodes. Refer to specific vehicle information for thecorrect procedure and to establish where the LED probetip should be connected.

One example of a procedure for instructing the ECUto output a blink code using an LED probe is where twoterminals on a connector or harness plug are linkedusing a switch. This process was used in some oldergeneration Audi models.


Figure 3.87 Dashboard warning light

ENGINE

Link tool

Diagnostic connectors to be linked

Check connector

(a) (b)

Diagnostic plug

Figure 3.88 Example oflinking terminals on thediagnostic plug to instructthe ECU to provide blinkcode outputa Link tool inserted intoconnector to initiatediagnostic sequenceb Plan view of connectorterminals

Figure 3.89 An LED probe for accessing the blink code

[647] Chapter 03 25/8/06 15:38 Page 158

A switch is connected across the two specifiedterminals, and then the LED probe connected acrosstwo other terminals of the harness plug. The specificinstructions will indicate when the switch should beclosed or open. This action makes or breaks the circuitacross the two terminals. If the switch is opened andclosed at the appropriate times, the ECU will output thefault codes as a blink code on the LED probe.

Obtaining codes and fault related informationusing a code reader/scan toolUsing a code reader/scan tool is the most informativemethod of retrieving fault related information from theECU.

A code reader/scan tool must be connected to theECU to obtain the information; connection is made to adiagnostic plug (Figure 3.90), which in most modernvehicles (with engine management systems) is locatedinside the passenger compartment. However, someother vehicle systems, such as ABS, have a diagnosticplug located in the engine compartment or elsewhere.

Note that European regulations now specify that theengine management diagnostic plug is to be locatedwithin a defined region of the passenger compartment;this is explained more fully in section 3.7.5.

Appropriate vehicle application softwareMany code readers can use different cartridges, pods orsoftware to enable them to function on different makesand models of vehicle and on different makes of vehiclesystem. Although some code readers are designed tooperate on only one (or a few) makes and models, mosthave the facility to use different application software,

which can be purchased from the code readermanufacturer. Depending on the equipment design, thesoftware is contained within replaceable pods orcartridges. In some cases new software can bedownloaded via the Internet.

Most code readers are capable of operating withmany different makes and models of vehicle. Softwareupgrades enable newer models to be covered withoutnecessarily requiring a complete new code reader.

Starting the communicationFor almost all vehicle systems, the ECU requires someform of password before it will output the fault codes orlive data to the code reader. The password is usually anelectronic code that is transmitted from the code readerto the ECU. In most cases, the appropriate softwarecartridge is installed in the code reader, or theappropriate vehicle is selected from the menu on thecode reader, causing the correct password to be sent tothe ECU. When the ECU has received the password, theECU and the code reader can then communicate.

Retrieving fault codes and dataOnce communication has been established between thecode reader and ECU, the capacity to retrieve codes andinformation will depend on the equipment being usedand the system being tested. In general, however,instructions are provided on the code reader display, orare included in an operator manual.

Following the correct procedures will then allowfault codes, messages or live data to be retrieved, aswell as enabling other functions to be performed, suchas actuator tests and intermittent fault detection.


Figure 3.90 Connection of acode reader to the vehiclediagnostic plug

[647] Chapter 03 25/8/06 15:38 Page 159

Different vehicle systems will provide different levels ofinformation. Some may provide only simple fault codes,whilst modern engine management systems provide aconsiderable depth of live data, etc. Additionally, lowercost code readers might not be able to display all of theinformation available from the ECU, and may not beable to initiate many of the other ECU test functions,such as actuator testing.

3.7.5 EOBD (European on-boarddiagnostics)

Emissions related legislation affecting on-boarddiagnosticsThe term on-board diagnostics refers in general to thefacilities already described relating to self-diagnosis.However, originally in America and more recently inEurope, legislation has imposed certain standards foron-board diagnostic systems, with particular regard toemission control systems. In reality, because emissionsystems are now under the control of the enginemanagement system, the operation of the enginemanagement system is influenced by the legislation.

The American market has had on-board diagnosticlegislation for some years (OBD 1 and OBD 2). InEurope, legislation was introduced in January 2000that applied to new models (subsequently embracingdiesel engines as well as petrol engines). In Europe, theon-board diagnostic legislation is referred to as EOBD(European on-board diagnostics) and is an adaptationof the American OBD 2, which was originallyintroduced in 1996. EOBD is part of a range ofregulations introduced by the EC aimed at controllingor reducing the emissions from motor vehicles. Thereare many aspects to EOBD regulations, but the relevantissues are discussed within this section.

Whilst the regulations are primarily aimed atemissions control systems/engine managementsystems, other vehicle systems also affect emissionlevels, such as the air conditioning. The air conditioningsystem may be operating at idle speed, thus causing theidle speed to reduce as well as demanding enginepower. An engine management system must ensure thatthe idle speed is maintained, and also ensure theair:fuel ratio is controlled, so power can be developedwithout increasing emissions significantly. It is commonto find fault codes on modern systems that embrace airconditioning functions, automatic transmissionfunctions, and also codes for other vehicle systems thatcommunicate or co-operate with the enginemanagement system.

Long term monitoring of emissions from thevehicleOne objective of the regulations is to ensure thatemissions levels from motor vehicles are maintained atacceptable levels for the operating life of the vehicle,and also that, when the emissions levels are

unacceptable because of a fault, vehicle systems areable to detect the incorrect emissions and identify thefault. In effect, the regulations impose a requirement formonitoring emissions and emissions control systems(including the engine management system). Thisrequirement effectively defines the technical functionsof the engine management system and some of the ECUdiagnosis software. EOBD regulations resulted in theintroduction of post-cat monitoring, i.e. an oxygen orlambda sensor located after the catalytic converter(section 3.5.7). This additional lambda sensor allowedthe ECU to monitor the efficiency of the catalyticconverter process. A second lambda sensor signalindicates the oxygen content of the exhaust gas in thecatalytic converter.

The self-monitoring and self-diagnostic capability ofthe modern engine management ECU is highlycomplex, and the software program embraces manyvariables that can occur during engine managementsystem operation. As a simple example, if the signalvoltage from a temperature sensor exceeds themaximum programmed limit (e.g. 4.8 volts, as used inearlier examples in section 3.7.2), this would beregarded as a fault on older systems. However, EOBDsystems must have a built-in tolerance that allows thesignal voltage to exceed the maximum limit for a verybrief period of time.

ECU programming must therefore permit anoccasionally faulty signal, so long as it occurs only foran exceptionally short period and on a limited numberof occasions; under these specified conditions, the ECUwould not regard the fault as permanent, so no faultcode would be stored and the warning light would notbe illuminated.

Standardisation of fault codes, communicationand diagnostic plugsStandardised codesAnother significant aspect of EOBD is thestandardisation of many engine system related faultcodes. Prior to EOBD, vehicle manufacturers used theirown coding system to indicate faults – there was noconsistency across makes or models. Whilst vehiclemanufacturers are still allowed to use their own faultcode system and codes (which can be used in theirauthorised repair workshops in conjunction with thededicated vehicle test equipment), there is astandardised list of codes and messages that apply toemissions related system faults. The standardisedsystem of codes must be accessible to general purposecode reading equipment, so that equipmentmanufacturers can produce code reading equipmentthat provides the same fault codes and messagesirrespective of the vehicle being tested.

The standardised codes, which number in excess of500, are referred to as P Zero or P0 codes. Vehiclemanufacturers’ own coding systems are referred to asP1 codes. The P0 codes must still be accessible even if avehicle manufacturer uses its own P1 coding system.


[647] Chapter 03 25/8/06 15:38 Page 160

The letter P denotes the power train (effectively theengine and emission systems). Other code letters referto different aspects of the vehicle, as shown below:

P = power train (standardised EOBD codes)C = chassis system codesB = body system codesU = network system codes (this relates to the

communication networks where differentcomputers or ECUs communicate with eachother).

A zero (0) following a letter (as in P0) indicates astandardised code such as EOBD, or a code establishedby the Society of Automotive Engineers (SAE) or otherorganisations. A number 1 or 2 following a letter (as inP1 or P2) indicates a manufacturer’s own code. Notethat EOBD does not at this time embracestandardisation or access to live data, but only faultcodes and fault messages.

Figure 3.91 shows the structure of the standardisedP0 fault codes and messages.

Standardised diagnostic connector plugEOBD specifies a standard diagnostic plug which has 16terminals or connector pins (Figure 3.92). The functionof the pins is also standardised, i.e. some of the 16 pinsare designated as part of the EOBD system and areallocated specific functions such as battery voltage,power supply and earth connection for the code readercommunication terminals (the terminals through whichthe codes and data are transmitted from the ECU to thecode reader). This allows a single code reader toconnect to any modern vehicle with a single connectorharness and plug. The result is that vehicle testingstations, police forces and workshops can have astandard code reader to retrieve EOBD informationfrom any vehicle.

The location of the diagnostic plug has also beendefined within EOBD regulations. In general, it is sitedclose to the driver’s seat, approximately between thecentre line of the vehicle and the steering column. Theconnector plug is found on many vehicles just under thedashboard, adjacent to the driver’s leg, although insome vehicles it is on the centre console.

Malfunction indicator lampThe malfunction indicator lamp (MIL) is another termfor the engine warning light. The MIL is intended foruse when emissions/engine management system faultsoccur, or emissions are outside predefined limits. TheMIL must be positioned on the dashboard and must notbe red when illuminated. When the MIL is illuminated,it indicates that a system fault exists, and that theengine management system might be operating usingthe ‘limp home’ function. There are two main operatingstrategies for the MIL.

1 If the MIL illuminates permanently (no flashing)when the engine is running, it indicates that theECU has detected a fault that could allow excessiveemissions to be produced. It also indicates that afault code or message is stored in the ECU memory.The fault code can be accessed using a code readeror scan tool.

2 If the MIL flashes continuously when the engine isrunning, it indicates the ECU has detected an enginemisfire. In such cases, it is possible for excess oxygenand fuel to ignite in the catalytic converter, causingpermanent damage, or at least to accelerate itsageing. If the MIL is flashing, the driver shouldideally reduce engine speed and load, and take thevehicle to a repair workshop as soon as possible.


X X X X X

Fault (00–99)

1. Fuel and air metering2. Fuel and air metering (injector circuit)3. Ignition systems or misfire4. Auxiliary emission controls5. Vehicle speed control and idle control system6. Computer output circuit7. Transmission8. Transmission

0. SAE1. MFG

B. BodyC. ChassisP. PowertrainN. Network

Figure 3.91 Examples of EOBD standard P0 fault codes

1 2 3 4 5 6 7 8

9 10 11 12 13 14 15 16

Figure 3.92 16-pin diagnostic connector plug

OBD2 and EOBD are similar – but not identical –take care when interpreting results!

Modern OBD systems use a standard diagnosticplug/socket located in easy reach of the driver’sseat

Standardised communication EOBD rules identify a common ‘language’ that shouldbe used in the computer system, so that itcommunicates with a standard type or general purposecode reader. The standardised language is in factreferring to the computer protocol used when the codereader and ECU communicate. Included in thestandardisation are the passwords to enable the codereaders to gain access to the information and datawithin the ECU.

Key

Poin

ts

[647] Chapter 03 25/8/06 15:38 Page 161

Misfire detectionAnother feature of EOBD systems is the facility to detectmisfires. There is a strong possibility that the catalyticconverter might be damaged because of a misfire, andhigh emissions of HC are produced when misfires occur,so the engine management system must be able todetect misfires and, in some cases, cut off fuel to theaffected cylinder.

Misfires can be detected using several methods aslisted below.

1 Monitoring the crankshaft speed andacceleration – when the combustion process occursin a cylinder, it produces power that forces thecrankshaft to accelerate. If the combustion processin a cylinder is not efficient, or produces less powerthan the other cylinders, the acceleration at thecrankshaft will not be as great as for the othercylinders. This difference enables the ECU toestablish which cylinder is misfiring or running lessefficiently.

2 Spark detection is a method whereby the ECUmonitors the voltage in the ignition coil. When amisfire occurs, the changes in secondary circuitfiring and spark voltage can also affect the voltagein the primary circuit. The process is not dissimilarto examining the ignition circuit voltages using anoscilloscope or test meter. By comparing voltagesacross the different cylinders, and referencingexpected or previous voltages for that cylinder, theECU can identify which cylinder is misfiring.

3 A misfire causes an increase in the unburnedoxygen content of the exhaust gas. Thelambda/oxygen sensor will detect the high levels ofoxygen, and the ECU will register this as a fault.This process alone does not identify which cylinderis misfiring, but by analysing the misfire detection

information provided by other methods such as in(1) or (2), the ECU can assess which cylinder is atfault.

Cylinder isolationIf exhaust oxygen levels are excessively high, which willusually be accompanied by high levels of unburned fuel(HC), it is possible for the ECU on many modernsystems to cut off fuel delivery to the affected cylinder.As well as reducing the high HC emissions, cutting offthe fuel supply to the affected cylinder reduces the riskof excessive combustion of fuel and oxygen in theexhaust system and catalytic converter, combustionwhich could create high temperatures and damage thecatalytic converter.





[647] Chapter 03 25/8/06 15:38 Page 162

ENGINE MANAGEMENT –DIESEL INJECTION

Cha

pter

4

4.1 MODERN DIESEL FUEL SYSTEMS


Development of modern diesel fuel systems

Development and electronic control of the rotary type diesel injection pump

Cold-start pre-heating systems

Electronic control of diesel injection (common rail systems)

4.1.1 Emissions, economy andengine performance

As with the petrol engine, legislation has forceddesigners of diesel engines to reduce emissions ofpollutants, whilst at the same time consumer demandfor improved engine performance and economy haspotentially conflicted with emission requirements.However, again as with petrol engines, improvedengine design and the development of the fuel systemshas resulted in improvements in almost all areas.

EmissionsIn comparison with most petrol engines, the dieselengine operates with very weak air:fuel ratios. Apartfrom some modern direct petrol injection enginesoperating at light load, a petrol engine generallyoperates with a lambda value (excess air factor) of closeto 1. Diesel engines, however, always operate with alambda value greater than 1 and as high as lambda 1.4.

Because of this high excess of air, carbon monoxide(CO) emissions are exceptionally low, with values aslow as 0.01% at idle and with a typical maximum ofaround 0.2% (200 parts per million) at full load.Hydrocarbon (HC) emissions are also low at less than50 parts per million (ppm) at full load, but this can riseto approximately 500 ppm at idle speed, which is in facthigher than emissions from petrol engines.

Emissions of oxides of nitrogen (NOx) are high withdiesel engines and, as with petrol engines, levels canreach 2500 ppm.

The diesel engine does not generally use any formof air volume or air mass control, i.e. there is normallyno throttle control to regulate the volume/mass of airentering the cylinder. The volume of air drawn into the

cylinders is entirely dependent on engine speed, loadand design; therefore power or torque control is totallydependent on fuel quantity (air:fuel ratio). Enginetorque and power are therefore controlled by thequantity of fuel injected into the cylinder. Exceptionallyweak mixtures will result in low power and torquelevels, suitable for light load operation. However, whenit is necessary for the engine to produce higher levelsof torque and power, the quantity of fuel must beincreased (a richer air:fuel ratio).

Since engine torque and power are dependent onthe air:fuel ratio, when power or torque are required arich mixture (reduced oxygen content) is providedwhich results in an increase in CO and also in NOx

emissions (due to combustion temperature increase).Additionally, the levels of soot emitted increase withricher mixtures (soot can be regarded as fuel dropletsthat have not vaporised, compared with HC emissions,which are vaporised fuel).

In recent years, legislation has imposedprogressively tougher emissions limits, with the Euro 4legislation for 2005 dictating further reductions inemissions. Various emissions reduction techniques aretherefore used on modern diesel engines with particularfocus on electronic control of diesel injection systems.In fact, the introduction of ‘common rail’ diesel injectionsystems is an interesting development, with theirsimilarity to direct injection petrol systems that havealso been relatively recently introduced.

The subject of diesel emissions is frequentlydebated by the politicians and scientists, with theresult that there is a continuously changing view as towhether diesel or petrol engines produce the moreharmful emissions.

[647] Chapter 04 25/8/06 15:39 Page 163

Economy and engine performanceTraditionally, because diesel engines operate withrelatively weak air:fuel ratios (compared with petrolengines), fuel consumption is generally much lowerthan an equivalent petrol engine. However, dieselengines have generally produced less power than anequivalent size of petrol engine, although torquelevels are usually higher. The diesel engine is,however, ideally suited to turbocharging, which canenable the diesel to produce high torque and highpower outputs.

Many modern passenger car diesel engines withturbocharging can in fact match the performance oftheir petrol equivalents (even in the sportier models).Equivalent performance with reduced fuel consumptionis a major consumer benefit.

Another significant factor for the consumer is thecost of diesel fuel. In general, in Europe, diesel fuel isless costly than petrol (with the notable exception of theUK). For this reason, many European countries havehigh percentages of diesel cars, with in excess of 50%being normal in some countries. In the UK, there hasbeen a trend towards a higher percentage of vehiclesales being diesel (beginning around 2000), with up to30% of the total cars sold now being diesel enginevehicles (especially for smaller cars).

One other factor that has noticeably changed withthe diesel engine is noise. Modern injection systemsand engine designs have resulted in a considerablereduction in the traditional diesel knock, an importantfactor for the consumer. The intrusion of diesel enginenoise into passenger compartments is much lower andthere is also a significant reduction in noise outsidethe vehicle.

In comparison with petrol/gasoline engines,diesels operate with very weak mixtures. Thisresults in very low carbon monoxide emissions

Diesel engine performance was generally inferiorto that of petrol/gasoline engines but the additionof a turbocharger makes the performance similar.However, diesel economy is still better

4.1.2 Diesel engine developmentsAs with the petrol engine, fundamental diesel enginedesign has improved. General improvements embracethe combustion chamber and the use of four valves percylinder, as well as improved intake port designs.However, it is fuel systems that have provided thegreatest change to the diesel engine. With theprogressive introduction of electronic control, whichwas initially used to enhance the operation oftraditional diesel injection pumps (in-line as well asrotary), through to fully electronic common railsystems, electronic control has made some verysignificant improvements to the diesel engine.

Key

Poin

ts

Inevitably, the improvements have been very muchforced by environmental considerations (emissionslegislation), but with the addition of consumer benefits,the diesel engine has become a much more acceptedengine type, especially in countries where the petrolengine was traditionally the preferred option forpassenger vehicles.

4.1.3 Decline of in-line pump androtary pump injection systems

Electronic control for in-line diesel injection pumps isgenerally used only on large engines in heavycommercial vehicles. Passenger cars and lightcommercial vehicles increasingly used rotary typepumps through the 1980s and 1990s. With legislationforcing further emissions reductions, the trend hasmore recently been towards electronic control for unitinjectors and more frequently, common rail injectionsystems which are similar in layout and operation to amodern petrol injection system.

Figure 4.1 shows an in-line type pump (for a six-cylinder engine), which has electronic control to takecare of functions that were previously mechanically orhydraulically/pneumatically controlled. Injectiontiming (start of delivery) and fuel quantity arecontrolled by the ECU which passes a control signal tosolenoids or other actuators that then alter theposition of the mechanical devices within the pump.In effect, the pump operates in much the same way asolder designs but electronic systems enhance itscapability and accuracy. The in-line pump is notcovered in this book because of its lack of use inpassenger and light vehicles. For an understanding ofthe operation of in-line pumps, see Chapter 2 inHillier’s Fundamentals of Motor Vehicle TechnologyBook 1.

164 Engine management – diesel injection Fundamentals of Motor Vehicle Technology: Book 2

Figure 4.1 Bosch in-line diesel injection pump

[647] Chapter 04 25/8/06 15:39 Page 164

4.2.1 Rotary pumps ‘without’electronic control

This section covers just one type of rotary pump. SeeChapter 2 in Hillier’s Fundamentals of Motor VehicleTechnology Book 1 for additional information on theoperation of diesel injection pumps.

Figure 4.2 shows a VE pump fitted in a self-bleedingfuel system layout similar to that used on light vehicles.As with other rotary pumps, this type has one pumpingelement and a number of high pressure outlets, one foreach engine cylinder.

The rotary diesel injection pump 165

4.2 THE ROTARY DIESEL INJECTION PUMP

Figure 4.2 Bosch distributor pump fuel system

In addition to the basic features associated with moderndistributor rotary pumps, various add-on modules canbe fitted to the VE pump; these include:

● a solenoid operated fuel cut-off to give the driver akey start/stop operation

● an automatic cold starting module to advance theinjection

● a fast idle facility to give even running during warm-up

● torque control for matching the fuel output with thefuel requirement.

The section through the pump (Figure 4.3) shows thelayout of the basic sub-systems; these include:

● a low pressure fuel supply● a high pressure fuel supply and distributor● a fuel shut-off solenoid● a distributor plunger drive● an automatic injection advance unit● a pressure valve● a mechanical governor.

Low pressure fuel supplyDriven at half crankshaft speed by a drive shaft, atransfer pump with four vanes delivers fuel to thepumping chamber at a pressure set by the regulatingvalve.

This fuel pressure, which rises with engine speed, isused to operate the automatic advance unit. It alsogives an overflow through the pump body, which aidscooling and provides the self-bleeding feature. After

Figure 4.3 Bosch VE distributor pump

[647] Chapter 04 25/8/06 15:39 Page 165

passing through a small restriction at the top of thepump the surplus fuel is returned to the fuel tank.

High pressure fuel supplyFigure 4.4 is a simplified view of the pumping chamberwith part of the distributor head cut away to show thepump plunger. Besides rotating in the head to give avalve action, the plunger is reciprocated through aconstant stroke to produce the high pressure. The axialmovement is provided by a cam plate moving over aroller ring. The quantity of high pressure fuel deliveredto the injector via the outlet bore is controlled by theposition of the control spool. The control spool variesthe effective pumping stroke: the stroke increases as thespool is moved towards the distributor head andtherefore increases the quantity of fuel delivered.

In the position shown in Figure 4.4a the rotation of theplunger has caused one of the metering slits to open theinlet passage. At this point all outlet ports are closed.Prior to this, the plunger had moved down the chamberto create a condition for the fuel to enter and fill thehigh pressure chamber.

Slight rotation of the plunger closes the inlet portand causes the single distributor slit in the plunger toopen one of the outlet ports. Whilst in this position theplunger is moved up the chamber to pressurise the fueland deliver it through the outlet bore to the injector.

The position of the plunger at the end of theinjection period is shown in Figure 4.4b. At this point,the control spool has already allowed a considerablemovement of the plunger before the cut-off bore in theplunger has been uncovered. The exposure of this port


Figure 4.4 Principle of the VE pumping unit

[647] Chapter 04 25/8/06 15:39 Page 166

instantly reduces the pressure and terminates theinjection. Further pumping movement of the plungercauses the fuel in the pumping chamber to be returnedto the pump cavity. With the spool set in this maximumfuel position, which corresponds to the fuel requirementfor starting, a movement of the control spool to anextreme position away from the distributor headreduces the output to a minimum; this is the spoolsetting for slow running.

Fuel shut-offThe ‘no fuel’ or ‘stop’ position is provided by a solenoidoperated valve. The solenoid cuts off the fuel supply tothe inlet passage when the ‘ignition’ key is switched off.

Distributor plunger driveThe plunger must be rotated and reciprocated. Figure4.5 illustrates how this is done.

The distributor pump driveshaft is rotated at halfcrankshaft speed (for a four-stroke engine), and istransmitted via a yoke and cam plate to provide rotarymotion to the pump plunger.

Reciprocating motion is provided by the rotation ofa cam plate as it moves over four roller followers fixedto a roller ring. In a pump suitable for a four-cylinderengine, four lobes are formed on the cam plate andcontact between the plate and rollers is maintained bytwo strong plunger return springs. A yoke positionedbetween the driveshaft and the cam plate allows theplate to move axially whilst still maintaining a drive.

Pressure valveA delivery valve is fitted in the distributor head at theconnection point to the high pressure fuel lines (see4.3). The valve is used to seal the pressure in the highpressure line when the fuel delivery to the fuel outlet


port stops; when the pump element ceases to supplyfuel to the outlet port, the delivery valve closes whichimmediately causes the pressure to drop in the highpressure line which, in turn, causes immediate closureof the injector. However, pressure remains sealed in thehigh pressure line.

Automatic injection advance unitThe roller ring assembly is not fixed rigidly to the casing;instead it can be partially rotated through an angle of upto 12º to allow the automatic advance mechanismshown in Figure 4.6 to vary the injection timing.

When the pump is rotated, fuel under pressure fromthe transfer pump is delivered to the timing advancechamber via the pump cavity. A rise in the pump speedcauses the transfer pump pressure and flow to increase.The increase in pressure moves the timing advance

Figure 4.5 Plunger drive

piston against its spring, which in turn, causes theactuating pin to rotate the roller ring in a directionopposite to the direction of rotation of the driveshaft.The rotation of the roller ring advances the injectiontiming.

GovernorThe VE pump is fitted with either a two-speed or an allspeed governor. The layouts of these types of governorare similar, but differ in the arrangement of the controlsprings.

Figure 4.7 shows the main construction of a two-speed governor, which controls the engine during theidling and maximum speed operation. At other timesthe driver has near direct control of the quantity of fueldelivered and hence the power output of the engine.

The centrifugal governor, which consists of a seriesof flyweights, is driven from the driveshaft through

[647] Chapter 04 25/8/06 15:39 Page 167


Figure 4.6 Principle of the automatic advance

Figure 4.7 Governor – mechanical type

[647] Chapter 04 25/8/06 15:39 Page 168

gears with a ratio that steps up the speed. The highspeed flyweight rotation given by this ratio ensuresgood sensitivity of the governor, especially during theidling phase.

An increase in engine speed, and the associatedcentrifugal action on the flyweights, produces anoutward force that pushes a sliding sleeve against acontrol lever system. This lever, which is connected atits lower end to the control spool on the pumpingplunger, can move only when the sliding sleeve is ableto overcome the reaction of the spring that is in use atthat time.

StartingWith the accelerator pedal half depressed and thegovernor stationary, the starting spring pushes thesliding sleeve towards the flyweights and moves thecontrol spool to the maximum fuel position.

IdlingWhen the engine starts, the release of the accelerator,combined with the outward movement of theflyweights, causes the lever to move the control spool tothe minimum fuel position. When the engine isoperating in this phase, smooth idling is obtainedthrough the interaction of the flyweights and idlingspring.

With the accelerator pedal lever against theadjustable idling stop, any small speed increase causesthe flyweights to exert a larger force on the slidingsleeve. This slightly compresses the idling spring and, asa result, the spool control lever moves the spool andreduces the fuel delivery.

Any slight drop in engine speed produces theopposite action, so smooth idling under governorcontrol is obtained.

Mid-range operationOnce the idling range has been exceeded, the largergovernor force puts the idling and starting springs outof action. At this stage the intermediate spring comesinto use to extend the idle control range and so smooththe transition from idle to mid-range operation. Theintermediate spring is stronger and provides a flexiblelink between the driver’s pedal and the control spoollever, so that, when the accelerator pedal is depressed, aslight delay in engine response is introduced.

Beyond this phase any movement of the acceleratorproduces a direct action on the control spool.

Maximum speedDuring mid-range operation, the pre-load of the maingovernor spring causes the spring assembly to act as asolid block. However, when the engine reaches itspredetermined maximum speed, the force given by theflyweights equals the spring pre-load. Any further speedincrease allows the flyweights to move the spool controllever. This reduces the quantity of fuel being deliveredand so keeps the engine speed within safe limits.

4.2.2 InjectorsThe purpose of the injector is to break up the fuel to therequired degree (i.e. to atomise it) and deliver it to thecombustion region in the chamber. This atomisationand penetration is achieved by using a high pressure toforce the fuel through a small orifice.

Many vehicles use a type of injector thatincorporates a valve. The closed system is responsive topump pressure; raising the pressure above apredetermined point allows the valve to open, and stayopen until the pressure has dropped to a lower value.The ‘snap’ opening and closing of the valve givesadvantages, which make this system popular.

The complete injector, shown in Fig. 4.8a, consistsof a nozzle and holder, which is clamped to form a gas-tight seal in the cylinder head. A spring, compressed byan adjusting screw to give the correct breaking(opening) pressure, thrusts the needle on to its conicalseat. Fuel flows from the inlet nipple through a drillingto an annular groove about the seat of the needle. Athrust, caused by fuel acting on the conical face X, willovercome the spring and lift the needle when thepressure exceeds the breaking pressure. The opening ofthe valve permits discharge of fuel until the pressuredrops to the lower limit. Any fuel which flows betweenthe needle and body acts as a lubricant for the needlebefore being carried away by a leak-off pipe.

4.2.3 Injector nozzle typesThere are three main types of nozzle:

● single hole● multi-hole● pintle.

Single hole nozzleSee Figure 4.8b. A single orifice, which may be as smallas 0.2 mm (0.008 in), is drifted in the nozzle to give asingle jet form of spray. When this nozzle is used withindirect injection systems, a comparatively low injectionpressure of 80–100 bar is used.

Multi-hole nozzleSee Figure 4.8c. Two or more small orifices, drilled atvarious angles to suit the combustion chamber, producea highly atomised spray form. Many engines with directinjection systems use a four-hole nozzle with a highoperating pressure of 150–250 bar. A long stem versionof this type makes it easier to fit the injector in the head.

Pintle nozzleSee Figure 4.8d. Swirl chambers can accept a soft formof spray, which is the form given by a pintle nozzlewhen it is set to operate at a low injection pressure of110–135 bar.

A small cone extension on the end of the needleproduces a conical spray pattern and increases thevelocity of the fuel as it leaves the injector. This typetends to be self-cleaning.


[647] Chapter 04 25/8/06 15:39 Page 169

The elimination of heater plugs on some small indirectinjection engines has been made possible by theinvention of a special pintle nozzle known as the‘pintaux’ type, as shown in Figure 4.8e. Startingconditions produce a small needle lift, and so fuelpasses through the small auxiliary hole and is directedto the hottest part of the chamber. Under normalrunning pressures, the full lift of the needle dischargesthe fuel through the main orifice.

4.2.4 Rotary pumps with electroniccontrol

With an ever increasing demand on the compressionignition engine to develop more power with loweremissions, together with an increase in fuel economy,electronic control of the diesel fuel system has nowbecome the standard for passenger vehicles with dieselengines.

Although the very latest generations of electronicdiesel systems are in fact very similar to petrol injectionsystems, i.e. the injectors are directly controlled by thesystem ECU (section 4.3), technicians may stillencounter early generations of electronic diesel systems

where the electronic control influenced the operation ofa rotary pump. One example of this type of electroniccontrol is therefore detailed below and illustrated inFigure 4.9. The general term used to describe thesesystems is electronic diesel control (EDC).

An electronic diesel control system can give thefollowing advantages:

● lower emissions● lower soot emissions● increased engine output.

The non EDC Bosch VE pump accurately controls thequantity of fuel delivered by the injectors with the useof the control spool as well as a governor and anautomatic advance unit. However, external influences,such as engine temperature and air density, will affectthe engine performance and also the emissions. Precisecontrol of the fuel system can be achieved with the useof electronic diesel control.

The EDC electronic control unit (ECU) controls thefuel system by using two actuators, a solenoid operatedcontrol spool and a solenoid operated timing advanceunit, which are located in the distributor pump (Figure4.9). The pump uses many of the components that arefitted to the VE-type distributor pump, including thefuel shut-off valve and the fuel delivery plunger.

The ECU monitors the engine operating conditionsfrom information supplied by sensors and provides thecorrect control signals to the actuators, giving precisecontrol of the fuel delivered to the injectors. The EDCsystem uses sensors very similar in operation to thoseused with petrol fuel injection systems (see Figure 4.10).

An accelerator cable between the throttle pedal andthe distributor pump is no longer required to control thefuel volume. The position of the throttle pedal ismonitored by the EDC ECU with the use of a throttleposition sensor fitted to the throttle pedal linkage. TheECU controls the volume of fuel delivered to theinjectors by using a solenoid operated control spool.

The engine speed is monitored by a sensor fitted tothe engine crankshaft; the sensor is usually of theinductive type. An additional sensor is fitted to thedistributor pump, which monitors the speed andposition of the fuel control spool in relation tocrankshaft position. The ECU uses the information fromthese sensors, together with additional sensorinformation to determine the volume of fuel and fuelinjection timing.

A manifold absolute pressure (MAP) sensor enablesthe ECU to monitor the volume of air entering theengine. The ECU calculates the air density from theMAP sensor signal in conjunction with the intake airtemperature sensor signal. The MAP sensor signal isalso used to monitor and control the turbo boostpressure. The ECU controls the turbo boost pressurewith a waste gate actuator solenoid.

Two temperature sensors are used: an engine coolanttemperature sensor to monitor engine temperature andan intake air temperature sensor. The ECU uses


Figure 4.8 Injectors

[647] Chapter 04 25/8/06 15:39 Page 170

temperature information for fuel volume control. Thisinformation is also used to control the length of timethat the glow plugs operate during starting.

An injector motion sensor is fitted to one of theinjectors (Figure 4.11), usually to number 1 cylinder. Atthe start of fuel injection, when the fuel pressureincreases and lifts the injector valve from the seat, thesensor produces a signal. The start of injectioninfluences engine starting, combustion noise, fuelconsumption and emissions. The ECU monitors thesensor signal and determines, in conjunction with the

engine speed sensor information, the fuel injectiontiming control.

To enable the modern diesel engine to meetemission regulations, many engines are fitted with anexhaust gas recirculation (EGR) system. During certainengine operating conditions, the exhaust gases aremixed with the fresh air in the induction system, whichlowers the combustion temperature, thus reducing theharmful emissions produced by the engine. The volumeof EGR is measured with a mass air flow sensor, either ahot wire or a hot film type. The ECU controls the EGR


Injector needle motion

Accelerator pedal position sensor

Vehicle speed

Manifold absolute pressure sensor

Engine speed

Air mass sensor

Control spool position sensor

Temperature sensors coolant/air

EGR solenoid

Glow plug control module

Control spool solenoid

Timing advance unit solenoidEDC ECU

Figure 4.9 An electronically controlled diesel rotary pump (EDC)

Figure 4.10 Inputs and outputs for an EDC system

[647] Chapter 04 25/8/06 15:39 Page 171

valve actuator accordingly to ensure the correct volumeof exhaust gases are recirculated to provide the correctemission levels.

The position of the control spool in relation to thedistributor plunger determines the volume of fueldelivered to each injector, in the same manner aspreviously described with the Bosch VE pump. Thevolume of fuel delivered dictates the engine speed andengine power. A mechanical governor is no longer fittedto the distributor pump; the position of the controlspool is electronically controlled by the EDC ECU with asolenoid. Depending on the position of the spool, thevolume of fuel is either increased or decreased. Theposition of the spool can be altered to providemaximum fuel for full load through to zero fuel toprevent fuel from being supplied to the injectors. Theexact position of the control spool is monitored by theECU with a position sensor.

As with the VE pump, the fuel pressure inside thepump is relative to engine speed. The timing advanceunit functions in a similar manner to that of the VE

pump, except that the fuel pressure applied to theadvance unit is controlled by the EDC ECU with the useof the timing advance unit solenoid. The fuel injectiontiming can either be advanced or retarded by alteringthe control signal to the solenoid.

The EDC ECU controls the engine idle speed bycontrolling the volume of fuel delivered. To ensure thatthe engine idle is as smooth as possible, the ECU willslightly vary the volume of fuel to each cylinder by thecorresponding amount.

The EDC ECU also incorporates a diagnosticfunction similar in operation to that of a petrol enginemanagement system. If a fault occurs with the system,the ECU will if possible operate with a limited operatingstrategy (LOS). If a sensor circuit fails, the ECU willsubstitute the value of the sensor circuit, to providelimited emergency operation of the system. If the ECUdetects a system fault, it illuminates a warning lamp inthe instrument panel to alert the driver that a fault hasoccurred; the fault will also be stored in the memory ofthe ECU in the form of a code. To diagnose the systemfault, the fault information can be retrieved from theEDC ECU memory with the appropriate diagnostic testequipment.

Many modern vehicles are prevented from beingdriven by the fitting of an engine immobiliser system.Early immobiliser systems prevented diesel enginesfrom being started by isolating the power supply to thedistributor pump stop solenoid, preventing fuel fromentering the plunger. Modern electronically controlleddiesel fuel systems are immobilised within the ECU. Ifthe ECU receives an incorrect immobiliser code from thedriver, it prevents fuel from being supplied to theinjectors by isolating the control signals to thedistributor pump solenoids.

Electronic diesel control (EDC) systems can varytiming and fuel quantity by acting on theautomatic advance unit and the control spool

Inputs to an EDC system are similar to those usedfor petrol/gasoline engine management

Key

Poin

ts


Figure 4.11 An injector with a motion sensor

4.3 COLD-START PRE-HEATING SYSTEMS

4.3.1 Cold startingDirect injection enginesAll compression ignition engines require some specialprovision for cold starting, although modern directinjection engines (injecting directly into the maincombustion chamber, as opposed to into a pre-chamber)may require cold-start assistance only at low ambienttemperatures. The heat generated during compression,even under cranking conditions, is usually sufficient tocause ignition of the vaporised fuel.

For most cold-start conditions, the direct injectionengines used in most modern passenger cars are able tostart relatively easily: the injection of a larger quantityof fuel (a rich mixture), and the greater amount ofeasily ignitable fractions contained in the injectedcharge, are generally sufficient to start a cold directinjection engine. However, with many modern directinjection engines, assistance during cold starting isneeded to reduce harmful emissions when the engine isinitially started.

[647] Chapter 04 25/8/06 15:39 Page 172

operate for is usually dependent on enginetemperature: the colder the engine, the longer the glowplugs function.

The glow plugs are usually controlled automaticallyby a timer relay or an electronic control unit (ECU).When the ignition is switched on, the controller usuallyilluminates a glow plug warning light in the instrumentpanel to warn the driver when the glow plugs areoperating. Modern direct injection engines have a glowplug fitted to each cylinder. The glow plugs used onmany modern diesel engines can remain switched on fora few minutes after the engine has started, ‘post-glow’,normally with a reduced electrical current to prevent theglow plugs from overheating and burning out. Thisprovides additional heat in the combustion chamber toimprove the combustion process and therefore loweremissions when the engine is started from cold.

Modern diesel injection systems are electronicallycontrolled, so it is convenient and more effective for thediesel fuel system ECU also to control the glow plugoperation. The ECU is already receiving informationfrom the temperature sensors and other sensors so itcan control the glow plug operating time, the currentflow (reduced current and heating after starting), andthe warning light (the indication to the driver when theengine can be started).

Modern glow plugs have heating elements that areeffectively resistances with a positive temperaturecoefficient or PTC; the PTC increases the resistance ofthe heating element as the element temperature rises,thus progressively reducing the current flow. In effect,the current flow is self-regulated to prevent overheatingof the heating element, while still allowing an initialhigh current to heat up the element for the cold start.

Figure 4.13 shows a glow plug in a combustionchamber of a direct injection engine. Figure 4.14 showsthe construction of a modern glow plug and Figure 4.15shows the arrangement for controlling the operation of

Cold-start pre-heating systems 173

Figure 4.12 Indirect diesel injection into a pre-chamber

Indirect injection enginesWith indirect injection (Figure 4.12), the pre-chamber isnot exposed to the same amount of heat as the maincombustion chamber (heat is lost to the cylinder wallsand combustion chamber walls). These greater heatlosses therefore dictate that indirect injection engineshave extra provision to ensure ignition of the fuelduring cold starting. A significant design difference isthat higher compression ratios are used for indirectinjection engines: ratios of about 16:1 are used withdirect injection engines, while indirect injection enginesuse higher ratios of the order of 22:1, and in some casesa ratio as high as 30:1 is used. These highercompression ratios increase the heat produced duringcompression, which aids cold starting. A highcompression ratio is also used in indirect injectionengines to raise their thermal efficiency, and henceeconomy, unlike direct injection engines; this tends tocounteract the greater heat loss caused by the largersurface areas of an indirect injection combustionchamber.

4.3.2 Cold-start assistanceManifold heatersSeldom used on modern passenger car engines,manifold heaters are electrical units fitted to pre-heatthe air as it passes through the inlet manifold to thecylinder.

Pintaux injectorA Pintaux injector is a pintle injector which has anauxiliary hole to direct fuel down the throat of the pre-chamber during the cranking period (see Figure 4.8e).This type of injector is suitable for indirect injectionengines.

Heater plugStill the most widely used form of cold-start assistanceon diesel engines, the ‘glow plug’ or ‘heater plug’ isfitted in the combustion chamber and is effectively anelectric heater that can be used during cold starting andin the early phases of cold running. When the air iscold, the air in the combustion chamber is heated by anelectrical heating element for a few seconds prior tostarting a cold engine. The time that the glow plugs

Figure 4.13 Glow plug location on a direct injection enginecombustion chamber

[647] Chapter 04 25/8/06 15:39 Page 173

the glow plug. Note that in the example (Figure 4.15),the ECU (which is receiving information from thetemperature sensors) is effectively controlling the ‘glowplug control unit’; the control unit regulates the highcurrent passing to the glow plugs.

Note: Modern electronic diesel systems operate in avery similar way to electronic petrol injection systems,especially direct petrol injection. See sections 3.1 to 3.4for information on sensors, actuators and controlsignals for electronic injection systems. Chapters 1 and2 include additional information about sensor andactuator operation.

4.4.1 Advantages and disadvantagesof direct injection

Direct injection systems have been used on larger dieselengines for many years, especially for heavycommercial applications. Since the 1980s, lightpassenger cars have also increasingly been fitted withsmaller direct injection engines. Direct injection is amore efficient method of fuel delivery: it develops morepower and has a lower fuel consumption than indirectinjection (where the fuel is delivered into a pre-combustion chamber). Direct injection does, however,have one major disadvantage: the combustion noise ishigher than that of indirect injection, which isundesirable in passenger motor vehicles.


Figure 4.15 Glow plug control

4.4 ELECTRONIC CONTROL OF DIESEL INJECTION (COMMON RAIL SYSTEMS)

The combustion noise is generally referred to ascombustion knock and is caused by ignition of the fuelafter injection has initially started. The short delaybetween ‘start of injection’ and the ignition of the fuelmeans that there is a relatively large quantity of fuelthat initially ignites; this causes a rapid combustion andpressure rise in the early phases of the combustionprocess, which causes an audible knock. On most fuelinjection systems used up until the late 1990s, whichwere relying on mechanical pumps to generate thepressure for injection, it was relatively difficult toprecisely control the fuel quantity delivered during theearly injection phase. If the initial quantity of fuelinjected is too large, ignition is rapid and initial gasexpansion is rapid, thus causing the combustion knock.

Therefore an objective with the modern directinjection diesel fuel system is to control the initialinjection quantity in what is termed the ‘pilot injectionphase’, which is when a small quantity of fuel isinjected, ahead of the main injection period. This smallquantity of fuel causes a small rise in pressure duringthe pilot phase, thus reducing the rapid speed ofcombustion, which in turn reduces combustion knock.

1 Sheathed-element glow plug2 Glow control unit3 Glow-plug and starter switch4 To battery5 Indicator lamp6 Control line to the engine ECU7 Diagnosis line

Most diesel systems use a glow plug in thecombustion chamber to help increase thetemperature of the injected fuel

Glow plugs are controlled by a timer relay or anECU so that the optimum heat time is useddepending on the engine temperature

Key

Poin

ts

Figure 4.14 Glow plug construction

1 Electrical connector terminal2 Insulating washer3 Glow plug shell4 Glow tube5 Control filament6 Filling powder7 Helical heating wire8 Heater element gasket9 Double gasket

10 Round nut

[647] Chapter 04 25/8/06 15:39 Page 174

4.4.2 Common railCommon rail diesel fuel systems have been widely usedin commercial vehicle and large diesel engineapplications for many years. Common rail refers to asystem whereby the injectors receive fuel from acommon supply rail, which is fed fuel under pressure byan engine driven pump. On early common rail systemsthe pump provided fuel at a relatively low pressure; thefuel was then passed to the injectors through a pipe (oreven through holes or tubes built into the cylinderhead). The injectors contained a pumping element(usually driven from the engine camshaft via a rockersystem), which produced the high pressure necessaryfor injecting the diesel fuel. Some common rail systemsdelivered fuel to a separate pumping element whichthen passed the fuel at high pressure to the injector.

One advantage of using a common delivery rail isthat the high pressure pumping element is located in orclose to the injector, so the high pressure created by thepumping element does not have to pass through a longdelivery pipe, which is the case for traditional dieselfuel pump systems, where the pump is a considerabledistance from the injectors. With a long delivery pipecarrying high pressure, when the pump delivers the fuelit causes a pressure wave to travel along the deliverypipe (which is full of fuel); the time delay in the highpressure wave reaching the injector causes timinginaccuracies of injector opening and closing. With theshort or non-existent high pressure delivery pipe oncommon rail systems, this delayed pressure waveproblem is eliminated or reduced: see the followingparagraphs dealing with unit pumps and unit injectors.

Unit pumps and unit injectorsFigure 4.16 shows a relatively recent type of ‘unitinjector’, where the fuel injector contains a pumpingelement that is driven by a cam and rocker system. Thecam lobe can be part of the normal camshaft used tooperate the inlet and exhaust valves. The injector is fedwith fuel at a relatively low pressure from a commonsupply rail (feeding all injectors). The high pressure isthen created by the pumping element in the injectoritself. Unit injectors can deliver fuel at typical pressuresof 2000 bar.

With this type of unit injector, a solenoid attached tothe injector controls a valve arrangement that opens orcloses an outlet or spill port. When the outlet port isopen, the pumping element will still function and buildup pressure, but the fuel will pass straight out of theoutlet port back to the low pressure fuel system. Whenthe outlet port is closed by the solenoid, this will causepressure to build up above the injector nozzle (due tothe action of the pumping element). The pressure build-up will then cause the injector nozzle to open anddeliver fuel. At the appropriate time, the solenoid willopen the outlet port again, which will cause animmediate drop in pressure above the nozzle. This willthen allow the spring in the injector nozzle to return to

the closed position (a closing spring is located in thenozzle, as in older injectors – see Figure 4.8).

The solenoid is controlled by an ECU, whichfunctions in much the same way as a petrol injectionsystem ECU: i.e. the ECU receives information fromsensors and is therefore able to control the opening andclosing of the injector, thus controlling fuel quantity(the injection duration).

Figure 4.17 shows a similar arrangement to that ofthe unit injector but the pumping element is separate

Electronic control of diesel injection (common rail systems) 175

Figure 4.16 Unit injector with combined pumping element for acommon rail system

1 Actuating cam 3 High pressure solenoid valve2 Pump plunger 4 Injection nozzle

Figure 4.17 Unit pump system for a common rail system

1 Injection nozzle 4 High pressure solenoid valve2 Nozzle holder 5 Pump plunger3 High pressure line 6 Actuating cam

[647] Chapter 04 25/8/06 15:39 Page 175

from the injector; a short delivery pipe is thereforerequired to deliver the high pressure fuel from thepumping element to the injector. This type of system isgenerally referred to as a ‘unit pump’ system and issuited to engines where the camshaft location may notallow a combined unit injector and pumping elementto be used.

For both the unit injector and the unit pump systems,the common rail supply (from the low pressure pump)can be via a low pressure delivery pipe or through a holeor port system built into the cylinder head.

4.4.3 Electronically controlledcommon rail systems using asingle high pressure pump

The logical evolution of the common rail system is to usea single high pressure pump feeding the common supplyrail. The injectors will therefore not contain a pumpingelement (or require separate individual pumpingelements), but can still be controlled using a solenoidwhich regulates the outlet port of the injector. Figure4.18 shows the general layout of this type of system.Note that this type of system is virtually identical inprinciple to a ‘direct injection’ petrol injection system(section 3.4). Although fuel at high pressure will passfrom the common fuel rail to the injectors, the openingand closing of the injectors is not dependent on pressurewaves passing through the pipe: it is totally dependenton the solenoid action, which causes the injector outletport to open or close (as on the unit injector).

With this type of common rail system, as well ascontrolling the injector timing and injector duration(fuel quantity control), the pressure at which the fuel is

injected into the combustion chamber can also bealtered to suit the engine operating conditions andcylinder pressure. The fuel delivered to the commonfuel supply rail (by the engine driven high pressurepump) can be monitored by a pressure sensor andcontrolled using a pressure regulator. The pressure ofthe fuel delivered to the injectors can therefore becontrolled so that it is always at the desired value.Typical injection pressures are around 1600 bar.

With electronic control, the injection timing can beaccurately controlled to allow the fuel to ignite andburn correctly within the combustion chamber.Conventional diesel fuel systems inject the total volumeof fuel required by the cylinder during one injectoropening. For this type of common rail system withhigher fuel pressure, the volume of fuel can be injectedinto the combustion chamber in stages: a pilotinjection, main injection and sometimes a postinjection. The pilot injection period can be controlled,so that only very small quantities of fuel are injected,thus reducing combustion knock. Post injection can beused to aid the control of emissions: a small quantity offuel is injected at the end of the power stroke or even onthe start of the exhaust stroke, the fuel vaporises andpasses through to an NOx catalyst which then reducesNOx emissons.

Using the information from the various systemsensors, the ECU determines the volume of fuel and thepoint in time at which the fuel is to be injected toprovide the required power from the engine. Figure 4.19shows a typical single high pressure pump common railsystem (sometimes referred to as a common railaccumulator system), whilst Figure 4.20 shows acomplete system with sensors and actuators; note thatthe illustration shows a turbocharged engine, where theboost pressures are also controlled by the ECU.

4.4.4 Fuel pressure systemLow pressure systemWith a low pressure system, the fuel pressure isproduced by a low pressure pump that supplies fuelfrom the fuel tank via a fuel filter to a high pressurepump. The ECU controls the operation of the lowpressure pump. The pump location and design are verysimilar to those of a petrol fuel injection system. Figure4.21a shows the layout of the fuel system. Figure 4.21bshows a typical low pressure fuel pump. Figures 4.21cand 4.21d show roller cell and gear type low pressurepump elements. Note that a conventional fuel filter isfitted in the low pressure system (Figure 4.22).

High pressure systemThe high pressure pump, normally driven at halfcrankshaft speed by the camshaft, generates the highfuel pressure which is stored in the common fuel rail,hence the name ‘common rail’ (Figure 4.23). The ECUvaries the pressure produced by the high pressure pump


Figure 4.18 Electronically controlled common rail system using asingle high pressure pump

1 High pressure pump 3 High pressure solenoid valve2 Rail (high pressure 4 Injector

fuel acccumulator) 5 Injection nozzle

[647] Chapter 04 25/8/06 15:39 Page 176

up to a maximum pressure of typically 1300 to 1600 barwith electrically operated solenoid valves within thepump assembly. The ECU varies the fuel pressureaccording to engine operating conditions; the fuelpressure is not relative to engine speed (apart fromduring engine cranking). The diesel fuel lubricates theinternal cams and plungers of the high pressure pump.

The high pressure pump (Figure 4.24) is driven bythe engine and is usually located in the same position asa traditional diesel pump. The fuel pressure ismonitored by the ECU with a fuel pressure sensorsituated in the common rail (Figure 4.23), so the fuel isalways delivered at the correct pressure to suit theengine operating conditions. The fuel passes from thefuel rail to the injectors through metal fuel pipes. Thesepipes are approximately the same in length andmanufactured without excessively sharp bends whichmight restrict fuel flow. Note that if any of the fuel pipesare disconnected during service or repair, they shouldbe renewed. The pipes are made from steel whichdeforms thus providing some flexibility for fitting andto allow for vibration as well as small movements thatwill occur due to engine expansion caused by heat. Thehigh pressure pump connections, when tightened,ensure a fuel tight seal.

The common rail acts as an accumulator or reservoirof fuel, damping pressure fluctuations in the highpressure system due to the pumping action andinjection. The fuel rail is also fitted with a fuel pressureregulator (Figure 4.25), so if the fuel pressure becomesabnormally high, the excess pressure will pass through

the pressure limiting valve and return to the fuel tank. Amechanical fuel pressure limiting valve was used onearly common rail fuel systems. With later systems, theECU controls the fuel rail pressure with an electricallyoperated solenoid valve. Note that the ECU monitorsthe fuel pressure in the fuel rail and controls thepressure with a solenoid valve on the side of the highpressure pump.

4.4.5 Fuel injection system(See Chapters 1 and 2 for information on electronicsystem sensors and actuators, and on electronicactuator control signals.)

The ECU controls the injectors by making use of asimilar principle to that of petrol fuel injection. Thecommon rail fuel system uses many of the sensors thatprovide information for electronically controlleddistributor pump diesel fuel systems and for petrolinjection systems. These include:

● an engine speed sensor fitted to the crankshaft● a camshaft position sensor● an accelerator pedal position sensor● a MAP sensor● an engine coolant and intake air temperature sensor● an air mass sensor.

These sensors are used to monitor engine operatingconditions. The accelerator sensor provides the ECUwith driver requirements: whether the driver wishes to


Figure 4.19 Common rail accumulator injection system

[647] Chapter 04 25/8/06 15:39 Page 177


Figure 4.20 A complete common rail injection system (single high pressure pump system)

[647] Chapter 04 25/8/06 15:39 Page 178


Key to Figure 4.20

Engine, engine management, and high pressure fuel injection components17 High pressure pump18 Metering unit25 Engine ECU26 Fuel rail27 Rail pressure sensor28 Pressure control valve (DRV 2)29 Injector30 Glow plug31 Diesel engine (01)M Torque

A Sensors and setpoint generators1 Pedal-travel sensor2 Clutch switch3 Brake contacts (2)4 Operator unit for vehicle speed controller (cruise control)5 Glow-plug and starter switch (‘ignition switch’)6 Road speed sensor7 Crankshaft speed sensor (inductive)8 Camshaft speed sensor (inductive or Hall sensor)9 Engine temperature sensor (in coolant circuit)

10 Intake air temperature sensor11 Boost pressure sensor12 Hot-film air mass meter (intake air)

B Interfaces13 Instrument cluster with displays for fuel consumption,

engine speed, etc.14 Air-conditioner compressor with operator unit15 Diagnosis interface16 Glow control unitCAN Controller Area Network(on-board serial data bus)

C Fuel-supply system (low-pressure stage)19 Fuel filter with overflow valve20 Fuel tank with pre-filter and Electric Fuel Pump,

EFP (presupply pump)21 Fuel-level sensor

D Additive system22 Additive metering unit 23 Additive control unit 24 Additive tank

E Air supply32 Exhaust gas recirculation cooler33 Boost pressure actuator34 Turbocharger (in this case with variable turbine

geometry (VTG))35 Control flap36 Exhaust gas recirculation actuator37 Vacuum pump

F Exhaust-gas treatment38 Broadband lambda oxygen sensor, type LSU39 Exhaust gas temperature sensor40 Oxidation type catalytic converter41 Particulate filter42 Differential pressure sensor43 NOx accumulator type catalytic converter44 Broadband lambda oxygen sensor, optional NOx sensor

[647] Chapter 04 25/8/06 16:46 Page 179


Figure 4.21 Low pressure fuel system and low pressure pumps

(a)

(b)

(c)

(d)

Key to Figure 4.21a

1 Fuel tank2 Pre-filter3 Presupply pump4 Fuel filter5 Low-pressure fuel

lines6 High-pressure pump7 High-pressure fuel

lines8 Fuel rail9 Injector

10 Fuel return line11 Fuel-temperature

sensor12 ECU13 Sheathed-element

glow plug

[647] Chapter 04 25/8/06 15:39 Page 180


Figure 4.22 Fuel filter

Figure 4.23 High pressure fuel system

accelerate, decelerate or allow the engine to idle whilststationary. The ECU uses the sensor information tocalculate the desired fuel pressure, injection volumeand duration to produce the required engine powerand torque.

The fuel within the common rail is at a constantpressure during injection and therefore the volume offuel injected is also constant during the injector openingperiod. Therefore a precise volume of fuel can bedelivered during the injector opening period.

The EDC ECU determines the injector opening timeperiod (injector duration) from sensor information andprovides a control signal to the injector accordingly. Thehigh fuel pressure exerts a great force at the injectorneedle valve, and therefore a very high voltage and

current are required to initially open the injector. Theinjector driver control module provides the necessaryhigh voltage control signal to the injector. The modulemight be located within the ECU, or in some cases fittedas a separate unit. The ECU uses the engine speedsensor to provide the timing control for each injector.Additional information is required to synchronise eachinjector with the cylinder cycle. A cylinder recognitionsensor monitors the camshaft position, which providesthe ECU with the information necessary to control thephasing of the injectors. The injectors are situated in thecylinder head and spray fuel into the swirling air withinthe combustion chamber, which is normally integratedinto the crown of the piston.

If the current is switched off to the solenoid circuitthe injector is not energised and the injector needlevalve is closed, which prevents the pressurised fuelleaving the injector nozzle (Figure 4.26). The highpressure fuel is applied to the needle valve at the lowersection of the injector and also a control chamber whichis located on top of the injector needle valve within thetop section of the injector. The pressure of the solenoidspring and the needle valve spring is higher than thefuel pressure applied and therefore the needle valveremains closed.

The ECU determines the injection period duringwhich the injector opens and injects a volume of fuelinto the combustion chamber. The ECU provides theinjector with a control signal that energises the injectorsolenoid (Figure 4.26). The solenoid valve lifts,allowing the fuel pressure to escape from the controlchamber into the chamber above. The fuel passing tothe chamber returns to the tank via the fuel returnsystem. The initial current required to lift the solenoid ishigh, because of the pressure of the spring. Once thesolenoid is open, a smaller current is required tomaintain the solenoid position: the ECU applies aholding current.

An orifice restriction prevents the high pressure fuelfrom rapidly re-entering the control chamber; thecontrol chamber pressure is lower than the fuel pressure

1 Fuel rail2 Pressure

control valve3 Return line

from fuel rail tofuel tank

4 Inlet from highpressure pump

5 Rail pressuresensor

6 Fuel line toinjector

[647] Chapter 04 25/8/06 15:39 Page 181


Figure 4.24 High pressure pump

1 Drive shaft2 Eccenter3 Pump element with

pump plunger4 Inlet valve5 Outlet valve6 Fuel inlet

1 Flange 6 Return connection2 Pump housing 7 Pressure control valve3 Engine cylinder head 8 Barrel bolt4 Inlet connection 9 Shaft seal5 High pressure inlet 10 Eccentric shaft

[647] Chapter 04 25/8/06 15:39 Page 182


Figure 4.25 Fuel pressure regulator

Figure 4.26 Common rail injector

1 Electrical connections2 Valve spring3 Armature4 Valve housing5 Solenoid coil6 Valve ball7 Support ring8 O-ring9 Filter

10 High pressure fuel supply11 Valve body12 Drain to low pressure circuit

[647] Chapter 04 25/8/06 16:47 Page 183

pressure increases is less severe, resulting in a reductionin combustion noise, lower fuel consumption and loweremission levels.

The length of time that the injector is opened(injector duration), together with the pressure at whichthe fuel is injected, dictates the volume of fuel deliveredto the cylinder. It should be noted that althoughchanging the duration of injector opening time wouldaffect the volume of fuel delivered, an increase in thefuel pressure is generally used as the primary means toincrease the volume of fuel delivered to a cylinder. Athigh engine speeds, insufficient time exists between thestages of injection and it is not possible to provide pilotinjection. The ECU combines both pilot and main stagesof injection and uses a single injector opening period toinject the volume of fuel required.

The ECU monitors any imbalance between thetorque generated between cylinders. After eachinjection period, the power stroke occurs, whichaccelerates the speed of the crankshaft. The ECUmonitors the acceleration speed of the crankshaftthrough the engine speed sensor signal. If all cylindersare producing an equal amount of power, theacceleration of the crankshaft between each cylinderpower stroke should also be equal. Engine wear willaffect the power produced by each cylinder, and the


Figure 4.27 Combustion pressures a without pilot injectionb with pilot injection

applied to the needle valve. The difference in fuelpressure between the control chamber and at the needlevalve causes the valve to lift from the seat and fuel isexpelled through the injector nozzle into thecombustion chamber. The pressure of the injected fuel isequal to the pressure in the fuel rail. The high fuelpressure, together with the design of the injectornozzle, allows excellent atomisation of the fuel injected,which promotes good mixing of the air and fuel withinthe combustion chamber. Thoroughly mixing the airand fuel reduces hydrocarbon and soot emissions.

To end the injection of fuel, the ECU switches off thecurrent flow through the injector solenoid circuit,allowing the solenoid plunger and valve to return to itsseat. The closing of the solenoid valve allows thecontrol chamber to refill with high pressure fuel fromthe fuel rail. The high fuel pressure in the controlchamber, together with the force of the needle valvespring, exerts a greater force than that of the high fuelpressure at the base of the needle valve, so the needlevalve returns to its seat and injection ceases.

Pilot injectionEarlier designs of diesel fuel systems (in-line and rotarypump systems) generally inject the total volume of fuelduring one injector opening period for one cylindercycle. There is a time period between the start ofinjection and the start of ignition of the fuel. When thefuel ignites, the cylinder pressure rapidly increases,which pushes the piston down the cylinder. The sharprise in cylinder pressure is heard and referred to asdiesel or combustion knock.

The common rail fuel system normally injects thetotal volume of fuel (for the combustion process) in twoinjection stages, often referred to as pilot injection andmain injection.

A small volume of fuel is injected before the pistonreaches TDC. This small volume of fuel, typicallybetween 1 and 4 mm3 is used to condition the cylinderbefore the main volume of fuel is injected. The pilotinjection raises the cylinder pressure slightly due to thecombustion of the fuel: therefore the temperaturewithin the cylinder also rises. If the pilot injectionoccurs too early in the compression stroke, the fuel willadhere to the cold cylinder walls and the crown of thepiston, increasing the hydrocarbons and soot in theexhaust gases.

Figure 4.27 shows the difference in the combustionchamber pressure rise when pilot injection is usedcompared with when there is no pilot injection. Notethe steeper rise in pressure that occurs just after TDCwhen there in no pilot injection; it is this steep pressurerise that is creating the diesel or combustion knock.

Main injectionThe time delay between the points at which the fuel isinjected and ignited is reduced because the pilotinjection provides a slightly higher cylinder temperatureand pressure. The rate at which the combustion

[647] Chapter 04 25/8/06 15:39 Page 184

ECU can alter the fuel volume and injection timing tocylinders to equalise the power each cylinder producesat low engine speeds. Unequal power between cylindersis very apparent at idle, but the ECU stabilises theengine speed, ensuring a smooth engine idle.

Turbocharger boost pressure controlIf a turbocharger is fitted, the ECU controls theturbocharger boost pressure. The ECU monitors theinlet manifold pressure with a pressure sensor. If thepressure is too high (over-boost) the ECU regulates thepressure by using a waste gate in the exhaust manifold.Some later vehicles are fitted with a variable geometryturbocharger, where the ECU alters the geometry insidethe exhaust turbine to vary the boost.

Exhaust gas recirculationThe ECU controls exhaust gas recirculation (EGR). Thisreturns some of the exhaust gases into the inductionsystem to reduce the harmful emissions emitted from theexhaust, i.e. oxides of nitrogen or NOx (see section 3.5).The ECU monitors the air mass sensor signal situated inthe air induction system, and the sensor provides anindication of the volume of exhaust gas recirculated.

Unlike earlier generations of diesel engine, manyinduction systems are fitted with a throttle plate in theinduction system. When the throttle plate (butterfly) isused in a petrol engine, it alters the air volume enteringthe engine and therefore alters engine power. However,a throttle plate in a diesel engine is used to alter the rateof EGR. At low engine speeds, the angle of the throttleplate is adjusted to provide a depression in themanifold, which induces the rate of EGR. At highengine speeds and loads the throttle plate is fully open


to prevent restriction to the flow of air into the engine.The throttle plate is either operated by a stepper motoror by the modulation of a vacuum switching valve.

Sensors used in common rail injection systems arevery similar to those used for petrol/gasolineinjection

A high pressure engine-driven pump supplies fuelto electronically controlled injectors

Common rail systems can operate at pressures upto 1600 bar

Web links

Engine systems informationwww.bosch.comwww.sae.orgwww.imeche.org.ukwww.picotech.comwww.autotap.comwww.visteon.com www.infineon.comwww.kvaser.com (follow CAN Education links)



Key

Poin

ts

[647] Chapter 04 25/8/06 15:39 Page 185

TRANSMISSION

Cha

pter

5

5.1 PURPOSE OF THE TRANSMISSION SYSTEM


Purpose of the transmission system

Transmission types

History of electronic control

Multiplexing

Electronic transmission control as a black box

Sensors and actuators used in transmission systems

Clutch electronic control

Manual gearbox electronic control

Torque converter electronic control

Automatic gearbox transmission management (epicyclic, fixed gear and CVT)

Light hybrid powertrain technology (starter–generator)

Electronic differential and four-wheel drive control

Transmission diagnostics

Future developments

Any vehicle equipped with a combustion engine as itsprime mover requires a transmission system to transmittorque at an appropriate speed to the driving wheels.Fundamentally, the transmission system is neededbecause internal combustion engines have quite alimited speed range at which useable torque isproduced (this varies, but generally lies in the range1500–5000 rev/min). Operation within this speedrange is also important for the engine to achievemaximum efficiency. This is clearly beneficial for fueleconomy and to minimise exhaust emissions. There areseveral other reasons why a transmission system mustbe incorporated in the powertrain:

● to provide variable torque at varying speedsselectable by the driver for the appropriatecondition, such as low speeds, overcominggradients, comfortable cruising speeds

● to provide a ‘neutral’ state, i.e. a situation where theengine can run without being connected to thedriving wheels, for example, when a vehicle isstationary

● so that the vehicle can be driven backwards (inreverse), for manoeuvring, parking, etc.

● to provide torque multiplication: the internalcombustion engine cannot produce any torque atzero speed. Therefore a transmission (including aclutch of some sort) is needed to overcome vehicleinertia (resistance to change of speed) at standstill.This provides a smooth application of tractive forcein a manner that can be controlled by the driver, andconsequent movement of the vehicle from zerospeed.

[647] Chapter 05 25/8/06 16:35 Page 186

Transmission types 187

The various types of vehicle powertrain transmissionscan be classified according to their operatingprinciples.

● Multi-stage transmissions have a number of fixedgear ratios which can be selected manually by thedriver, or automatically by a mechanical orelectrical control system according to the vehicleoperating status.

● Continuously variable transmissions (CVTs) areinfinitely variable between certain boundary limitsachieved through hydraulic or mechanical means.

They can also be classified by their construction.

● In-line transmissions have an input shaft on oneend and an output shaft at the other. They arepredominantly used in front engine, rear wheeldrive applications.

● Dual shaft transmissions have their input andoutput shafts misaligned or eccentric, typically forfront wheel drive applications.

Multi-stage transmissions rely on fixed, geometricallylocked elements (i.e. gears), whereas CVTs use frictionlocking principles to achieve the necessary ratios. Thisfriction locking function needs an additional energyinput (for example, from the oil pump used togenerate hydraulic pressure for gearbox operation inan automatic gearbox or CVT), which reduces theoverall efficiency of the gearbox itself. Thisinefficiency is offset by the fact that, because of the

infinitely variable transmission ratios, the engine canoperate closer to its maximum efficiency, whichincreases the efficiency of the powertrain as acomplete unit.

Another factor that differentiates transmissionssystems is their level of automation. In Europe,traditional manual transmissions are dominant,whereas in America or Asia, automatic shiftinghydraulic, or electrohydraulic, transmissions have thelargest share of the market. These traditionaltransmissions are now being displaced by moderntransmissions that incorporate the latestdevelopments in mechatronics and software such thatthey have the efficiency of traditional manualtransmissions with all the benefits of an automaticshifting transmission. In addition, they can operate inmanual or semi-automatic mode according to driverpreference and it is possible to integrate the shifting oftransmission ratios with other vehicle control or safetysystems for maximum driver benefit.

Table 5.1 summarises the common transmissiontypes and their characteristic features.

All vehicles with an internal combustion enginerequire a clutch and gearbox for reasons ofperformance and efficiency

The main types of gearbox are manual orautomatic and both types can be electronicallycontrolled

Key

Poin

ts

5.2 TRANSMISSION TYPES

Table 5.1 Transmission types

Transmission type Transmission Comparative Comparative Comparative Gearshift ratio via mass noise efficiencya comfort factorb

Manual, Fixed gears Low Low +10% –synchromeshAutomatic, Planetary gears and Medium Low 0% 0.90hydraulic shift torque converterCVT Belt/chain type drive High Medium +5% 0.95Toroidal variator drive Friction wheel variator Very High Low +7% 0.95(Torotrack)Automatic shifting Fixed gears, electro- Low Low +15% 0.63manual transmission mechanical actuationDual clutch Fixed gears, electro- Medium Low +8% 0.87transmission hydraulic actuation

a Efficiency when compared to an automatic, multistage transmission at given operating point with a petrol/gasoline engineb Measure of the quality of the change of transmission ratio: 1 = completely smooth, 0.1 = rough transition

[647] Chapter 05 25/8/06 16:35 Page 187

188 Transmission Fundamentals of Motor Vehicle Technology: Book 2

5.3.1 First developments in electricalcontrol for transmissions

Traditionally, transmissions and powertrains invehicles were purely mechanical systems. One of thefirst developments to incorporate an electrical controlelement was the overdrive system. This system wasfitted to many sporting or GT cars in the 1950s and1960s and was also available as an option on manyother car models during that period.

The overdrive was a self-contained unit, fitted to theexisting gearbox casing, providing an additional gearratio which could be engaged in third or fourth gear.The extra gear ratio was less than one, giving theoutput shaft a higher speed than the input shaft.(Usually, transmissions fitted at that time had a top gearratio of 1:1, known as ‘straight through drive’). Thus,with overdrive selected, for a given road speed, enginespeed was reduced and this provided more relaxedcruising and better economy. An inhibitor switch on thegearbox prevented engagement of overdrive in first andsecond gears.

The heart of the overdrive system was a singleepicyclic gear set (as used in many automaticgearboxes) with engagement effected via ahydraulically actuated cone clutch (Figure 5.1). Thisprovided the most redeeming feature of the system:even though the system provided an extra gear, it wasnot necessary to declutch (i.e. disconnect the enginetorque from the gearbox input shaft) in order to engageor disengage the extra gear. This provided improveddrivability and a sporty overtone to the vehicle. Thedriver could just ‘flick’ a switch to engage or disengagethe extra gear.

The engagement of the gear was implementedhydraulically via an oil pump to generate pressure,which caused an actuator to engage or disengage theclutch. The control of the hydraulic circuit wasimplemented electrically via a simple solenoid valve and

this is where the electrical control begins. The systemallowed the driver to engage overdrive via a switch onthe gear knob. This was a simple circuit, with aninterlocking switch, to ensure that overdrive could beselected only in third or fourth (top) gear.

This is a very simple control circuit (Figure 5.2) toensure that the overdrive operates only when thecorrect conditions exist. This was only the start of theintegration of electrical control into vehicletransmissions: from this point, the growth insophistication developed rapidly.

5.3.2 Integration of electronics fortransmission control

In the 1980s the next major step forward fortransmission system control was the integration ofelectronic control for automatic gearboxes. This was alogical progression which allowed greater degrees offreedom and flexibility, such as adapting shift control todriving style, simplified hydraulics in the gearbox andreduced costs.

A further advantage of this system, exploited duringthe late 1980s and 1990s, was the integration oftransmission control with engine control. Control unitscould share information from common sensors,reducing costs. Strategies to improve gear shifts byusing engine control parameters (such as ignitiontiming) could also be used to improve drivability andperformance.

These systems have been developed in combination,even further in more recent years. Tighter integrationbetween transmission/engine control andmanagement, brought about by increasingly stringentemission regulations, means that complete control ofthe powertrain is essential for modern vehicles. Nolonger can the engine and transmission be consideredas separate units. Current developments in vehicle

5.3 HISTORY OF ELECTRONIC CONTROL

Figure 5.1 An early overdrive assemblya Direct driveb Overdrive

[647] Chapter 05 25/8/06 16:35 Page 188

Multiplexing 189

dynamic control systems require open, fastcommunication between control elements (steering,braking, transmission, engine, etc.) and tightlyintegrated systems to achieve high levels of vehiclesafety and performance.

5.3.3 Future developmentsFuture developments will involve improvements incontrol performance. One of the latest enginetechnologies under development (homogeneous chargecompression ignition, or HCCI) requires dynamic realtime control to achieve the required performance and tooperate within, for example, emissions limits. This levelof control performance, in combination with fastcommunication between electronic control units, allowsfurther refinement of transmission and powertraincontrol.

Transmissions for hybrid engine technologies requirespecial consideration. Electric motors can producemaximum torque at zero speed and hence do not need agearbox as such. When these motors are used inconjunction with a combustion engine, a sophisticatedpowertrain control system must be used. This uses theappropriate prime mover according to the drivingconditions and operates the whole system as efficientlyas possible (the control system needs to provide energyrecovery, battery management, switch over fromelectric to IC engine power and vice versa, etc.)

There is no doubt that this technology will evolveand change shape as powertrains are developed toproduce more efficient, cleaner vehicles withincreasingly higher performance and drivability.

Engine and transmission electronic systems mustcommunicate to improve efficiency

Powertrains are constantly evolving

Key

Poin

ts

5.4 MULTIPLEXING

5.4.1 Integrated electronic controlunits

The transmission system can be improved and mademore sophisticated by integrating the transmissioncontrol unit (TCU) with the engine control unit (ECU).This is a purely electronic control/softwaredevelopment. The transmission, engine and equipmentremain unchanged.

Linking together the operation of these two units(Figure 5.3) provides integrated powertrainmanagement. This is now common practice withmodern vehicles and it is a logical step forward toimprove the vehicle powertrain system as a whole unitrather than considering engine and transmission asseparate systems. Some vehicle manufacturers are now

opting to produce a single electronic control unit forengine and transmission called a PCU, or powertraincontrol unit. This represents the ultimate step inharmonisation and physical integration of the controlsystems.

Typical technology adopted by most manufacturersfor this type of integration and communication is viathe controller area network (CAN) bus. Most electroniccontrol units have inbuilt CAN capability and the levelof system integration and interaction is chosen by themanufacturer but not limited. Integration of the vehiclecontrol systems can provide a number of advantages.

● Maximum efficiency and performance can be fullyrealised through integrated control of the engine,transmission and powertrain components. For

Figure 5.2 Basic overdrive circuit diagram

Rear

Overdrivesolenoid

Gearboxswitch

Ignitionswitch

Vacuumswitch

Manual switch

Fusebox

13

24

RHS

LHS

Relay

[647] Chapter 05 25/8/06 16:35 Page 189


example, the gearbox shift function can beincorporated with the engine torque control and thiscan be adjusted during the shift to optimisedrivability.

● Sensor information can be shared and used fortransmission operation and optimisation as well asengine functions (for example, engine speed,throttle position and engine temperature). Overallsensor count is minimised (reducing manufacturingcosts) and wiring system complexity is reduced.

● Vehicle control systems can be harmonised to worktogether such that their function and operation canbe complementary. This is particularly importantwith safety systems. Operation of transmissioncomponents such as four-wheel drive, activedifferential locks and torque distribution can be fullyintegrated with safety systems such as stabilitycontrol, traction control, anti-lock braking andbrake force distribution, and engine controlfunctions such as throttle position and torque.

For more detailed information about the technology forinterfacing ECUs, see Hillier’s Fundamentals of MotorVehicle Technology Book 3.

5.4.2 Electronic transmission controlas a black box

Electronic control of the transmission is an inevitablestep forward for motor manufacturers, who havealready implemented full electronic control andmanagement of the engine fuel, ignition and controlsystems. This step is necessary to optimise the overallefficiency of the vehicle power unit. No longer canengine and vehicle be considered separately, so full

integration and overall electronic control of thecomplete powertrain is a reality today. This enablesmanufacturers to achieve the levels of performance,drivability and economy that the market demands.

It is important to remember though that thepowertrain or transmission control is a simple system! Itis similar to any other system on the vehicle. As such,understanding of its operation can be broken down intomanageable elements. This is particularly importantwith fault diagnosis.

The powertrain or transmission control ECU can beconsidered as the central component in this system. It issupplied information about the powertrain status froma number of strategically placed sensors. Thisinformation would typically include pressure,temperatures, rotational speeds (wheels or gearboxshafts), linear speeds (vehicle speed) and driverrequirements (throttle position or gear lever position).The ECU processor runs a software program in real timewhich responds to these inputs and calculatescorresponding actions to be taken. These actions areimplemented by actuators connected to and driven bythe ECU. For example, these could be solenoid valves tosupply pressurised oil to brake bands or clutches in thegearbox, or signals to other engine systems toimplement some required action (for example, to retardthe ignition during a gear shift).

Sophistication in transmission and powertraintechnology is shifting from mechanical to electronictechnology or mechatronics. Sophisticatedmechanically based control systems (such as ahydraulic valve block for an automatic transmission)are being replaced with software and electronics withmuch higher degrees of freedom and flexibility. Theconsequence of this is that the remaining mechanical

Figure 5.3 Powertrain control unit showing inputs and outputs

Output speed sensor

Crank sensor

Throttle position

Oil temperature

Inhibitor switch

Vehicle speed sensor

Air conditioning

Idle switch

Stop lamp switch

Selector switch

Up-shift switch

Down-shift switch

Control relay

Solenoid actuator 1

Solenoid actuator 2

Solenoid actuator 3

Solenoid actuator 4

Lock up clutch

Gear position

Other

EngineECU

TransmissionECU

Diagnostics

[647] Chapter 05 25/8/06 16:35 Page 190

Multiplexing 191

parts become simpler, since there is less intelligenceneeded within them. This means that the skills of themotor vehicle technician or engineer must also shift inthe same direction. These days it is not possible to avoidthe prospect of facing electronic or electrical faults onvehicle powertrain or transmission systems. This doesnot need to be a problem! A logical thought process inconjunction with simple functional decomposition orbreakdown of the system will overcome any fault thatcould occur.

Figure 5.4 shows a simple functional breakdown ofa powertrain control system.

5.4.3 General commentsThe above developments in electronic control systemshave reduced the mechanical complexity of the gearboxdesign itself. The distinction between automatic andmanual gearbox construction is becoming less clear. Forexample, in the 1970s, the construction of an automaticgearbox was completely different from that of amanual, constant mesh gearbox. The automaticgearbox was expensive to manufacture. In place of themanual gearbox’s clutch there was a torque converter.In place of the gears and synchromesh there wereplanetary gears and in place of a gear lever there was asimple selector lever connected to an extremelysophisticated and complex all hydraulic system ofvalves to control gear shifting and selection.

Developments in electronics, sensors and actuatorshave simplified the mechanical construction ofgearboxes to the point where a modern, automaticgearbox is very similar to a manual gearbox, except that

clutch operation and gear selection can be implementedusing electroactuators. A system like this uses allelectronic control and therefore offers many possibilitiesfor the use of advanced control methodology andcommunication with other vehicle systems.

The first developments in electronic control oftransmissions consisted of replacing hydraulic controlwith electronic control, but still using planetary gears,brake bands and clutches; so the mechanical automaticgearbox was still of conventional construction. Afterthis step, integration with engine control andmanagement systems became usual to increaseefficiency and drivability. Current technology is suchthat less direct control is given to the driver in manyvehicle systems. Brakes and dynamic stability areelectronically controlled and monitored. Throttle andtraction control are fully electronic. In line with thistrend, gear selection and clutch control of a manualgearbox can be implemented and greatly improvedwith the addition of electronic control and monitoring(for example, a Tiptronic or a direct shift gearbox(DSG)). We are now reaching the point where thedifference between an automatic gearbox and amanual gearbox is just in software function rather thanhardware or construction.

All complex systems can be broken down andrepresented in a diagram showing inputs andoutputs

Some systems use a single ECU for engine andtransmission control. Others allow the ECUs tocommunicate using a CAN system

Key

Poin

ts

Figure 5.4 Electronic control of transmission shown as a block diagram

Crankshaftposition

Engine speed

Enginetemperature

Battery

Throttleposition switch

Air flow sensor

Vehicle speed

Selector leverposition

Program switchKick-downswitch

PRND 3 2 1

+ –

Electronic control unit

Outputstage

Outputstage

RAM/ROM

Ignitioncoil

Injectionvalves

Failureindication

Relay

Pressureregulator

Shift valves pickup

Reverse gear protection

I/Otrans-mission

Components usedfor transmission

Input signals fortransmission

Input signals forMOTRONIC

Input signals forboth systems

Legend

Inpu

t-si

gnal

con

ditio

ning

CPU

I/OIgnitionInjection

Outputstage

Outputstage

Outputstage

[647] Chapter 05 25/8/06 16:35 Page 191


Value to bemeasured(temperature for example)

Sensor convertsthe value to avoltage (or otherelectrical signal)

Signal is converted from analogue to digital(A/D)

ECU acts on the informationreceived andsends outputsignals to the actuator

5.5 SENSORS AND ACTUATORS USED IN TRANSMISSION SYSTEMS

Figure 5.5 Representation of sensor operation

The sensors and actuators provide the essential dataand control elements of the powertrain control system(Figure 5.5).

5.5.1 SensorsMost of the sensors used in powertrain or transmissionsystems are similar in technology to the equipment usedfor engine management or engine control systems, sothey are subjected to a similar harsh operatingenvironment. They have to work reliably in extremes oftemperatures and pressures and perform their requiredtask for as much of the life of the vehicle as possible.

Normally the sensor has a simple transfer functionrelating or converting the measured value to an outputsignal. In most cases, for an automotive system, theoutput signal would be a voltage. This voltage would beconnected to the ECU input. An analogue to digitalconverter inside the ECU digitises this signal ready forprocessing by the central processing unit (CPU). Theprocess for driving the actuator is similar but reversed.The CPU outputs the demand value digitally to a digitalto analogue converter, and then the analogue signaloutput will operate the actuator or actuator controllerto move the actuator to the required position (note thatthe actuator controller could be inside the ECU; forexample, it could be a stepper motor driver circuit).

An important point to consider is the developmentof smart sensor technology. Sensors fitting thisdescription are being developed and implemented morefrequently in modern vehicles. A smart sensor could bedescribed as one that has ‘local intelligence’. This meansthat, rather than just being essentially a converter,converting some physical quantity into an electricalquantity, this sensor can have additional functions, suchas analogue to digital conversion, digitalcommunication with the driver, self monitoring, aplausibility check, etc. The development of thistechnology has only been possible because of thedramatic miniaturisation of CPU technology. Theadvantages are that:

● there is distributed intelligence in the powertraincontrol system; a greater overall intelligence in thesystem; and less load on the ECU CPU

● signal transmission can be digital, with a vastlyimproved signal quality and reliability, and is lesssusceptible to interference

● the bus system can be standardised; sensors can bedaisy chained on the bus; and there is a significantreduction in cabling/wiring.

Powertrain sensors are similar in their technology tosensors used for engine control (for example, speed ofrotation sensors). Below, there is a brief overview of thespecific technology.

Speed (rotation) sensorSpeed sensors are used in engine control. For example,all ECUs need a crankshaft position sensor and mostneed a camshaft position sensor as well. Speed sensingof gearbox shafts is important for auto shiftfunctionality and to provide closed loop feedback forshift quality control. The two main technologies usedare inductive analogue sensors and digital Hall effectsensors. Both of these, when used as rotation sensors,measure relative velocity and form part of anincremental rotation sensing system. They are used inconjunction with an encoder or toothed wheel. Atypical inductive sensor’s cross section is shown inFigure 5.6.

Figure 5.6 Inductive type speed sensor

[647] Chapter 05 25/8/06 16:35 Page 192

Sensors and actuators used in transmission systems 193

The inductive sensor is simple and reliable; it works onthe principle that a moving magnetic flux will induce avoltage into an adjacent conductor. The constructionshown includes a permanent magnet and soft iron core,around which an induction coil is wound. The sensor ismounted in close proximity to an encoder wheel or agear wheel with a tooth profile. When this gear wheel isrotated, the movement (passing) of the teeth in closeproximity to the sensor disturbs the magnetic flux andhence induces a small voltage in the coil. The voltageapproximates a sine wave; its amplitude increases asthe speed increases (up to a saturation point). Thefrequency is a function of the number of teeth on thewheel (which is fixed) and the speed of the wheel. Thusthe frequency of the signal is proportional to theinstantaneous speed of the encoder or gear wheel. Theair gap and tooth profile dictate the overall outputsignal profile and a reference mark can be providedwith a missing tooth or teeth. Through special pulseconfiguration and teeth profiles, the shaft position anddirection of rotation, as well as speed, can be measured.One limitation of this sensor is that it is passive innature: it cannot detect zero movement or position.

The Hall effect sensor (Figure 5.7) is similar inoutward appearance and use to the inductive sensor.The main difference is the fundamental operatingprinciple. The Hall effect is widely used in industrialsensing technology as well as automotive sensingapplications and can be described simply as a voltageproduced by the interaction between the current flowingand the magnetic field around a current carryingconductor. This voltage is a function of the magnitude(size and direction) of the current and magnetic fluxstrength, but also of the material of the current carryingconductor. In all automotive sensor applications thesensing element (the current carrying conductor) is asemiconductor and hence optimised to provide a highquality output signal (normally a voltage).

Hall effect sensors are also used for current sensingapplications, where they can sense the current flowingin a cable via the magnetic field around it. The mainadvantages of a Hall effect sensor when used as anincremental rotation sensor are that:

● the output signal it produces has a fixed amplitude;only the frequency varies

● the signal is normally a square wave, which is easilyprocessed with simple electronic circuitry in theECU

● the sensor is active and can be used for positionsensing: it can sense zero position, because theencoder wheel does not have to be moving togenerate a signal.

One example of the use of these sensors in transmissionsis to provide information about torque converter turbinespeed. This would be used in the control strategy forshifting, for torque converter lock up control and also fordetermining the correct line pressure during shifts.Additionally these sensors would be used in CVTapplications for determining actual gear ratios.

Gearbox RPM sensors comply with two mainprotocols:

● the signal frequency is proportional to the speed● the frequency and pulse width modulated (PWM)

signals give information about speed (includingzero speed, i.e. standstill) and direction of rotation.

Temperature sensorsPowertrain or transmission control systems need tomonitor the temperature of the transmission. Asophisticated system will be calibrated to adjust systempressures and responses as the temperature changes toachieve the optimum drivability under all engineconditions. Such a system will also act to prevent failureof the transmission from excessive temperatures inextreme working conditions.

Is

RM

Supply voltage

Measuring resistor

Figure 5.7 Hall effect speed sensor

[647] Chapter 05 25/8/06 16:35 Page 193


The sensing technology is similar to that used in enginetemperature sensors (Figure 5.8). The sensors arecontinuous function devices that change resistance withtemperature (they are resistive elements with a negativetemperature coefficient or thermistors). A temperaturesensor would typically be mounted near the torqueconverter, or the oil sump of the transmission, or both inautomatic transmission applications. For automatedmanual transmission (such as Tiptronic or DSG) there isa mechatronic assembly mounted on the gearbox. Thisunit houses an electrohydraulic control assembly as wellas the transmission ECU itself. This is a harshenvironment, so temperature sensors are mounted notonly to sense oil temperature but also to sense ECUcontrol unit temperature. The latter is integrated withinthe assembly.

Transmission temperature sensing will become morecommonplace as engine and transmission controllersbecome more tightly integrated (with full powertraincontrol), and also, as emission legislation and onboarddiagnostic legislation become more comprehensive.

Driver’s lever position sensorThe lever position sensor is the main user interface tothe driver (Figure 5.9). Depending on the manufacturerand the vehicle, it will have a number of functions:

● mode selection for the desired gear (park, reverse,neutral, etc.)

● reversing light operation● selection of overdrive● shift up and down for sequential gearboxes.

Additional switches (Figure 5.10) can be incorporatedfor the driver to select different shift modes to suitdriving styles or conditions. These will change the shiftstrategy, line pressure and torque converter controlaccording to the selected mode. For example, winter

driving mode could set the transmission in third gearand disable torque converter lock up. For driving inslippery conditions this mode reduces torque at thewheels to prevent wheel spin. Another typical selectionis ‘sport’ mode, which changes the shift point selectionand kick down trigger point to optimise accelerationand give a sporty feel to gear shifting.

Figure 5.8 Temperature sensor

Figure 5.9 Transmission gear selection lever

Figure 5.10 Selection switch

Pressure sensorIt is important to control pressure in the transmissionsystem, particularly in traditional automatictransmissions, to ensure good performance of thecontrol and actuator system. The valves and actuatorsall need the hydraulic pressure to be correct to maintainoptimum system performance under all conditions. Astemperature changes, the viscosity of the lubricant canchange, which in turn can affect the system pressure.

[647] Chapter 05 25/8/06 16:35 Page 194


changes in permeability by measuring changes in itsown magnetic field.

One design of magnetoelastic sensor is constructedas a thin ring of steel tightly coupled to a stainless steelshaft. This assembly acts as a permanent magnet whosemagnetic field is proportional to the torque applied tothe shaft. A magnetometer converts the generatedmagnetic field into an electrical output signal that isproportional to the torque being applied. In anotherproposed design (Figure 5.12), a portion of the shaft ismagnetised and the non-contact sensor measureschanges in the magnetic field caused by the torsionalforces in the shaft.

The technology used is similar to that used for enginecontrol (Figure 5.11): a pressure sensing element (athick film or semiconductor) is mounted in a housingand exposed to the system pressure via a channel in afitting or mounting screw. With this sensor technologyadditional circuitry can be included for linearisationand temperature compensation of the raw signal. Thissignal can be amplified to provide an appropriateoutput voltage signal level.

Drivetrain torque sensorA recent development is a sensor to measure drivetraintorque in production vehicles. This is particularlyuseful, since it could provide true closed loop feedbackand control of drivetrain torque distribution underoperational conditions. Additionally, real time limitmonitoring of transmission torque throughput couldhelp prevent damage to any transmission or powertraincomponents. Such a sensor is also useful for monitoringthe efficiency of the powertrain. The sensor technologyused is non-contact magnetoresistive. Sensor units havebeen developed that can be integrated into existinginstallations with minimal design changes.

Torque sensing can use one of two principles:measurement of stress in the shaft material (this is afunction of the shaft torque); or measurement ofangular displacement due to torsion between twopoints on the shaft. Torque sensors for powertrainapplications use the former method and measure thisstress via a magnetoelastic principle.

The ability of the shaft material to concentratemagnetic flux (i.e. its magnetic permeability) varieswith torque: a magnetoelastic torque sensor detects

2 cm

Connector

Evaluationcircuit

Sensingresistors

Pressure connection

Mounting thread

P

Figure 5.11 Pressure sensor principle

Flux gates measure the axial magnetic fieldgenerated by the shaft under torsional load

based on the magnetoelastic principle

Magnetised shaft

Figure 5.12 Magnetoelastic torque sensor measurementprinciple

Position sensorsFor certain applications a number of sensors could bemounted in strategic positions to confirm that an actionhas taken place, or to provide an input signal. Forexample, sensors could detect gear lever movement forsequential shift application, or travel sensors coulddetect movement of shift forks in an electronicallyautomated manual gearbox. For these applicationssmall Hall sensors are commonly used for theirrobustness and reliable operation (Figure 5.13).

Indirect sensor signals fortransmission/powertrain controlA number of sensor inputs to the transmission orpowertrain control are necessary to fully integrate andharmonise the operation of the engine andtransmission. Even though these signals may not beobviously linked to operation of a gearbox, the currenttrend to close integration of engine and transmissionto provide an integrated powertrain unit means thatthis type of integration becomes essential to support

[647] Chapter 05 25/8/06 16:35 Page 195


● Air conditioning status – This signal is used by thepowertrain control to compensate for the additionalload placed on the engine by the air conditioningsystem. It is derived from the compressor clutchsignal. The transmission control line pressure istrimmed slightly according to the additional load.

● Mass air flow/manifold pressure – This signal isused as a fundamental indication of engine load. Itis measured by an air mass flow sensor. Certainmanufacturers prefer to derive engine load frommanifold pressure and throttle position. In eithercase, this load signal is a parameter used in thecontrol of line pressure and lock up clutch function.

● Throttle position – Another fundamental input tothe engine and transmission control, this signal isimportant not only for absolute throttle position butalso for rate of change during fast changing driverdemand (e.g. tip in or WOT). The signal will beused to control shift scheduling, and to control theline pressure and torque converter.

● Vehicle speed – The vehicle speed is one of themain powertrain outputs (in addition to torque) andhence is a critical parameter to monitor and feedback. The existing vehicle speed sensor is used forthis purpose to determine shift scheduling and linepressure control.

Most sensors convert a physical variable to anelectrical signal

The main types of sensor used in transmission are:speed, temperature, pressure and position

Torque sensors are under development for usewith transmission control systems

Indirect sensors supply information from othervehicle systems

5.5.2 ActuatorsActuators are devices that convert the low level electricalsignal from the ECU into actual, physical movement.This could be continuous, cyclic movement ormovement to a set position in accordance with ademand value. With the current trend towards indirectoperation of powertrain components (drive by wire),actuators form an important element of the powertraincontrol system. Clearly, an actuator needs to activate aphysical output to the required position, in the requiredtime. The sensor must be able to repeat this as and whendesired, reliably, for many operating cycles throughoutthe life of the vehicle. The actuator must also be capableof operating at the extremes of temperature that couldbe encountered around the globe.

The current trend is towards totally electricalactuation as opposed to electrical servo operation. Forexample, rather than having an electrical solenoidopen a hydraulic valve to provide oil pressure to a

Key

Poin

ts

Figure 5.13 Gear lever sensor control unit

overall targets of low fuel consumption, gooddrivability and optimum performance. Typical inputsappropriate for an electronically controlled automaticgearbox are as follows.

● Engine coolant temperature – This is sensed via anegative temperature coefficient (NTC) sensor forengine management functions. With thetransmission system certain functions, such as thetorque converter lock-up clutch, are disabled untilthe engine reaches operating temperature.

● Engine speed (rev/min) via crank positionsensor – A fundamental input to the transmissioncontroller, as well as to the engine control system,this signal is the transmission input speed. Thesignal is a conditioned version of the ECUcrankshaft position sensor signal. It is needed tooptimise the transmission strategy for the variousengine operating states.

● Brake on/off – This simple digital state signal isderived from the brake pedal in a similar way to thebrake light switch. It is normally provided via aseparate switch mounted on the brake pedal or via adouble pole brake light switch, for safety reasons.This signal is used to disengage the torque converterclutch under braking conditions.

Hall sensors

Sensor control

Hallsensors forTiptronicoperation

[647] Chapter 05 25/8/06 16:35 Page 196


hydraulic actuator for physical movement, the actuatoris totally electrical/electronic, with an electricalmechanism providing the physical movement (i.e.direct rather than indirect electrical actuation). Thereason for this is that, as electronics and electricaltechnology has developed, it has become possible toproduce reliable highly efficient electroactuatorscheaply, with the required performance. These do notneed large amounts of electrical power, yet can stillprovide the required force.

These actuators are compact and self-contained(they need only electrical signal and power cables: noadditional power source (such as oil pressure) is neededfor the necessary force to be generated). Henceintegrating these units into existing powertraininstallations or new installations where space is at apremium (always the case for modern vehicles) is less ofan issue for manufacturers. Another important benefit,becoming increasingly significant, is that these actuatorscan have ‘local intelligence’. The benefits of this areclear: the actuator can provide reliable feedback to thecontrol system about its performance. An actuator canprovide not just positional feedback (which is importantfor closed loop control) but also status monitoring, suchas temperature or a signal plausibility check. This allowscomplex safety and redundancy to be integrated in theoverall control system. For example, a clutch actuatorcould monitor the system for wear of the clutch andinform the driver when clutch replacement will soon beneeded. Such actuator intelligence can be used by thecontrol system to provide real time monitoring of thepowertrain components, checking for faults andefficiency of operation. This technology will becomeessential as requirements for on board diagnostics andsystem monitoring become more sophisticated and aslegislation becomes tighter.

The following section discusses the different typesand basic operating principles of commonly foundpowertrain actuators.

Electrohydraulic actuatorsSimple on/off actuatorSimple solenoid actuators (Figure 5.14) are commonlyused within the transmission system to direct flow of oilinto and out of components as required, typically toimplement gear shifting and for torque convertercontrol. Their basic construction consists of a spoolvalve to control fluid flow connected to a simplesolenoid arrangement. Current supply to the solenoidchanges the valve state. They can be constructed for useas on/off or change over valve (normally open orclosed). They are usually switched at battery voltagebecause they demand a high current at initial actuation(pull in of the solenoid). The benefits of these actuatorsinclude their simple, robust construction.

Variable position bleed actuatorThe variable position bleed actuator is used inapplications where a variable movement and position

are required for fluid control (see Figure 5.15). Fortransmission applications, such an actuator would beused in a gearbox as a bleed valve, for example inpressure control applications where the output pressurefrom the valve would be a function of the supply current.An important design criterion is to reduce ‘stiction’ orstatic friction in the valve to allow precise positioningwith minimal error. This can be achieved through carefulport design. An electronic control circuit with pulsewidth modulation would typically be used to operatethis valve by supplying the appropriate varying currentaccording to demand. An important feature of the circuitis that it should be able to provide an appropriatecurrent to activate the valve quickly and then supplysufficient current to hold the valve position withoutoverheating (different currents for pull in and holdingare provided by an intelligent driver circuit). This issimilar to a fuel injector driver circuit in the engine ECU.

It is very common to find all the requiredelectrohydraulic actuators mounted in a completeassembly on or in the gearbox casing. This has anumber of advantages:

● interconnecting hydraulic paths between the valvesare short, minimising pressure drop

● one casting contains all the valves, reducingmanufacturing costs

● the system is easier to integrate into the gearboxassembly or design

● its single interface point is easier to integrate intothe system electronics.

Figure 5.14 Solenoid actuator

Figure 5.15 Schematic of a variable position bleed actuator

Fmag

FfeedPfeed

1

Pclutch

[647] Chapter 05 25/8/06 16:35 Page 197


motion via a worm driven crank assembly. This in turnoperates a piston inside the unit which provideshydraulic pressure to a clutch slave cylinder assembly. Inthis case, therefore, the existing clutch actuator (ahydraulic slave cylinder) remains and theelectroactuator replaces the master cylinder assembly.Clearly this system also incorporates an ECU and aclutch pedal position sensor. The main advantage ofsuch a system is that it can be (and has been) introducedin a production vehicle. Minimal changes or adaptationsare needed in the vehicle design to incorporate thissystem.

The disadvantage is that retaining an existinghydraulic system means that this system still needs to bemaintained. In addition, the dynamic response of such asystem is reduced because there is more inertia in thesystem due to the increased component count and theinterfaces between the actuator and the clutch itself.

An improvement is to dispense with the existingactuator system and mount the electroactuator as closeas possible to the clutch itself. This is feasible for acompletely new design of gearbox or powertrain. Designof the actuation mechanism can also improved withgreater operating force and higher efficiency. One designapplies the force directly at the release bearing via aconcentric release mechanism (Figure 5.18). It uses anaxial screw ball bearing track (a helix) with a rotatingcollar such that, as the collar turns, an axial force isapplied to the clutch release bearing via the ballbearings. The collar is rotated by a brushless DC motorwith a position feedback signal. This motor drives a leadscrew assembly to convert rotary to linear motion. Thislinear force is applied to the collar via a Bowden cable(shown in Figure 5.19).

An alternative mechanism is to apply the force tothe release bearing via a lever mechanism (Figure5.20). In this case the actuator can also be mountedclose to the clutch on the bell housing. The actuatoruses a brushless DC motor with positional feedback.This provides a high level of reliability and longevity.The rotary motion is converted to linear movement viaa lead screw or spring band system incorporated in the

Another recent development is to combine within theelectrohydraulic assembly, the electrical and electronicparts, including the ECU! This creates a complete, self-contained mechatronic unit for control andmanagement of the transmission. Very close interfacingof hydraulics and electronics gives greater reliabilityand reduces the installation space needed. In addition,the complete transmission assembly forms a compact,modular unit that can be easily integrated into currentor new vehicle designs. A typical example is shown inFigure 5.16.

Figure 5.16 Mechatronic unit

Figure 5.17 Worm gear drive actuator

ElectroactuatorsClutch actuatorThere are several designs for electronic control andelectrical actuation of the clutch. The type chosen byeach manufacturer depends greatly on whether anexisting gearbox design is being adapted, or whetherthe gearbox design is new and will integrate electricalclutch operation from inception.

Figure 5.17 shows a current design used in aproduction vehicle. This basically consists of a packagecontaining a brushless DC motor with feedback ofposition. The rotary motion is converted to linear

Figure 5.18 Direct acting clutch actuator

[647] Chapter 05 25/8/06 16:35 Page 198


actuator. This is transmitted by a simple leverarrangement as a force at the release bearing. Thissystem is capable of transmitting high forces (forclutches transmitting up to 900 Nm) and can also beused in dual clutch applications.

Gear shift actuatorClearly, with a normal automatic transmission, gears arechanged via a number of actuators and a combination ofbrake bands and clutches, which operate hydraulically.For electronic control applications, they are activated bygearbox system oil pressure which is switched viasolenoid type valves as described above. Therefore theactuators are hydraulic but controlled electrically. In thissection we will focus specifically on electroactuators:actuators where the force is generated by some electricalmeans and controlled electronically.

Actuators of this type are used in the latestgeneration of highly efficient automatic shiftingtransmissions. These transmissions are generally, inconstruction, similar to a manual gearbox, except thatthe driver has only indirect control of the gear shiftprocess. Most vehicles using this system receive driverinput from switches on the steering wheel (typically

Figure 5.20 Dual clutch actuation system

Figure 5.21 Retrofit gear shift actuator

flipper controls as in racing cars). The driver tells theECU that a shift is required; the ECU then calculates thebest time to do this and therefore has the ultimatecontrol over the process, taking into account otherfactors, such as vehicle stability, braking, steering andother dynamic factors.

There are a number of methods of gear shiftactuation depending on the gearbox type. The maindifferentiating factors are whether the transmission isretrofitted with actuators or is a new design where thistechnology will be used. Another factor is whether thetransmission is an auto shift gearbox (a manualtransmission with electroactuation of the shiftingprocess) or is a parallel shift gearbox (such as theVolkswagen direct shift gearbox (DSG), where the nextgear is engaged while the vehicle is in the existing gear;this requires a twin clutch arrangement).

The system shown in Figure 5.21 is a retrofit unitused with an existing gearbox design for a small frontwheel drive vehicle. In this particular application gearselection is via a push pull and twist rod. This rod entersthe gearbox and has a selector ‘finger’ attached whichpushes forward and backward the appropriate selectorrail for the required gear (Figure 5.22).

The actuator unit consists of two brushless DCmotors, one to provide forward and backwardsmovement (for gear selection) via a quadrant gear. Theother provides movement side to side (across thegearbox ‘gate’) via a rack mechanism. The existing gearinterlock mechanism is employed to prevent selectionof two gears at the same time. It is important to notethat this actuator assembly was designed for the firstgeneration of automatic shift gearboxes, wheremodularity (i.e. the ability to reuse existing componentsin gearbox designs) was the most important factor.

Assist

Clutch bell housing

Lever

Releasebearing

Figure 5.19 Mechanical concentric clutch release system forclutch actuator

Bowden cable

Lead screw drive

Brushless electricmotor

Mechanical concentricrelease system

[647] Chapter 05 25/8/06 16:35 Page 199


Current designs of actuation mechanism are in line withcurrent trends in transmission and powertrain design:manual constant mesh gearboxes are used, but withelectrical actuation of clutch and gear selection. Hencethe housings for these actuators will be designed infrom concept. The choice of whether the gearboxbehaves as a manual or automatic transmission (orboth) will simply be controlled via software andperhaps the driver interface (such as whether or notthere is a clutch pedal; whether the gear lever is onewith a manual style gate, or is an auto-style selectorlever, a sequential selector lever or a steering wheelpaddle change).

Figure 5.23 shows a shift drum with an integralmotor which could be installed directly into a gearboxhousing.

This selection mechanism is similar to those used inmost motor cycles, in that the drum has groovesforming tracks into which the gear selector forksengage. The tracks are designed such that, as the drumrotates, the appropriate selectors move thesynchromesh hubs in and out of engagement to selectthe appropriate gear. A brushless DC motor is used,integrated within the drum, complete with a planetarygearbox to provide the required operating speed andforce. The advantage of this design is its compactnessand ease of installation into the gearbox housing. If asingle motor is used to drive the assembly the systemcan be relatively low in cost but there are some points toconsider:

● the gearbox design has to accommodate thearrangement from concept; it cannot be retrofitted

● the shift drum arrangement means that gear shiftshave to be sequential; arbitrary shifts cannot bemade (for example, from fourth to second gear)

Figure 5.22 Manual gear shift actuator for front-wheel drive

Figure 5.23 Double shift drum with internal drive

Figure 5.24 Double shift drum with external drive

Gearbox

SupportIntegrated brushless motor

● the latter can be avoided by using a double shiftdrum arrangement: Figure 5.24 shows an examplewith external drive motors.

Reduction and torque multiplication are provided by aspur gear. For these systems both motors must becapable of sufficient force to provide gear selectionreliably, compared with the previous retrofit design,where only one motor needs to be capable of generatingthis force (the other just moves across the gearbox‘gate’). This factor increases the overall cost of theassembly. Another important point is that interlockingof the gear selection must be prevented: in theory, bothactuators could engage a gear at the same time.Interlocking can be implemented with a mechanicalarrangement as well as through software.

A double shift actuator arrangement is currentlymore appropriate for parallel shift gearboxes or DSGsystems where maximum advantage can be gained froman optimised shift process. By using two actuators, withactive interlocks, reduced gear change times cansignificantly improve vehicle performance andacceleration. This provides a clear sales argument to thecustomer for the additional cost of this technology.

[647] Chapter 05 25/8/06 16:35 Page 200

Clutch electronic control 201

5.6.1 IntroductionThe clutch is a simple device which can be employedin any rotating drive system where it is necessary todisengage or re-engage the transmission of torque. Afriction clutch is used for vehicle transmission systems.

The clutch system consists of a driven and adriving member (Figure 5.25). In a transmission, thedriving member is usually also the engine flywheel.The driven member is a friction disc which connectsvia a spline to the gearbox input shaft. The frictiondisc is clamped by a spring disc in the rest position,and hence drive and torque can be transmitted fromdriving to driven member. The spring clamping can berelaxed as required by the operator (in a vehicletransmission, via the clutch pedal); this allows thefriction disc to spin freely and hence no torque istransmitted. If the spring pressure is graduallyreleased, the clamping force can be gradually appliedto the friction disc and hence torque can beprogressively transferred between the driving anddriven members as required. This allows the smoothtake up of drive and transfer of torque between theengine and gearbox. The clutch in a vehicletransmission allows the effective disconnection of theengine and gearbox as required by the driver. This isneeded for three main reasons:

● as mentioned, it allows the smooth application oftorque from the engine to the gearbox, which isnecessary for the vehicle to start from stationary,because a combustion engine has to be running ata minimum speed to produce power and torque

● to allow disconnection of the engine, so that notorque is transmitted through the gearbox; this

allows the driver to shift from one gear to anotherin a fixed ratio manual gearbox

● to provide a temporary neutral condition for stopand start driving, for example, in traffic.

5.6.2 Electronic clutch managementFor many years, mechanical friction clutches, operateddirectly via a foot pedal (mechanically or hydraulically)have remained completely unchanged. As the clutchforms such a critical part of the transmission/powertrainsystem, simple mechanical operation directly by thedriver leaves much room for improvement. Electroniccontrol of the clutch, in combination with fullintegration of clutch operation with the overallpowertrain control system provides several possiblebenefits:

● improved drivability, with more comfortable andsmoother clutch and gear shift action, and with anti-stall protection in automatic mode

● effortless operation, particularly when fitted tovehicles with high engine torque

● improved pedal feel for the driver, with reduced legpressure required

● improved reliability, which is achieved throughreduced driveline wear and tear, and the totalabsence of mechanical linkages for clutch actuation

● an automatic shifting mode when used with asuitable gearbox

● improved crash protection for occupants becausethe pedal box is less intrusive

● data acquisition of clutch use for maintenancescheduling.

These actuators are becoming increasinglycommonplace on modern vehicles and will certainly bemore common in the future. Current actuator driveunits, brushless DC motors, provide reliable technologythat can last the life of the vehicle, and they are suitablefor use in safety critical applications (as above). Thecompact design of the motors means that the actuatorsrequire no greater space than the manual mechanismsthey replace. Additional technology can be integrated inthe actuator design, such as:

● incremental travel measurement● electronic commutation● position sensing.

This means that additional sensor mounting to supportoperation of these actuators is not specifically requiredbecause it is built in.

Actuators (most of them) convert an electricalsignal into a physical action

A common transmission actuator is a solenoidoperated valve that controls fluid pressure

The current trend is towards full electricaloperation, so that the actuator, for example, actsdirectly on the component and no fluid pressure isrequired

Smart actuators are being developed that containlocal intelligence

DC motors are popular as they can even be usedto retrofit for ‘manual’ gear shifts

Key

Poin

ts

5.6 CLUTCH ELECTRONIC CONTROL

[647] Chapter 05 25/8/06 16:35 Page 201


Such a system has been developed and can be fitted toexisting production models as well as to new designs ofpowertrain systems for road vehicles. Such a systemreplaces the existing hydraulic or mechanical actuationassembly with an electromechanical actuator (see Figure5.26). A pedal position sensor is attached to the clutchpedal for driver input and the whole system is controlledby an ECU integrated in the powertrain control system,which takes additional information on vehicle behaviourfrom the engine and powertrain ECUs and sensors. TheECU controls the clutch actuator directly and thus themechanical link between the clutch pedal and clutch iscompletely eliminated.

The basic functions of this unit are exactly the sameas those of a manual transmission: to accelerate thevehicle from rest and stop it. Movement from restrequires precise control of clutch engagement and thiscan vary according to the operating conditions (a hillstart will require different control from a start on thelevel). Changing gear is a highly dynamic process andrequires precise control for the gears to change

Figure 5.25 Standard vehicle manual clutch

smoothly and seamlessly. Stopping (declutching) is asimple process when carried out manually, but addingelectronic control opens up possibilities during extremeconditions for clutch operation to be more controlledand fully integrated with the other vehicle safetysystems (such as stability control, ABS and tractioncontrol) to improve vehicle safety.

Additional functions can be added to the systemaccording to the application: for example, one couldadd:

● drivetrain condition monitoring● creep control for accurate slow speed movement● active vibration damping for powertrains without a

dual mass flywheel.

The system can implement clutch operation in less thanone-tenth of a second to optimise safety and drivercomfort. Additionally, it operates seamlessly to correctdriver errors. The pedal stroke and resistance are fullyadjustable to suit the driver and the vehicle type. Thesystem’s compact design can save weight and offersgreater flexibility of installation in a vehicle.

[647] Chapter 05 25/8/06 16:35 Page 202

Clutch electronic control 203

CBW transmission system

The driver can select the clutch operation mode – with or without the pedal

The system reduces foot travel and the force required to operate the pedal. The slightest mistake in controlling the clutch (for example, sudden pedal release) is corrected

Actuator optimises clutch operation

Powertrain efficiencyEnhanced driver comfort using a mechatronic clutch

Electronic control unit

Clutch-by-wire (CBW)supersedes the mechanical link between the pedal and the clutch

Characteristics• Offers automatic and semi-automatic clutch modes• Small system dimensions enhance driver leg protection in

front-end impacts• Technology is designed for integration with other systems

(ABS, ESP, etc.)• Enhances vehicle’s reliability by reducing transmission wear

(by automatically correcting operating mistakes) and eliminating mechanical linkages

• Eliminates noise and vibration feedback into the cabin• Enhances driving comfort. CBW makes clutch use easy,

provides smoother gear changes, and eliminates the risk of stalling in automatic mode

Force feel system

The engine does not stall if the driver fails to disengagethe clutch, for example in emergency stop situations

effectively spilt into two halves. During operation, i.e.upshift or downshift, an appropriate gear can be pre-selected while the torque is still transmitted through anexisting gear. This dramatically reduces the gear shifttime, torque transmission to the wheels is practicallyuninterrupted and acceleration times can besignificantly improved. This technology will be discussedin more detail in the appropriate section of this book.

In this application, clutch engagement anddisengagement are handled electronically. Normally,wet clutches are used, actuated by hydraulics. Theelectronic controls are interfaced via VBA valves, drivenwith a PWM signal which is generated from the ECUaccording to demand.

DSG technology will be discussed and explained inmore detail in a later section of this chapter.

Parallel shift boxes pre-select the next gear whilean existing gear is still in use

Electronic clutch control improves efficiency anddriveability

Clutch control can be fully electrical or an actuatorcan be used to operate the slave cylinder, forexample

Key

Poin

ts

Figure 5.27 Twin multiplate clutch in section

Figure 5.26 Electronic clutch control system

5.6.3 Twin clutch arrangementsThe latest developments in gearbox automation (DSG orparallel-shift gearboxes) require a twin-clutcharrangement (Figure 5.27), because the gearbox is

[647] Chapter 05 25/8/06 16:35 Page 203


5.7.1 Introduction

Manual gearboxes (Figure 5.28) have not beendramatically affected by developments in electrical andelectronic systems until recently. Early sliding meshgearboxes with straight cut gears were noisy and harsh,and were difficult to use. Later, constant meshgearboxes with helical gears were quieter, but stilldifficult to use (double de-clutching was required).Modern synchromesh gearboxes are quiet and easy touse, but gears cannot be engaged unless they aresynchronised. These developments all happened in theearly years of the mass produced car industry: since theintroduction of the synchromesh gearbox on massproduced vehicles in the 1960s little has changed inmanual gearboxes, apart from their increasing numberof gears (initially from four to five, and now six gearsare commonplace).

Engine electronic systems have developed rapidlyover the past 20 years. Now these developments arestarting to impact on the transmission, even on acompletely mechanical device, the manual gearbox.Drive-by-wire systems are becoming more common

within the industry and in production vehicles. The nextstep with manual transmission (already available at thetop end of the market) will be the introduction ofelectronic shift control and actuation (even if manuallycontrolled by the driver) in conjunction with electronicclutch actuation and monitoring (where the drivercontrols the clutch position with virtual clutch pedal).

This technology is particularly interesting inapplications where improved fuel economy is the maintarget: for maximum economy, driver behaviourbecomes more critical. Therefore, by automating thegear shift process, the powertrain system can beoptimised to give the best possible fuel consumption:the system can eliminate some driver errors and drivingstyles which can compromise performance and fueleconomy. Where economy is the ultimate target, it ismuch better to use automated manual transmissionsthan other types of transmission, such as CVT orhydraulic automatic transmission. This is becausemanual transmissions do not suffer from the internallosses (due to the hydraulic power needed for shiftactuation) associated with automatic gearboxes andhence offer greater efficiency.

5.7 MANUAL GEARBOX ELECTRONIC CONTROL

Figure 5.28 Constant mesh gearbox

[647] Chapter 05 25/8/06 16:35 Page 204

Manual gearbox electronic control 205

Contrary to the above reasons of economy, automatedmanual shift gearboxes are often used in highperformance vehicles because the manual gearbox ismore efficient than other types! Hence, for ultimatevehicle performance, it is the preferred choice. Withelectronic control of shifting and clutch actuation, shifttimes can be optimised and improved considerablycompared with manual control by the driver. Advancedstrategies for acceleration from rest can be implementedto improve vehicle performance. In addition, as notedpreviously, the transmission control can be integratedwith other vehicle control systems for stability andtraction to provide the ultimate in high performancedriving. For example, in a critical situation such asdownshifting on a slippery surface, the clutch can bereleased instantaneously if there is excessive engine dragtorque, so that the car will not lock the driving wheelsand slide.

This technology has been developed and marketedsuccessfully by manufacturers of high-performancevehicles, such as BMW, Ferrari, and Aston Martin, and itis widely used in motor sport, where the goal is ultimatevehicle speed and performance.

Table 5.2 shows gear shift times for differentvehicles with automated manual gearboxes.

● fast and precise gear shifting within 80 milliseconds● sequential shift actuation via a selector lever or

steering wheel buttons● shift warning lights on the dashboard● self-learning adaptation of the gearbox over time● ‘Drivelogic’ personal settings – 11 driving programs

ranging from balanced dynamic to strictly sport● a sequential shift mode● a fully automatic shift mode● special functions – slip recognition, a climb assistant

and an acceleration assistant.

5.7.2 Case study: VW electronicmanual gearbox

Volkswagen has produced the world’s first production3 litre car! That is, the company has developed a carthat uses only 3 litres of fuel every 100 km. The factorsaffecting fuel economy are many and varied: namelyaerodynamics, rolling resistance, powertrain design andvehicle mass. These all have to be optimised to achievethe required efficiency from the vehicle to lower the fuelconsumption consistently to this level.

The vehicle used is a VW Lupo, adapted to givegreater powertrain efficiency. One of the most importantadaptations was to implement electronic switching inthe transmission (Figure 5.29). This was done to reducethe possibility of increased fuel consumption caused bydrivers’ gear shifting habits. This system also ensuresthat the vehicle is in the correct gear, to give the best fuelconsumption, at all times relative to the drivingconditions. The system covers three main componentareas:

● Mechanical – The transmission is the manualsystem used in current VW production small cars.The gearbox was made lighter via additionaldrillings in internal components and by reducing theoil capacity. Additional mechanical components inthe system include the selector mechanism shaftand levers

● Hydraulics – The manual shift mechanism wascompletely replaced by an electro-hydraulic unit.The clutch is operated through an actuatormechanism. Hydraulic power for the shift andclutch is provided by an electric hydraulic pump andpressure accumulator

● Electronics – The driver uses a throttle pedal sensorand selector lever, with the gear shift and clutchoperated via electro-hydraulic valves. Informationabout shift position and selector lever position is fedback through potentiometers and micro switches. Atthe heart of the system is the transmission ECU.

The system can operate in manual, sequential mode orautomatic mode. Pushing the selector lever sends a shiftdemand signal to the ECU via micro switches, and thelever assembly contains a potentiometer to detect theabsolute position of the lever for selection of neutral,

Table 5.2 Gear shift times

Gearbox/model Minimum shift time (ms)

BMW SMG 1 (M3 E36) 220BMW SMG 2 (M3 E46) 80Ferrari F1 (575M) 220Ferrari F1 (360 F1) 150Ferrari F1 (Maserati 4200GT) 80Bugatti Veyron 200Aston Martin Vanquish 250Alfa Romeo Selespeed 700

An important additional feature of electronic control isits ‘self-learning’ capability. The transmission controlsoftware can be designed to be adaptive to a person’sdriving style, and a driver can select a driving mode,such as sport mode. BMW calls this feature ‘Drivelogic’,when it is offered with their sequential manual gearbox(SMG) on their high performance M series models.This function allows the driver to choose thetransmission shift characteristics from 11 differentdriving programs. These range from a balanceddynamic program (program S1) to a very sportyprogram (S5). Finally, the driver can also choose aprogram (S6) where the system’s dynamic stabilitycontrol (DSC), which comes standard, is deactivated.Here, the transmission will shift and respond with adynamic performance similar to that of a racing car,thus giving the driver the ‘ultimate driving machine’experience!

The highlights of the BMW SMG system are asfollows:

[647] Chapter 05 25/8/06 16:35 Page 205


reverse or automatic shift mode. When a shift is made,the ECU calculates if this is appropriate from a numberof input signals. These are:

● engine speed and torque● throttle position● brake pedal and pressure● vehicle speed.

The signals are provided by sensors connected directlyto the transmission ECU or via the CAN bus interface tothe engine ECU. The process of changing gear is exactlythe same as a driver would implement manually. Firstthe clutch is opened via the hydraulic clutch actuator(Figure 5.30); pressure is supplied and controlled at theactuator via a solenoid valve operated by the ECU. Theclutch position is fed back to the ECU through amovement sensor mounted on the actuator. The clutchlimit positions are monitored by the ECU at regularintervals to compensate for clutch wear.

Once the clutch is open, gears are shifted by thehydraulic pistons in the electro-hydraulic shift actuator(Figure 5.31), and controlled via solenoid valves. Thereare two pistons for gear selection and two for gateselection (each piston pair provides forward andbackward force), each piston has its own controllingsolenoid valve to apply or release hydraulic pressuresmoothly and progressively. This is essential for smoothsynchronisation during gear shifting. Potentiometers

are fitted for gate and gear selector movement andthese send the measured position back to the ECU.

Once the next gear is fully engaged the clutchactuator is released and the clutch closes to reinstatetorque into the gearbox. During transient operationthe clutch is kept approximately 20% open to ensuregood response during gear changing and to reducetransition times.

Figure 5.29 VW electronic manual gearbox overview

Figure 5.30 Clutch slave cylinder

[647] Chapter 05 25/8/06 16:35 Page 206


An additional feature of the system is the stop/startfunction. This eliminates fuel wastage when the vehicleis idling. When the vehicle is stationary, the enginestops if the brake pedal is pressed for more than 3seconds. When the pedal is released, the engine isrestarted automatically and first gear is engaged sothat, when the driver presses the accelerator pedal, thevehicle accelerates immediately. This feature ismanaged by the transmission ECU and is an integratedpart of the overall fuel consumption reduction strategyfor this vehicle.

5.7.3 Case study: Volkswagen directshift gearbox (DSG)

There are a number of advantages and disadvantages ofautomatic and manual transmissions.

● Manual gearbox – It is the most efficient type ofgearbox with minimal power losses; the driver hasfull control over shifting and hence is provided withsportier driving.

● Automatic gearbox – This has the greater level ofsmoothness and comfort, with no interruption intorque transmission during driving.

The ultimate gearbox would be one that combines thebest attributes of both types with the latest in controland system integration technology. VW has proposedthis in the form of their direct shift gearbox (DSG); thistechnology is also known as a parallel shift gearbox(PSG). This attempts to combine the transmissionconcepts of automatic and manual systems into acompletely new generation of gearbox.

The main system highlights are:

● a six speed synchromesh gearbox (plus reverse)● selectable, pre-programmed driving modes (sports,

etc.)● a sequential shift via a lever or steering wheel

buttons● a completely integrated mechatronic control unit

which houses ECU electronics and electro-hydrauliccontrols, mounted on the gearbox, providing asystem with minimal external interface connections

● a hill-holder function and creep regulation, withenhancements for low-speed driving andmanoeuvring

● system fault handling, with full electronic diagnosticcapability, and a limp home mode.

Figure 5.31 Selector assembly

[647] Chapter 05 25/8/06 16:35 Page 207


Input gear of differential

Installation position in gearbox

Parking lock gear

Differentialcrown wheelOutput shafts

Reverseidler

Inputshafts

Twin clutchassembly

Reverseidler

Output shafts

Figure 5.32 Gearbox shaft layout

Mechanical constructionThe torque is transmitted into the two transmissionunits via an integrated twin clutch assembly withhydraulic actuation for both clutches. A dual-massflywheel is used to insulate the transmission fromengine torsional vibrations (see the diagram in theClutch section). The two input shafts are combinedconcentrically, each one fitted with a pulse wheel so theECU can detect rotating speed (Figure 5.32). The twooutput shafts hold the gear synchromesh units and bothtransmit torque to the differential gear. The differentialalso includes a gear wheel for a locking pawl to providea ‘park’ position (with the wheels locked by thetransmission).

Gear selection is via selector forks (Figure 5.33), ina similar way to a normal, manual gearbox, exceptthat, in this case, the selector forks are hydraulicallyactuated through oil pressure. A small piston mountedat each end of each selector fork is supplied withpressurised oil according to shift requirements fromthe control system. A small permanent magnet fittedto each selector fork allows the ECU to detect theprecise fork position and hence gear engagement via asensor in the gearbox. Once the selection is made thepressure is released and the selector fork is held inposition by a locking mechanism. Figure 5.33 Selector mechanism

[647] Chapter 05 25/8/06 16:35 Page 208


them; they monitor output shaft speed and directionof vehicle travel (by offsetting the two signals)

● clutch pressure – sensors are used in the regulationof clutch operation

● gearbox oil temperature – sensors protect thegearbox from overload and are used to initiate awarm-up function

● control unit temperature – sensors protect thesystem electronics

● selector fork travel – sensors monitor the actualposition of the selector fork for gear selectionstatus

● selector lever – sensors provide information aboutabsolute position (park, reverse, neutral, etc.) andsequential shift (up, down).

The electro-hydraulic part of the control system consistsof a number of actuators:

● a pressure control valve – this modulating valveregulates the main system pressure according toengine torque

● clutch pressure control valves – modulating valvescontrol the clutch operation

● an oil pressure control valve – a modulating valvecontrols cooling oil flow

● gear actuator valves – on/off solenoid valves engageand disengage gears via selector forks

● a multiplexer control valve – an on/off valvecontrols the position of the so-called multiplexer;this unit is used together with the gear actuatorvalves for gear selection and reduces the number ofgear actuator valves required

● safety valves – modulating valves isolate hydraulicpressure in sections of the gearbox in the event of asafety related fault; they also allow rapid opening ofeach respective clutch, if necessary, when anoverpressure occurs.

Additional interfaces are provided via the CAN to:

● the anti-lock braking system (ABS), the electronicdifferential lock (EDL) and the traction/stabilitycontrol (ESP) system

● the diesel or gasoline engine management system● the selector lever control unit● the steering column electronic control unit.

Basic principle of operationThe DSG consists in essence of two manual,synchromesh gearboxes in one unit. Each one has itsown clutch and torque input from the engine. Theseclutches are wet, multi-plate clutches that are actuatedhydraulically under the control of the transmission ECU.Look at Figure 5.35: the first, third, fifth and reversegears are within transmission unit 1 and the second,fourth and sixth gears are within transmission unit 2.

The fundamental principle behind splitting thetransmission in this way is that one transmission can beengaged (i.e. in gear, transmitting torque) while theother transmission can be in the next gear (i.e. in gear

An important part of the gearbox is the oil lubricationsystem. This provides not just lubrication and coolingbut also hydraulic power for the actuators to shift gearsand operate the clutches. The oil pump is drivendirectly from the engine input through a shaft. Thelubrication system also incorporates its own filter andheat exchanger, since it is so critical to correct systemoperation and performance.

Electronics and control systemAt the heart of the system is the mechatronicstransmission module (MTM) (see Figure 5.34). Therobust construction of modern electronic technologymakes it commonplace with electronic transmissions tointegrate the electronics, electrics, hydraulics andmechanics into a single module, mounted at thetransmission itself. This has the advantage that itprovides the highest level of component integration withthe minimum number of external interfaces andconnections to the vehicle, which greatly improvesreliability.

In the DSG system the MTM is at the centre of thesystem and all sensors and actuators are connected to it,since all actions are initiated and monitored by it. Thisunit also houses the ECU itself.

The sensors in the system measure the following:

● clutch oil temperature – at this position in thegearbox the lubricant is under the greatest thermalstress, and by monitoring the temperature at thispoint the control unit can regulate the flow of oilaccordingly

● gearbox input speed – this is basically the same asthe engine speed

● input shaft speed – a speed sensor on each inputshaft monitors the speed input to each half of thegearbox; they are mounted on the opposite side ofthe clutch to the above sensors, which allows thesystem to monitor the clutch status and slip ratio

● output shaft speed – two sensors are mounted on asingle pulse wheel but with phase shift between

Figure 5.34 Mechatronics module

[647] Chapter 05 25/8/06 16:35 Page 209


but not transmitting torque), in preparation for the nextgear change. This gear change can be made veryrapidly, just by switching torque transmission from oneclutch to the other. It can be implemented in acontrolled manner, such that there is minimal loss oftorque at the road wheels.

With its double-clutch design and sophisticatedelectronic control, the system is as comfortable to use asan automatic transmission. In addition, its capacity toimplement lightning quick gear shifts means no loss oftorque transmission at the wheels, so performancedriving is particularly rewarding.

One important point to note though is that, becausehydraulic power is required to operate the gearbox,there are some small parasitic losses, which reduce theoverall efficiency slightly. Parallel shift gearboxes thatare currently under development by other

manufacturers use twin plate dry clutches with fullyelectrical actuation of the clutches and gear shifting.This improves efficiency as there are no powerrequirements if the gearbox is in a quiescent condition(i.e. a gear is fully engaged, with the gearboxtransmitting torque).

Manual gearboxes under full electronic control areeffectively fully automatic – but are more efficientthan epicyclic or CVT gearboxes

A clutch actuator is needed to control manualboxes automatically

Most manual systems can operate in manual,sequential mode or automatic mode

Electronic control reduces shift times considerably

Key

Poin

ts

Transmission unit 2

Transmission unit 1

Multi-plate clutch K2

Multi-plate clutch K1

Engine torque

Figure 5.35 DSG basic principle

5.8.1 IntroductionA torque converter (Figure 5.36) is standard inautomatic transmissions of all types and basicallyreplaces the clutch. It converts a high speed/low torqueinput from the engine into high torque/low speedoutput to drive the transmission and therefore allowsthe smooth take-up of the vehicle from rest. Also,because of the ‘slip’ effect, it infinitely multiplies thenumber of gear ratios available by effectivelyinterpolating between each fixed gear ratio.

A torque converter is basically an opposed pump-turbine unit, enclosed in a casing partially filled withhydraulic oil.

The pump side of a torque converter is directlyconnected to the prime mover (the combustion engine)and the rotation torque circulates the oil inside thecasing. This imparts energy into the fluid in the form of

5.8 TORQUE CONVERTER ELECTRONIC CONTROL

Figure 5.36 Torque converter

[647] Chapter 05 25/8/06 16:35 Page 210

Torque converter electronic control 211

kinetic energy. The turbine is placed directly in the pathof the moving fluid (i.e. directly opposite the pump), sothe dynamic energy in the fluid is recovered by theturbine and converted back into torque. This simplesystem is known as a fluid flywheel. The torqueconverter (Figure 5.36) has an impeller mountedbetween the pump and turbine to improve efficiency.The impeller (or stator) is mounted to the casing via aone-way clutch which allows it to rotate in onedirection only. The stator redirects the fluid path undercertain conditions to ensure that the fluid strikes theturbine at the correct angle and speed under allconditions, thus greatly improving the overall torqueoutput of the unit.

5.8.2 Torque converter lock-up clutchMost modern automatic transmission systemsincorporating a torque converter will use a lock-upclutch across the converter to improve fuel economy.The basic problem with a torque converter is that itrelies on a speed difference between the input andoutput shafts to be able to operate and transfer ormultiply torque. If the speeds are the same,hydrodynamic oil circulation will not take place andtorque will not be transmitted. Loss of speed in thetorque converter is known as slip, and causes somepower to be lost as heat.

For overall powertrain efficiency, losses must beminimised so a lock-up clutch is fitted to the converterbetween the turbine and impeller (on the casing)(Figure 5.37). When required by the control system (forexample, when a vehicle is cruising in top gear) thelock-up clutch can be engaged and the converter

bypassed. This clearly prevents slip and the parasiticlosses that are unavoidable in the converter itself duringnormal operation. Integrating operation of this clutch isparticularly easy when the control system itself iselectronic.

The clutch can be engaged or disengaged by thetransmission ECU with an electro-hydraulic controlsystem. This could be a simple on/off arrangement suchthat the clutch is engaged only in top gear, or it could beengaged in top and second-to-top gears. This would bea relatively simple way of increasing the overalltransmission efficiency in cruise conditions only.

5.8.3 Optimisation of lock-upclutch – slip control

A further improvement in efficiency can be gained byactivating the torque converter lock-up as often aspossible or practical. This would increase the efficiencyof the transmission even further but would also create aproblem. The torque converter also has a dampingfunction: it isolates the transmission and powertraincomponents from engine induced vibrations and henceimproves the drivability and smoothness of the vehicle.When the torque converter lock-up clutch is engaged,this damping is lost, which could severely affect thedriving comfort at low speeds. Therefore, with a simplelock-up clutch on/off arrangement, a compromisebetween efficiency and economy has to be found.

A recent development, made possible by advancesin control electronics, is a provision for the lock-upclutch to operate progressively. This is implemented viaPWM VBA valves to control hydraulic pressure to thelock-up clutch, allowing progressive, controlled

Stator

Impeller

Turbine

Clutchapplypiston

Clutchfrictionsurface

Lock-up clutch

Torsionalisolatorsprings

Transmissioninput shaft

Engine crankshaft

Stator

Non lock-up position(piston released)

Input Output

Lock-up position(piston engaged)

Input Output

Figure 5.37 Torque converter with lock-up clutch

[647] Chapter 05 25/8/06 16:35 Page 211


5.9 AUTOMATIC GEARBOX TRANSMISSION MANAGEMENT

engagement and disengagement of the clutch. Withsophisticated control algorithms and methodologies, aclutch slip control system can be created to optimiseconverter efficiency, while still providing the superiordrivability of an automatic gearbox.

A typical system uses the converter and clutchtogether and monitors the power distribution via theslip speed. This is established from the speed differencebetween the engine and torque converter turbine.Hence the power transmission distribution can becontrolled and optimised, so that the system canprovide a distribution ratio that gives instantaneouslythe best balance between power loss andnoise/vibration. In particular, this feature can improvetransmission efficiency at low vehicle speeds, with aconsequent reduction in fuel consumption.

5.8.4 Torque converter developmentsThe torque converter remains an important part ofcurrent transmission system developments and itsefficiency can be improved further with the integrationof electronic control. Using the latest computermodelling techniques at the design stage can increasethe basic unit efficiency and specific power capacityconsiderably by optimising internal fluid flows and flowpaths. Further enhancements and developments madepossible through the flexibility of electronic control are:

● idle disconnection of the torque converter – anadditional clutch controlled via the transmissionECU, mounted between the engine and torqueconverter, can be designed into the pump housing(it is similar to a lock-up clutch); this allows

disconnection of the converter and gearboxinternals to reduce drag and friction losses when theengine is idling

● a reverse torque converter – modifying the basicconstruction of the torque converter to allow thestator to transmit force (rather than being mountedon a one-way clutch fixed to the casing) means that,with the turbine locked, the reaction force againstthe stator could be used to provide a reverserotation, which would be useful for CVTapplications, where a planetary gear set is stillneeded within the gearbox to provide this functionat significant extra cost

● an integrated torque converter andstarter–alternator – a multi-function unit thatcombines traditional engine starting and electricalpower requirements, which, with electronic control,provides additional features, such as trafficstart/stop functions, electrical assistance duringacceleration, hybrid drive options and powerregeneration, improving the overall efficiency andadaptability of the powertrain.

Almost all torque converters now contain a lock-up clutch. Using electronics, the slip can becontrolled to improve efficiency and drivability

Under electronic control a torque converter can bemade to produce a reverse rotation and bedisconnected completely at idle

Integrating the torque converter with astarter–alternator provides features such as trafficstop/start and acceleration assistance

Key

Poin

ts

5.9.1 IntroductionAutomatic transmissions have been available in vehiclessince the early days of the development and massproduction of cars. In the UK, the automatic gearboxhas always been seen as a luxury option, while, in theUS, the automatic gearbox is standard on mostpassenger cars.

The basic automatic transmission (see Figure 5.38)consists of a sophisticated hydraulic–mechanical systemincorporating a torque converter (a hydrodynamicdevice) in place of the clutch. A system of planetary(epicyclic) gears provides the various forward andreverse ratios. The whole system is controlled via asophisticated mechanism of valves supplyingpressurised oil to brake bands and clutches forengagement of the appropriate gears. Such systemsworked well and were fully adopted by motormanufacturers before the revolution in microelectronics

which brought in sophisticated engine managementsystems. The additional degrees of freedom provided byelectronic control have enabled the shift processes to beoptimised and have resulted in improvements in theoperation and efficiency of automatic gearboxes.

Where electronic control is implemented the basicmechanical arrangement of the gearbox remains thesame as in hydraulically controlled units (Figure 5.39):drive into the gearbox from the engine via a torqueconverter, planetary gear sets to provide the fixedratios, with gear changes implemented via brake bandsand one-way clutches. With electronic systems, anelectronic control unit (ECU) with a central processingunit (CPU) has a number of inputs from the vehicle,engine and driver. The required gear is calculated andthen implemented via the electrical actuation ofhydraulic valves to provide pressurised oil to theappropriate brake bands or clutches.

[647] Chapter 05 25/8/06 16:35 Page 212

Automatic gearbox transmission management 213

Figure 5.38 Exploded view of ZF 6 speed automatic gearbox fitted to some Jaguar models

Figure 5.39 Gear train layout in the Borg Warner 55 gearbox (a previous generation)

[647] Chapter 05 25/8/06 16:35 Page 213


The main advantages of using electronic control are:

● multiple shift patterns can be selected by the driver(such as sport, and winter patterns, etc.)

● gear shifting can be made smoother● systems can be easily adapted to different vehicles,

reducing integration costs for manufacturers● the hydraulic control system is simplified, with no

need for one-way clutches.

The main technical features of an electronicallycontrolled transmission (Figure 5.40) are as follows.

● With electro-hydraulic selection and implementationof gearshifts, shift points are determined by thecontrol system according to driver requirements andvehicle conditions. Gears are changed via a numberof electro-hydraulic solenoid valves that feedhydraulic pressure to the brake bands and clutches.Electronic control algorithms can optimise theengagement and disengagement of these internalcomponents to allow for inertia, drag, etc. Thismakes for fast, smooth gear changes. Most recentsystems dispense with internal overrunning clutches,since shifts can be performed smoothly and safelypurely with electronic control. This reduces theweight and size of the transmission.

● Electronically controlled systems have asophisticated torque converter lock-up strategy tooptimise transmission efficiency yet still provide thehigh degree of shift quality needed.

● The number of gears has increased to extend therange of gear ratios. With electronic control, shiftsare seamless and of high quality, so the number ofratios can be extended to five or more (mechanicallycontrolled transmissions have three or four gears).

● The system pressure is controlled and adaptedaccording to the transmission status via an electro-hydraulic valve. This improves shift quality andtransmission efficiency over the life of the vehicle(and with high-load or low-temperatureconditions). It ensures that wear of the frictioncomponents (clutches and brake bands) does notimpair the performance of the transmission.

In some electronic control systems, the basicmechanical arrangement of the gearbox remainsthe same as with hydraulically controlled units

Electronic control of an automatic box allowsdifferent modes to be selected by the driver

Electronics simplify the hydraulic control systems

Converter lock-up strategy can be improved

5.9.2 Automatic transmissionmanagement

Basic requirements for electronic transmissioncontrolThe control system must be capable of providing thefollowing features:

● gear selection – shifting to or selection of thecorrect gear ratio for the current driving conditions,taking into account the system information from thesensors at all times

● shift quality – adapting system pressure controldynamically to provide seamless shifts, andimplementing torque converter lock-up formaximum efficiency

● driver input – allowing additional input from thedriver, such as kick-down or sequential shifts

● fault handling – detecting system faults and errors,ensuring all shift operations are plausible,preventing shifts operations that could causedangerous driving conditions, and providing a limphome capability

● an adaptive response – the ability to recognise andadapt to individual driver styles and drivingpreferences.

The basic control functions of an automatictransmission electronic control are as follows.

Shift point controlGenerally, the actual shift point is determined from anumber of shift maps stored in the ECU which can bepre-selected by the driver with a manual switch.Typically, these maps would allow for driving modessuch as ‘sport’, ‘economy’ or ‘winter’. The shift points area function of accelerator position and driving speed andtake into account boundary limitations such as enginespeed limits (maximum and minimum). Theyincorporate an element of hysteresis to preventunnecessary shifting, which could reduce driver comfort.

The shifting operation is time-critical: the finite timetaken to release and apply the friction components is animportant factor and, in the most sophisticatedapplications, is taken into account in the softwarecalibration. The latest generation of transmissions withno overrun clutches require overlap control of thehydraulic clutch operation to allow smooth transitionfrom one gear to the next. This is particularly

Key

Poin

ts

Trondo inc.

4760–95/B

Mechanical

Hydraulics

Electronics

Figure 5.40 Electronically controlled automatic transmissiongearbox

[647] Chapter 05 25/8/06 16:35 Page 214


demanding on the control system and requires highCPU capability and real-time performance.

A further, more recent development is adaptive shiftpoint and transmission control. This means that theECU itself must be capable of adapting the shift pointsaccording to driver and vehicle conditions. Fromevaluation of the driver inputs (kick-down, use of thebrake, accelerator and selector lever) the basic shiftpoints can be adapted to suit the current driver style.The driving conditions can be established via interfacesto other control systems (such as the ABS and enginemanagement ECU). The following conditions can easilybe recognised and control adapted accordingly:

● gradients are detected by comparing current andrequested acceleration with engine torque; thesystem adapts by moving gear shifts to give higherengine speeds

● cornering is detected through differences in wheelspeeds, so requested shifts can be delayed orprevented to optimise vehicle stability

● in winter driving in ice or snow, wheel slip isdetected by measuring wheel speeds, so lower gearsare avoided to reduce tractive force on surfaces withlow friction coefficients

● for traction control the shift strategy is adapted toprovide maximum traction.

Torque converter lock-up controlAs mentioned previously, the torque converter usuallyhas a lock-up clutch to bypass it under certainconditions. This improves the efficiency of thetransmission but must be implemented carefully so thatshift quality and drivability are not adversely affected:the torque converter acts as a damping element fortorsional vibrations at lower engine speeds. Theconverter lock-up clutch has three states: open, closedand controlled. These states are defined and determinedin a similar manner to the gear shift point control andare a function of engine speed and throttle demand. Anoptimised characteristic curve for the converter lock-upprocess is available for each gear and is stored in theECU calibration. The settings take into account the needto optimise fuel consumption and tractive force.

Engine torque control during shiftingThe evolution of automatic transmission with electroniccontrol has resulted in certain developments such astorque converter lock-up and an increased number ofgears. These make it possible to design a sophisticatedand efficient powertrain system. However, thesedevelopments place additional demands on the controlsystem to produce an efficient and smooth shiftingprocess, which can only be realised via a harmonisedengine and transmission control system. The shiftprocess can be optimised through engine interventionduring shifting using torque control. Of course, thisrequires an interface between the engine andtransmission controls and, as discussed previously,current technology supports this easily (via CAN).

The main aims of engine intervention control are:

● to improve shift smoothness and drivability● to reduce wear by shortening slip times and forces● to transmit higher power● to improve synchronisation during shifts.

Torque can be controlled in one of two ways. Mostcommonly it has been controlled by retarding theignition angle from the set position. This can be doneeasily and has a fast response but clearly can only beapplied to gasoline engine vehicles. With more recenttechnology, where there is a torque based functionalstructure for engine control, with a CAN interface,torque can be controlled through a torque interface,which would be used with a number of other vehiclecontrol systems, such as ABS and TCS.

Pressure controlSystem pressure control is an important factor in shiftcomfort, second only to torque control. It is responsiblefor controlling the forces in the friction elements in thegearbox during shifts and is a key factor in maintainingconsistent performance in shift quality throughout thelife of the vehicle. As with shift point control, adaptivealgorithms can be used to allow for life cycle variationsin the transmission (the wear of friction components)and engine (changing tolerances), as well as changes inresponse caused by variations in the temperature of theautomatic transmission fluid (ATF). The systemcompares actual shift times with stored reference valuesand uses this as the basis for making incrementaladjustments to the system pressure up to a maximumdeviation of ≠ 10%. This limit is imposed for reasons ofoperational reliability. The adjustment values are storedin the ECU memory so that they can be reinstated eachtime the system powers on when the vehicle starts up.

Safety functionsSeveral safety related features must be included in thetransmission control system; these generally preventcritical driving conditions caused by driver error orfailed components. Uncontrolled shifting is particularlyundesirable, especially downshifting, which could causeserious problems for the driver, or destruction of thetransmission.

Monitoring of the electronics system andcomponents themselves is particularly important. TheCPU in the ECU is monitored via internal and externalcircuits; the software code execution is monitored forplausibility during run time; and sensors and actuatorsare continuously checked for correct and plausibleoperation.

If a sensor or actuator fails, in most cases substitutevalues can be used and the system switched to ‘limphome’ mode. For example, the transmission outputspeed signal can be substituted by a wheel speed signal,temperature sensor values can be replaced with fixedvalues. This would be sufficient to allow the vehicle tobe driven home or for repair. These safety

[647] Chapter 05 25/8/06 16:35 Page 215


enhancements and error handling methods give anincreased degree of confidence and acceptance in themarketplace for full electronic control systems.

Diagnostic functionsThe diagnostic capability and functionality of currentsystems now forms approximately 30% of the systemsoftware resources. The main reason for this is therequirement for vehicles to comply with on boarddiagnostics (OBD) in order to meet current and futurelegislation. The diagnostic function handles the storageof fault information and the communication of this datato the service tester. The fault memory inside the ECU issub-divided into:

● primary fault memory – non-volatile memory thatcontains fault code and type, plus a warm-upcounter and system flags (an OBD requirement)

● secondary fault memory – one memory slot per faultcode, containing filters, time stamps and flags

● back-up memory – optional memory with deletedfault codes from the primary fault memory

● snapshot memory – for use with diagnostictesting.

The most important monitoring functions for thetransmission are:

● solenoid valve monitoring● pressure regulator monitoring● run-time monitoring of program code.

In the past, access to fault information has beenmanufacturer specific but, with the introduction ofOBD, a standardised protocol for communication to atester, with standard fault codes, is now in general use.

Transmission management is the equivalent ofengine management – but for the transmission

Transmission and engine ECUs are linked to allowcontrol of engine torque during shifting

ECUs communicate using the CAN protocol

Modern systems can recognise and adapt toindividual driver styles

Key

Poin

ts

5.9.3 Case study: Tiptronic gearboxIntroductionTiptronic is the name used by Porsche to describe theirsequential shift automatic gearbox with intelligent shiftstrategy adaptation. This technology was a jointdevelopment between Porsche, ZF and Bosch. The name‘Tiptronic’ is also used by other manufacturers todescribe this technology. The system consists of astandard automatic transmission (i.e. torque converter,epicyclic gear sets) with full electronic control. Thedriver has the choice of driving in automatic mode, or ofusing the manual lever (Figure 5.41) for sequentialshifting. Later versions (Tiptronic-s) also incorporatesteering wheel buttons for up and down shifts.

OperationWhen the system is driven in automatic mode anintelligent driving program runs. This takes intoaccount basic information about vehicle and enginespeed, vehicle acceleration and throttle position, andadapts this according to the dynamics of the vehicle(road resistance) and the ambient conditions (altitude).This information is made available to the Tiptroniccontrol unit, which then decides which shift map to useand what adjustments are necessary to it. This processis carried out continuously and ‘steplessly’ and isinvisible to the driver.

Additional functions are:

● a warm-up map – an optimised transmissionoperation for fast warm up of the engine andcatalyst; upshift points are delayed, and theconverter clutch remains open

● an active shift to sports map – rapid movement ofthe accelerator initiates the sports map; the systemshifts back to the economy map automatically

● kick-down – this does not initiate a change of shiftmap

● downshift during braking – this allows enginebraking assistance

● overheat protection – this is automatically initiatedby restricting engine torque via the engine ECU

Figure 5.41 Tiptronic gearbox

[647] Chapter 05 25/8/06 16:35 Page 216


● gear hold in manoeuvres – lateral acceleration isdetected and the current gear is maintained formaximum stability

● torque reduction during shift – for smooth gearshifts● torque converter lock-up – from second gear up,

depending on engine load and shift characteristiccurve.

Figure 5.42 illustrates the shift curve adaptation processfor the Tiptronic system.

5.9.4 Case study: Honda’sfour-speed all clutch-to-clutchsystem

The Honda clutch-to-clutch automatic transmissionsystem (Figure 5.43) is of particular interest as it is adeparture from the usual design of an automaticgearbox. A traditional automatic gearbox has epicyclicgear sets, with brake bands and clutches to provide theappropriate forward and reverse gears. These are allactivated by electro-hydraulics and controlled by anECU. For a traditional front-engine–rear-drive vehiclethis can be easily accommodated in the powertrainlayout. For transverse front-engine–front-drive vehicles,this arrangement becomes difficult to fit within theengine compartment. Honda’s solution is closer to amanual gearbox in form and is specifically designed foruse in front-wheel drive Honda cars.

Of particular interest is the gear shifting andselection mechanism. The system does not use epicyclicplanetary gears, but is closer in design to a standardconstant mesh gearbox. The system provides fourforward gears plus reverse. The schematic is shown inFigure 5.44.

The shaft layout consists of an input or mainshaft,which drives a secondary shaft via an intermediate gear.Between these two shafts is the countershaft. Theoutput from this shaft drives the road wheels via thedifferential. The countershaft has all the fixed, forwardgears mounted on it, as well as a servo-operated dogclutch selector for selecting forward or reverse.

The fixed gears on the countershaft mate withcorresponding freewheeling gears (of different ratios)mounted on the main and secondary shafts. Thefreewheeling gearwheels are engaged or disengagedwith the shaft via hydraulic, multi-plate clutches, onefor each gear. When a gear is required, the appropriateclutch is engaged and torque is transmitted via thatgear. Changing gear is simply a matter of disengagingone clutch and engaging another.

A simple analogy is to compare this unit with anormal manual gearbox. The countershaft of thisgearbox can be compared to the layshaft of a simple in-line manual transmission and the main/secondary shaftcan be compared to the primary/mainshaft. Instead ofthe required gear being selected and engaged manuallyvia a synchromesh dog clutch from the gear lever, thegears are changed by using a small clutch inside thegearbox for each gear, activated hydraulically andcontrolled electronically with an ECU.

Torque input to the gearbox from the engine isthrough a traditional torque converter with a lock-upclutch for maximum efficiency. The gear selectorclutches are engaged via hydraulic oil pressure andcontrolled with solenoid operated shift valves. Theseshift valves in turn are activated by the automatictransmission ECU, which changes to the appropriategear for driving conditions. It is also possible for thedriver to select the required gear manually (a semi-

Measuredvariables

Level Shift-curve adaption (transmission and converter lockup clutch)

• Data summing• Filtering• Averaging• Weighting

ModulationfactorMF1…MF5MFA

Shift curvesCLC curves

2 Short-term adjustmentSpecial function

Inhibition of trailing-throttle upshiftsahead of curves

Gear retentionin curves

Active shiftingjump to shiftcurve SK5

Upshift duringbraking on low μ

3 Manual tip (nudge) shifting P + Upshifting

– DownshiftingComputer support toprevent excess rpm

D

1 SK5

N

V

SK1

%

SK3SK4

SK2

Engine speed3

45

6

71

2

Throttle valve Vehicle speed120

160200

240

28040

80

Lateralacceleration

Linearacceleration

Figure 5.42 Tiptronic shift strategy

[647] Chapter 05 25/8/06 16:35 Page 217


3rdReverse

Counter shaft

Main shaft

Secondary shaft

Idle gear

Torque output to tyre

4th

D R

Servo

1st 2nd

Figure 5.44 Gear train schematic

automatic). Table 5.3 shows the operating conditionsof the clutches for each of the available gears.

The latest development of this gearbox has reducedits overall length by 22 mm. This does not sound a lot,but in a modern engine bay, where space is already at apremium, this could be very advantageous. In addition,the internal component count is reduced by dispensingwith the first gear one-way clutch. This means that thegearbox needs only four clutches and aservomechanism to provide all the required gears. Inthe development of this latest version specific problemshad to be overcome.

In previous versions of the gearbox a one-wayclutch was fitted to the first gear clutch to improve shiftquality by transferring some torque during upshift.Without this element a drop in torque during the shiftcould be perceived by the driver, which adverselyaffects drivability. This happens because hydraulicclutches have a small, finite, delayed response causedby the necessary refilling of the piston cavity withhydraulic oil before clutch pressure is generated. Thepiston cavity is emptied of oil through a check valveafter each operation to prevent displacement of the

Engine

Wheeldrive

Final drive

Wheeldrive

Torqueconverter 3rd and 4th gear

with clutch packs

1st and 2nd gearwith clutch packs

Main shaft

Countershaft

Secondaryshaft

Figure 5.43 Four-speed all clutch-to-clutch system

Table 5.3 Operating conditions

Elements Engaged Clutch Servo

Actual car 1st 2nd 3rd 4th D/R

1st • D2nd • D3rd • D4th • DReverse • R

piston due to centrifugal effects. To overcome this, acentrifugal cancellation mechanism is fitted to the firstand second gear clutch hydraulics (Figure 5.45), whichallows precise operation and timing of the first andsecond gear clutch operation for smooth upshifting.This precision timing is achieved with high-performance linear solenoids, which can give therequired degree of control. In addition, the mainshaftspeed and acceleration are monitored by the automatictransmission ECU, so that shifting can be monitoredand optimised in real time operation by the ECU.

[647] Chapter 05 25/8/06 16:35 Page 218


Catastrophic failure of the gearbox at speed and aconsequential downshift to first gear due to controlsystem failure would cause over-revving of the engineand possible road-wheel lock. This was previouslyprevented by using a one-way clutch to stop drive

Compensation cavity

Piston cavity

Centrifugalhydraulics

Centrifugalhydraulics

Clutch piston

Figure 5.45 Clutch piston mechanism

through the first gear being engaged under theseconditions. Since the current design dispenses with theone-way clutch, a safety mechanism must be used toprevent engagement of first gear. This is done withdirect hydraulic control using linear solenoids and newshift logic with three solenoids, plus a fail-safe valve.Thus gears are shifted through a combination of eightsolenoid signals (Figure 5.46) and this provides thefail-safe mechanism.

The overall dimensions and form of this gearboxare ideal for front-wheel drive vehicles, providingsafety and improved drivability for the user.

Tiptronic is the name used by Porsche (andothers now) to describe a sequential shiftautomatic gearbox with intelligent shift strategyadaptation

Features such as ‘gear hold in manoeuvres’ can beprovided. With this feature, lateral acceleration isdetected and the current gear is maintained formaximum stability

For transverse front-engine–front-wheel drive, asolution in use by Honda is closer to a manualgearbox in form

Key

Poin

ts

1st accumulator

Shift valve B

Shift valve A

Shift valve D

Shift valve C

CPC valve B

CPC valve A

Shift valve E

1st clutch

2nd clutch

3rd clutch

4th clutch

Pressure switch

Shift solenoid B

Shift solenoid C

Shift solenoid A

Linear solenoid B

Linear solenoid A

LG solenoid

Modulator valve

Servo valveRegulator valve ATF pump

Pressure switch

2nd accumulator

3rd accumulator

4th accumulator

Figure 5.46 Hydraulic control system

[647] Chapter 05 25/8/06 16:35 Page 219


5.10.1 IntroductionA problem with any combustion engine is thatmaximum power, torque and fuel economy all occur atdifferent engine speeds within the range of normaloperation. In a conventional, selectable, fixed ratiogearbox, in any particular gear, the engine speed andvehicle speed have a fixed relationship, so maximumtorque, power or economy can be achieved only at oneroad speed per target in each gear. Driving a car along aroad is transient in nature, so achievement of any ofthese targets for any reasonable period of time, will benearly impossible.

Figure 5.47 shows that only a limited fit to the idealcurve can be achieved with a standard transmission. Acontinuously variable transmission (or CVT), as itsname implies, can provide an infinite number of ratiosbetween some absolute limits. Thus this transmissionsystem can give a powertrain performance curve that iscapable of matching the ideal tractive effort curve for aparticular vehicle.

CVT transmissions can be operated mechanically,hydraulically or electrically and various designs havebeen developed, proposed and utilised, although theyhave never made the market breakthrough that wasanticipated. One popular design, adopted by DAF andVolvo for small cars, was the Variomatic system shownsimplified in Figure 5.48.

The Volvo system satisfied the basic requirementsfor a CVT for light vehicle applications. However, thetechnology was considered ‘quirky’ within the marketplace and the system layout was not easily adaptablefor general application in the front-engine–rear-drivecars that were popular at the time that the system was

developed. The system needed a specific powertrainlayout (a front engine, with rear drive via a transaxle(Figure 5.49)). Composite rubber drive belts were used,which needed adjustment and replacement. These beltslimited the maximum torque transfer possible, so thesystem was suitable only for small cars.

The next generation of CVT developed for compactcars used a steel belt instead of a rubber one. Thepackaging is clearly different as the application is forfront-wheel drive vehicles (Figure 5.50) but theprinciple of operation is exactly the same as that of theVariomatic system. One important difference is that thissteel belt is actually used in compression to push, ratherthan pull, the drive force. A major advantage of thissystem is that parasitic losses inside the gearbox aresignificantly reduced compared with the losses in atraditional automatic gearbox.

The system incorporates the differential assemblyand forms a compact single unit. The steel belt limitsthe torque that can be transferred, so the system can beused only with smaller engines of up to 1.6 litresdisplacement.

5.10 CONTINUOUSLY VARIABLE TRANSMISSION (CVT)

Figure 5.47 Tractive effort curvesa Ideal tractive effort curveb Curve for conventional stepped transmission

Figure 5.48 CVT system

[647] Chapter 05 25/8/06 16:35 Page 220

Continuously variable transmission (CVT) 221

Basic outline of CVT transmission operationThe CVT systems described above both work on thesame basic principle. They have similarly constructeddriving and driven members that consist of opposedcones both mounted on the same shaft. Theseeffectively form pulleys with v-shaped grooves, with abelt running between them to transmit torque. Thecones that form the pulley on the primary (driving) sidecan be moved closer together or further apart: it isthrough this mechanism that the overall gear ratio canbe varied infinitely. In operation, with a fixed widthbelt, if the pulley sides (cones) are moved apart, theeffective working diameter of the belt is reduced. If theyare moved together, the diameter is increased (Figure5.51).

Gear ratios are shifted via an actuator: by movingthe relative positions of the cones on the primary pulley,the actuator changes the effective working diameter ofthis pulley and consequently the gear ratio. Thesecondary (driven) pulley has the cones spring-loadedagainst each other, which maintains the correct belttension as the ratio varies with the fixed length belt.

The CVT has great potential to reduce fuelconsumption and emissions because the engine can beoperated continuously at its optimum operating point.The greatest problem is the power loss caused by theinternal energy requirements of the transmission. Withelectronic systems, oil flow and pressure can be

Figure 5.49 Variomatic transmission

Figure 5.50 Ford CTX transmission

[647] Chapter 05 25/8/06 16:35 Page 221


It prevents excessive forces being used, which wastepower and reduce efficiency, but maintainssufficient pressure to prevent belt slip.

● A driving program – this enables preselected drivingmodes to be implemented as well as adaptivefunctions, and the selection of fully automatic orsemi-automatic (sequential) modes.

● Torque converter lock-up – this improvestransmission efficiency by bypassing torqueconverter slip.

● Pump control – control of the pump flow rateimproves transmission efficiency and preventsexcessive flow at high speed.

● Limp home mode – in the event of failure, limphome and fail-safe features must be built into thecontrol system structure, in addition to diagnosticmonitoring capability.

A CVT system has only two gears – forward andreverse

CVT transmission can be operated mechanically,hydraulically or electrically

Gear ratio shifting is implemented through anactuator which alters the relative positions of thecones

In most CVT boxes the drive belt is pushed – notpulled

CTX (the Ford transmission) stands for constantlyvariable transaxle

Key

Poin

tsb2

a2b

a1a

1

2

3

b1

a ‘Low’ ratiob ‘Overdrive’ ratio1 Input (primary) pulley2 Push-belt or chain3 Output (secondary) pulley

a1, b1 ‘Low’ ratioa2, b2 ‘Overdrive’ ratio

Figure 5.51 Simple CVT

controlled accurately to suit the working conditions ofthe gearbox, so overall efficiency can be significantlyimproved. A most important factor is the availability ofreliable sensors and actuators to support reliableimplementation of the control strategy.

Basic functions of the transmissionAs well as the variable-ratio pulley assembly, additionalcomponents are needed in the transmission system forit to be suitable for road vehicles. These are:

● mechanisms to provide a neutral gear(disengagement of the drive to the road wheels) anda reverse gear

● an arrangement to allow the progressive take-up ofdrive from standstill, such as an electric or multi-plate clutch or torque converter, which should alsoallow the vehicle to ‘creep’ at low speeds formanoeuvring

● an appropriate fixed gear assembly to drive the finaldrive/differential at the appropriate speed anddivide torque equally between the driving wheels

● a suitable control system to select the correct ratioaccording to driver’s requirements and drivingconditions. This could be a hydraulic system, but thepreferred solution for a modern vehicle would beelectronic or electro-hydraulic.

Electronic control functionsThe control system for a CVT could also incorporate thefollowing specific requirements.

● Contact pressure control – this provides adjustmentof the belt clamping force in relation to load forces.

[647] Chapter 05 25/8/06 16:35 Page 222


5.10.2 Case study: Audi Multitronictransmission

IntroductionAudi has taken the principle of the basic CVT systemand extended its capability to make it suitable forapplications in larger vehicles where the torque transferrequirements are much higher. By adapting andimproving some of the basic mechanical parts,particularly those that limit the capabilities of thetraditional CVT, and with the addition of sophisticatedelectronic control, Multitronic transmission (Figure5.52) gives improved fuel consumption compared witha traditional automatic transmission. The accelerationperformance of the vehicle is also marginally improvedwhen compared with a normal five speed manualtransmission.

Overall the system provides very wide ranging gearratios of up to 6.05 to 1, both higher and lower ratiosthan any other automatic transmission. With thissystem Audi has produced a highly efficient CVT withbroader application possibilities and some importanttechnical developments and improvements.

System highlights● Link plate chain drive – The component that limits

torque transfer in a CVT is the connecting elementbetween the driving and driven parts of the variator.Early designs used rubber belts (DAF 66, Volvo340). Later designs, as mentioned above, use a steelthrust belt designed by Van Doorne. Both are limitedin terms of the maximum torque that they cantransfer. For the use in the Audi application the beltmust transfer nearly 300 Nm of torque! The solutionwas the development of a link plate chain drive,designed jointly with LuK (Figure 5.53). This chainis constructed of 1025 links with 75 pins, all made

of high-strength steel. Torque is transferred throughthe contact between the pulley flanks and the endsof the pins.

● Multi-plate clutch – A wide range of gear ratios isavailable, so the torque converter is replaced by amuch more efficient, hydraulically operated, wetmultiplate clutch. This avoids the traditional lossesassociated with a torque converter but also, withelectronic control, allows implementation of anumber of starting strategies according to driverpreference. These strategies are established viamonitoring of throttle demand and rate of change.An additional feature is a ‘creep’ function that isautomatically initiated by the electronic controlsystem for low speed manoeuvres.

● Dual piston variator with torque sensor – Thistechnology ensures that the variator grips the chainwith sufficient pressure, depending on torquetransmitted, to prevent slip but with no morepressure than is necessary. This is measured by thetransmission ECU monitoring the clamping forces(via the torque sensor) and variator speeds (viaspeed sensors); the hydraulic pressures are thenadjusted accordingly. The advantage is that thiskeeps the internal gearbox power requirements tothe absolute minimum, thus increasing efficiency, aswell as reducing heat build-up. The dual pistonarrangement for the variator (one piston forclamping the chain and a smaller piston for ratiochange) ensures that the required dynamic responsecan be obtained with a small, more efficienthydraulic pump (Figure 5.54).Figure 5.52 Multitronic system overview

Engine

Starting clutch

Variator

Link-plate chain

Oil pump

Hydraulic control

Electronic control

Figure 5.53 Link plate chain

[647] Chapter 05 25/8/06 16:35 Page 223


Figure 5.54 Variator with torque sensor

● DRP (dynamisches regelprogramm – dynamiccontrol program) – The transmission controlmodule (mounted on the gearbox itself) has ahighly dynamic program for calculating the targettransmission input speed. Based on driver input andoperating conditions, the system can determine theappropriate driving style and shift pattern selection,for example performance or economy. The drivingfeel is designed to be similar to the feel of a manualshift mode. The system also eliminates the so-called‘rubber band’ effect typical of traditional CVTs: itensures that the engine speed increasesproportionally with road speed in a similar way to amanual transmission when accelerating. Probablythe most intriguing feature is the manual, sequentialshift mode. This allows six predefined ratios (eventhough there are an infinite number available) to be

selected by the driver manually with the shift leveror steering wheel mounted buttons. This gives thedriver more control over shifting and a more sportydriving experience (Figure 5.55).

5.10.3 Case study: Nissan electroniccontrol

IntroductionNissan has recently been at the forefront of thedevelopment, improvement and integration of CVTtechnology in its vehicle range. As noted earlier, thistechnology is only suitable for smaller, lower-powervehicles. Nissan has added significant technical featuresto the basic belt drive CVT to extend its application tomore powerful vehicles up to the 2.0 litre engine class.In addition, the inclusion of an electronic control systemhas increased the efficiency, improved performance andextended the driving appeal of a CVT equipped vehicle.

Technical highlightsThe main focus of the development was to:

● improve power and economy● improve acceleration in terms of performance and

drivability● provide a manual, sequential shift mode.

To achieve these targets the technical developmentswere as follows.

1 Torque converter with lock-up clutch – Thiscomponent was added to the transmission system toprovide improved acceleration from rest; anadditional benefit was smoother transmissionthrough the damping properties of the torque

Driver inputEconomical

Sporty

Evaluation of signal from theaccelerator pedal module.

Acceleration rate and position of accelerator pedal

Calculation of target transmission input speed

Influencing factors(e.g. engine warmup)

Transmission control

ResultActual transmission input speed (and hence engine speed)

Evaluation of road speed androad speed changes

Evaluation of road speed androad speed changes

Vehicle operating stateAccelerationDeceleration

Constant speed

Vertical section of routeUphill

DownhillLevel

Figure 5.55 Dynamic control program

[647] Chapter 05 25/8/06 16:35 Page 224


converter. An electronically controlled lock-upclutch ensures that the efficiency of the transmissionremains high. This clutch also features a highlydurable facing material that allows lock-up at lowerspeeds than is normal (down to 20 km/h), whichgives a further improvement in economy.

2 Expanded belt width – The belt width is increasedby 25%, and, by optimisation of the pulley ranges,the wider belt can easily accommodate the torquegenerated by a 2.0 litre engine.

3 Electronic control system – The system consists ofsophisticated electro-hydraulic elements with a fullyelectronic control system (ECU). This systemmonitors and controls all aspects of transmissionoperation to optimise performance and economy.The system overview is shown in Figure 5.56.

4 Manual mode – To enable greater driverinteraction, a multi-speed manual mode is included.This has sequential shifting to allow the driver toselect and hold gears during sports driving.

SummaryThe system shows measurable benefits in terms ofeconomy and performance. The fuel economy isimproved by 20% compared with the standard four

speed automatic. This is possible because of theextended range through which the lock-up clutch canoperate (reducing slip losses) and because the systemhydraulic pressure adapts to engine load (via theelectronic control system) to reduce internal energylosses.

The torque multiplication capability of the torqueconverter improves acceleration performance by 30%,and drivability is increased through the smooth, steplesstorque delivery of the CVT system. The manual shiftmode is an enhancement which will be appreciated bydrivers who want a greater level of interaction for amore rewarding driving experience.

The Audi Multitronic system provides a very widerange of gear ratios (from 6.05:1 to 1:1). Thesystem does not use a torque converter

The Nissan system monitors and controls allaspects of transmission operation to improveperformance and economy

Using a wider belt can increase the torque capacityand allow CVT systems to be used with largerengines

Key

Poin

ts

Primary pulley speed Primary

pulley

Secondarypulley

Forwardclutch

Reverseclutch

Torqueconverter

Control valve assembly

Lubrication and cooling systems

Input signal system

oil pump

Secondarypulley speed

CVTcontrol unit

Line pressuresolenoid

Lock-upsolenoid

Steppermotor

Shift controlvalve

Lock-upcontrol valveEngine

speed

Throttle position

Clutch regulator valve

Torque converterregulator valve

Pressureregulator valve

Manualvalve

Figure 5.56 System overview

[647] Chapter 05 25/8/06 16:35 Page 225


5.11.1 IntroductionA hybrid powertrain is one with more than a singleprime mover and can incorporate a number oftechnologies for power conversion and energyaccumulation. The prime objective of a hybrid drive isto harness the advantages of each particular drivetechnology under its optimum operating conditions.The consequential increase in efficiency of thepowertrain as a whole offsets the increased initial costand reduces the harmful exhaust emissions from thevehicle.

The internal combustion engine and electric motorcan both be considered as torque sources or primemovers; their respective attributes are show in Table5.4.

The hybrid drive solution offers great potential forimproving fuel consumption and reducing emissionsduring low- to medium-speed operation of the vehicle.This is because the internal combustion engine hasgreatly reduced efficiency under part-load conditions. Itis likely that hybrid powertrain designs will becomecommonplace in future as the technology develops andimproves. Figure 5.57 shows the classification of hybriddrives.

A full hybrid drive powertrain with integratedcontrol needs to be a fundamental part of a vehicle’sinitial design. It is is impossible to introduce a hybriddrive retrospectively into an existing vehicle withoutconsiderable reworking of the powertrain (toaccommodate both the internal combustion engine andthe electric drive) as well as the vehicle chassis (toaccommodate the energy storage medium – thebattery).

5.11.2 Light hybridsA compromise can be found in the form of theintegrated starter–generator (ISG), which is a naturalprogression in the development of automotive electricalsystems because of the continuously growing demandfor electrical power in the modern vehicle. It is expectedthat electrical power of up to 10 kW could be needed infuture and the standard 14 V electrical system will need

Table 5.4 Internal combustion engine and electric motorcomparison

Internal combustion Electric motorengine

Torque Uni-directional Bi-directionalResponse Slow: > 300 ms Fast: < 5 ms

at low speedCharacteristic Maximum torque Maximum torque

not available available at lower at lower speeds speeds and from zero

5.11 LIGHT HYBRID POWERTRAIN TECHNOLOGY (STARTER–GENERATOR)

Hybrid electric vehicle system functionality classification

Full hybrid

Mild hybrid

Micro hybrid

Generatorstop–start

Stop–startautomatic Cold start Direct start

Boostregeneration

Coasting +regenerative

braking

Pureelectricdrive

Pure electrical driving

Drivability

Efficiency

Regenerativebraking

• Combustion engine• E-machine(s)• Combination

Battery andcapacitorcharge

Electric boostand launch

assist

Intelligentbelt tensioner

SOCcontrol

Batterymanagement

Figure 5.57 Hybrid electric vehicle classification

[647] Chapter 05 25/8/06 16:35 Page 226

Light hybrid powertrain technology (starter–generator) 227

to be supplemented or replaced by a 42 V system(Figure 5.58). The separate starter and generator unitsused in modern vehicles can be easily integrated into asingle unit to start the IC engine, provide on-boardelectrical power and recharge the battery. This unit,depending on the installation, could also provide anumber of additional features such as:

● a short-term power boost function to supplementthe internal combustion engine during acceleration,particularly in the lower speed ranges

● a retarder to supplement the brakes, which couldalso regenerate energy during deceleration: theelectric motor/generator is highly efficient at doingthis and can achieve efficiencies of up to 80% underthese conditions

● an engine stop/start function for use in stationarytraffic to reduce harmful emissions; engine startingtimes of less than 0.5 seconds can be achieved.

A considerable advantage of this technology is that itcan be integrated into an existing design with minimaleffort. Basically the system consists of:

● a three-phase AC motor integrated with the internalcombustion engine design

● an AC/DC converter which rectifies the ACelectricity generated by the three-phase motor

● a DC/DC converter that provides the requiredvoltage levels

● a control electronics system for the ISG powertrainsystem

● an energy management system controlling the ISGand the vehicle power requirements.

There are two principal designs for an ISG currentlyproposed or under development by manufacturers (seeFigure 5.59), which are discussed in more detail below.

1

Micro hybrid

Generator/starterBelt driven SG

Intergrated SGMild hybrid

Micro hybrid functionalities+ Boost+ Regeneration

stop–start

1 Combustion engine2 Electric machine (SSG)3 Air cooled ECU4 Smart switching unit5 14 V battery6 DLC (supercapacitors)

1 Combustion engine2 Water cooled ECU3 DLC (supercapacitors)4 Electric machine5 Clutch6 Gearbox

2

3 4

5

6

1

2 3

4 5

6

Figure 5.59 Micro and mild hybrid systems

42 V

14 V

High levelload

Low levelload

DC link capacitor

DC

DC

DC

AC3~

Inductionmachine

Sensors

ISG ECUControl ECU

In/out

CAN bus

Hybrid drive

Figure 5.58 Integrated starter–generator control system

[647] Chapter 05 25/8/06 16:35 Page 227


that, with these constraints, this technology must beincorporated into the powertrain design. It would bemore difficult to incorporate this system than someothers and retrofit would not be feasible.

The system has a high performance, providing up to200 Nm of torque for short durations. The unit iscapable of generating up to 8 kW of electrical powercontinuously for the electrical system. A flywheelstarter–generator is usually fully integrated with avehicle’s power management system. Managed by thepowertrain control system as an available torquesource, the system is integrated via a standard interface,such as a controller area network (CAN). A torquemanagement system will then distribute the torquefrom the starter–generator and the internal combustionengine according to driver requirements and drivingconditions.

In addition to providing engine starting andstop/start functions (as does the belt drivestarter–generator), the flywheel starter–generator canbe configured to improve efficiency, because it is farmore integrated with the powertrain control system.Additional features that can be added to do this includethe following.

● A torque booster can provide additional torque tocomplement an internal combustion engine undercertain conditions. The torque booster also gives thevehicle designer the option of using a smallerinternal combustion engine. An important factorhere though is the battery state: the chargecondition of the battery must be monitored carefullyto avoid any significant deterioration in vehicleperformance.

● Regenerative retarder – The ISG can be used torecover energy during braking, and can beincorporated into an active brake managementsystem for maximum benefit. Again though, thebattery’s state of charge is a critical factor: theremust be enough capacity to store all the energyrecovered.

Figure 5.60 Belt driven starter–generator

5.11.3 Belt driven starter–generatorThe belt driven starter–generator is also known as amicro-hybrid system (Figure 5.60). It consists of a beltdriven electric machine mounted at the front of theengine, with a control unit and belt tensioner. It providesstart/stop functionality as well as starting and chargingfunctions. The system works at 14 V and as such doesnot require major modifications to the vehicle electricalsystem: the standard starter battery technology is used.This means that the system can be retrofitted to existingengine designs, since it can be fitted in place of aconventional alternator using the same fixings, freeingup the space where the starter would normally appearon the engine, which also gives a small weight saving.

In starter mode, the machine can start the enginesilently (with its belt drive) and up to three timesquicker than a traditional starter with a ring gear. Ingenerator mode, the machine is over 80% efficient, upto 15% more so than a standard alternator because ofthe advanced power management system.

In operation, when the vehicle is at a standstill intraffic, the engine cuts out, so all sources of noise andpollution are eliminated and fuel consumption is zero.When required, the engine restarts automatically andimmediately in less than half a second; this is possiblebecause of the high torque/inertia ratio of the electricmachine. In modern traffic conditions, in suburbanareas, vehicles are standing for up to 35% of journeytime: overall fuel consumption in these conditions canbe improved by up to 10%.

5.11.4 Flywheel starter–generatorAlso known as a mild hybrid system, the flywheelstarter–generator is mounted in place of the vehicleflywheel and acts directly at the crankshaft between theengine and transmission (Figure 5.61). Typically thisunit is a highly efficient synchronous or asynchronousAC machine running at up to 60 V. Normally a system ofthis type is integrated within a vehicle electrical powersystem incorporating a higher voltage line, at 42 V, formore efficient transmission and conversion. It is clear

Figure 5.61 Flywheel starter–generator

[647] Chapter 05 25/8/06 16:35 Page 228

Electronic differential and four-wheel drive control 229

The ISG system can be part of an intelligent powertrainmanagement system, controlling and managing allavailable torque sources, and deploying the mostefficient one depending on operating conditions. Aswell as managing traction demands, the system alsomanages the vehicle’s power usage. It controls theentire drive train by enabling communication betweenall electronic units in the vehicle. It constantly monitorsand exchanges data between the internal combustionengine, the powertrain, and the electrical controlsystems, such that the auxiliary drive system always hassufficient electrical power for the engine to be able toself-start when required.

A hybrid powertrain is one with more than oneprime mover (an engine and a motor)

ISG stands for integrated starter–generator

A mild hybrid can provide a short-term powerboost, a retarder function to supplement brakesand an engine stop/start function to reduceharmful emissions in static traffic

A micro-hybrid system has a belt driven electricmachine mounted at the front of the engine (acombined starter and alternator)

Key

Poin

ts

5.12 ELECTRONIC DIFFERENTIAL AND FOUR-WHEEL DRIVE CONTROL

5.12.1 IntroductionThe rapid development of electronic systems controllingthe modern motor vehicle powertrain has progressivelyextended from the engine, to the transmission and morerecently to the final drive system. Electronic control isintegrated into modern vehicles to ensure that torquedelivered by the powerpack (engine and transmission)is efficiently and effectively transferred as tractive force.Vehicle electronic control systems interact to maximisethe potential of the vehicle propulsion system toimprove safety and performance.

Vehicle wheels are often in contact with surfaceswhere the adhesion varies significantly, so optimisationof the tractive force is essential. Nearly all vehicles havea standard mechanical differential, which, undernormal driving conditions, provides uniform torquedistribution between the driving wheels, givingpredictable vehicle behaviour. However, the force isalways distributed evenly, irrespective of speed; so, ifone wheel lacks grip and slips, no force is transmitted,so the other wheel also transmits no force and thevehicle is immobilised. Stability is another importantissue with modern vehicles. Powerful engines canproduce more force to accelerate the vehicle than thesystem can transmit. Depending on the road surfaceand driver behaviour, the wheels can spin, leading toloss of traction and erratic performance.

Several electronic control solutions are available todeal with these problems, and to enhance vehicleperformance as a whole.

Automatic brake differential (ABD)/tractioncontrol (ASR)An automatic brake differential is not a differentiallock in the traditional sense: it is not like a mechanicallock, which positively locks and prevents differentialaction completely. ABD uses the anti-lock brake systeminfrastructure (wheel speed sensors, hydraulics, etc.)and detects any difference between the rotationalspeeds of the driving wheels. When it detects a speed

difference, the system can activate the brake on theoffending wheel to counteract the slip. This is done ina progressive manner, allowing controlled transfer oftorque via the open differential to the driving wheelwith the greatest adhesion. There are some limitationswith this system: it cannot operate effectively at zerospeed and therefore is of no use for true off-roadapplications; and because of the temperature limits ofbraking systems, it is not available at higher vehiclespeeds (greater than 25 mile/h).

The system normally operates in conjunction withthe engine control electronics to provide active tractioncontrol: as well as decelerating the slipping wheel withthe brake, the ECU can actively reduce engine power tofurther control the torque distribution. The enginetorque is usually controlled via the electronic throttle(e-gas) actuator and the ignition timing (spark retard).The control units communicate using the high speedCAN bus line.

Electronic control of differentialThe above system is not specifically suitable for off-roadvehicles or sports utility vehicles (SUVs). So, for suchuse, it needs to be supplemented by another system,capable of sustained prevention of the differentialaction on the driving axles for maximum traction at lowspeed. Mechanical solutions here include differentiallocks and limited slip differentials (LSDs). However,these solutions require manual intervention by thedriver (the former) or can only respond in a simplisticway (the latter). An electronic control system providesthe following benefits:

● a smart system that can respond proactively, notonly to speed differences, but also to otherparameters monitored by the vehicle electroniccontrol systems (wheel speeds, throttle demands,engine load, etc.)

● a differential action that can easily be integratedwith other vehicle systems, such as ABS, tractioncontrol and stability control.

[647] Chapter 05 25/8/06 16:35 Page 229


Multiplate clutchDifferentialgears

Locking balls

Current coil

Figure 5.62 Electronically controlled differential

The system can be configured in a number of ways butFigure 5.62 shows a solution for a live rear axle. Itconsists of a standard differential, but with theaddition of multi-plate clutches between the planetwheels and differential cage. These clutches areengaged electromagnetically by a current in the coil.This current determines and is proportional to thetorque transfer. When activated, the current coilcreates a magnetic field which pulls the cone intofrictional engagement with the differential cage. Thefrictional torque created by the cone causes the ballsto ride up a ramp machined into the side gear. Thelateral movement of the side gear applies a force ontothe centre block and that load is transmitted from thecentre block to the opposite side gear, compressing theclutch pack and locking the differential.

These systems could either be directly integrated in apowertrain control unit (PCU), or they couldcommunicate with each other on a high-speed bussystem (e.g. CAN).

Differential control systems can operate inconjunction with the engine control electronics toprovide active traction control

The differential action can be integrated with ABS,traction control and stability control

Multi-plate clutches between the planet wheelsand differential cage are usually controlledelectromagnetically via current in a coil

5.12.2 Case study: Porsche tractionmanagement (PTM)

IntroductionThe Porsche Cayenne is equipped with one of the mostsophisticated four-wheel drive systems. The systemmechanics are similar to any off-road or SUV with full-time 4WD, and the system intelligence lies in theelectronic controls. The fundamental principle isextremely simple: to actively distribute torque to thosewheels that can utilise it most efficiently.

Key

Poin

tsB

ias

torq

ue, l

b in

Current, amps

Dynamometer test showing the relationship between bias torque and coil current. The tight wheel was held to zero RPM, the differential was driven at 50 RPM, and the loose wheel was allowed to freely rotate at 100 RPM. The coil current was slowly increased until the tight wheel torque reached 18,000lb in.

Figure 5.63 Control current: excitation current v locking torque

This design is particularly intelligent as it does notrequire any additional hydraulic or pneumatic powerand hence is highly efficient. It can easily beintegrated into an existing installation simply byadding the hardware and driver stage to an existingpowertrain ECU. It does not need to have its ownspecific control unit.

Figure 5.63 shows excitation current plotted againstlocking torque.

Integration of safety systemsA clear benefit of these solutions is that it is possible toconnect the differential control system to other vehicledynamics controllers to provide a harmonised efficientsystem with maximum tractive force control, providinggreater safety. The systems that could be linked include:

● anti-lock braking (ABS)● traction control (ASR)● brake force distribution (EBD)● cornering stability control (CSC)● dynamic drift control (DDC).

[647] Chapter 05 25/8/06 16:35 Page 230

Electronic differential and four-wheel drive control 231

The system consists of:

● a transfer case incorporating epicyclic differentialand range selection, with an electrically operatedprogressive differential lock via a multiplate clutch(Figure 5.64)

● a front axle with electrical (ABD) lock capability● a rear axle with electrical, progressive differential

lock via a clutch● harmonised electronic control and monitoring of

brakes, traction and stability.

The front final drive does not have a mechanicaldifferential lock. Potential slip of one front wheel isprevented by the automatic brake differential (ABD, seeabove). The negative effects of a mechanical differentiallock, such as increased weight and limitations onsteering and handling can thus be avoided. If bothwheels on one axle are in danger of slipping, the controlsystem intervenes through the electronic enginemanagement and reduces power in order to maintaingrip. By combining these functions, the PTM ensuresoptimum traction on almost every surface.

All the PTM functions are fully automatic, resultingin inherently more dynamic handling characteristics, aswell as greater active safety.

5.12.3 Case study: Haldex couplingIntroductionFour-wheel drive transmissions (4WD) have been inproduction models of performance cars for some 25years. During this time the mechanics of thesetransmissions have evolved to improve further thetraction efficiency of the system as a whole. The heartof a 4WD system is the centre differential which dividesthe torque between the front and rear axles. Over theyears this technology has evolved from simple lockingdifferentials, to limited slip differentials, to viscousdifferentials, and, the latest iteration, the Torsendifferential. This development process has improved theperformance of 4WD vehicles, but all of these systemssuffer from the same shortfall: wheel slip can berecognised only by a difference in shaft speeds betweenthe front and rear axles. The problem is that althoughthese systems can sense slip, they are not able to detectthe cause of it.

Developments in electronics have moved thistechnology forward significantly with the introductionof the Haldex coupling, which is now used by a numberof large motor manufacturers. Owing to themechatronic nature of this system, the coupling is fullycontrollable and can take into account not only slip butalso vehicle dynamic state.

Overview and operationThe Haldex coupling is mounted in the rear differentialassembly and is driven by a propshaft connected to thefront transaxle (Figure 5.65). Engine torque istransmitted via this propshaft from the front transaxle,directly from the front differential drive. There iseffectively no centre differential. Inside the unit theinput shaft is connected to the output shaft and rear axlevia the Haldex multi-plate coupling; torque to the rearaxle is provided and controlled through this coupling.

The basic operation of the mechanics and hydraulicsof the system are as follows (Figure 5.66). When aspeed difference occurs between the input and outputshafts of the coupling, the swash plate drives the smalloil pump plungers and this generates oil pressure. Thispressure is used in the clutch pistons to compress the

Figure 5.64 Four-wheel drive traction management unit

Drive torear axle

Drive tofront axle

Engineinput

OperationThe system provides permanent torque distributionbetween the front and rear axles which, under normalconditions, is a 38/62% front/rear split. This rear wheelbias provides stable grip on almost any surface andgives the driver the level of road feel and chassisfeedback that would be expected from a high-performance vehicle.

The transfer boxes provide high- and low-gear ratiosfor on/off road use. The set of ratios is selected with atoggle switch near the gear lever. The functions areprogressive and sequentially selected and shift between,for example, on- and off-road conditions.

● The first actuation of the switch selects ‘low’ rangethrough the transfer box. This is engaged via aselector fork and electric motor actuator. Inaddition, with the vehicle sensors, the activeelectronic system continuously measures traction atthe wheels, as well as vehicle speed, lateralacceleration, steering angle and operation of theaccelerator pedal. From this, the systemautomatically calculates the optimal degree oflocking for the differentials at the drive axles. In thisway, more power is applied at the front or rearwheels, depending on the driving situation.

● The second actuation of the switch fully locks thecentre differential using a clutch and an electricmotorised actuator.

● The third actuation of the switch fully locks the reardifferential, again using a clutch and actuator formaximum off-road traction.

[647] Chapter 05 25/8/06 16:35 Page 231


Transverse FWD layout

Propshaft to rear axle

Rear axle includingHaldex coupling

Figure 5.65 Four-wheel drive control layout

clutch plate pack and hence engage the clutch. Once theclutch is locked then 100% of the torque is transmittedthrough the coupling. The clutch pressure is maintainedby a regulating valve. The system includes an electric oilpump which runs at engine speeds above 400 rev/min.This provides a background system pressure of 4 barwhich pre-loads the clutch pack to remove any play andensures a minimum system response time.

Electronic control systemThe ECU for the 4WD system/Haldex coupling ismounted directly at the rear axle casing. This is in linewith current trends for powertrain control systems:there is a fully integrated mechatronic control moduleincluding electronics and hydraulics, with minimalexternal connections and interfaces, to improvereliability. Figure 5.67 shows the system overview.

The sensors and interfaces used in the system are:

● a CAN interface to the engine ECU, providinginformation about engine speed and load throughcrankshaft position and throttle position sensors

Figure 5.66 Hydraulically controlled clutch assembly

Wet multi-plate clutch

Clutch piston

Hydraulic piston pump

Controllable throttle valve

Oil pressure sensor

Haldex LSCOther onboardsystems

Oil temperature sensor

Hand brake switch

Brake light switch

Enginecontrol

ABSESP

4x wheel speedsBrake light switchABS activeESP activeYaw rateLateral acceleration

Steerwheel angle

Engine torqueEngine speedAccelerator position

ECU

CAN

Engine

Figure 5.67 Electronics control system links

[647] Chapter 05 25/8/06 16:35 Page 232

Transmission diagnostics 233

● a CAN interface to the ABS ECU, providinginformation about the vehicle dynamic state,including wheel speed, vehicle acceleration, brakestatus (on or off) and handbrake status (on or off)

● a temperature sensor for the Haldex coupling,enabling the coupling to react to changes in oiltemperature, and to detect high temperature oroverload conditions so the system can protect itselfin extreme conditions.

The actuators are:

● a positioning motor for the regulator valve(controlled directly by the ECU) which controls theclutch pressure and hence the operation and controlof the system

● an oil pump, which provides background systempressure and lubrication while the unit is in standbymode.

The system’s main features and advantages are:

● active torque distribution via an electronicallycontrolled multi-plate clutch, with instant activationand high torque transfer for maximum tractionwhen required; an alternative separate off-roadmode (switch) can lock the coupling

● an extremely fast responding, highly dynamicsystem but with the feel of a normal two-wheeldrive car for predictable handling

● no strain on system components or the vehicleduring low-speed manoeuvring

● compatibility with different tyre sizes, for example,when a space saving spare wheel has to be fitted; analgorithm in the ECU detects differences in thediameters of the tyres and adjusts the characteristicsaccordingly

● no restriction on towing or testing (a chassisdynamometer or brake test) of the vehicle, becausethe system is inactive when the engine is notrunning

● full compatibility with other vehicle dynamiccontrol systems such as ABS, EDL, TCS EBD andESP; the coupling communicates on-line with safetysystems in the vehicle.

The fundamental purpose of four-wheel drivetraction management is to distribute activelytorque to the wheels that can utilise it mostefficiently

Four-wheel drive control systems are fullyintegrated with other dynamic control systems formaximum driver safety and benefit

Key

Poin

ts

5.13 TRANSMISSION DIAGNOSTICS

5.13.1 IntroductionAlthough they are sophisticated pieces of modernelectromechanical technology, like anything else,electronically controlled transmissions can and do gowrong. Generally, considering the environment inwhich they operate, these units are extremely reliable.In most cases, failure of a sensor or actuator initiatesthe limp home failure mode which means that thevehicle can still be driven (although with reducedperformance) for repair or to get the occupants home.

An important point though is that the workingcomponents are expensive and can be difficult toreplace if they are deeply embedded inside thetransmission unit. So correct diagnosis of the cause offaults is important to prevent wasted time and theunnecessary replacement of parts.

5.13.2 Typical diagnostic procedureA logical approach is essential! For diagnosis, analysis ofa system should be broken down into its elemental,functional parts. A typical test procedure would involvesome or all of the following.

● Simple, basic manual and visual checks – Themost basic checks should be done first, for example,the lubricating fluid level and condition. Thetransmission should be at the correct temperaturewhen the fluid level is checked (see manufacturers’data). Then a visual check should be made forobvious problems, such as fluid leaks or damagedelectrical connections.

● A road test and report – A complete understandingof the problem is essential in order to be able todiagnose a fault efficiently. This can be achievedthrough a test drive. Findings should be recordedefficiently as a report during or after the test drive.Malfunctions and correctly functioning componentsshould be noted. Particular attention should begiven to shift quality, both downshifts and upshifts.A stall test should be performed. It is necessary totest the correct operation of different shift programsand converter lock-up clutch operation (by drivingfeel).

● System mechanical checks – Basic checks shouldbe performed on the mechanical and hydrauliccomponents. The whole system can operatecorrectly only if the system pressure is correct. If

[647] Chapter 05 25/8/06 16:35 Page 233


ASC DME

+

Diagnosis – programming

Range

Range displayProgram display

A5S 440ZGS 8.55

GS 8.60.2

M62

Shift lock

Valve body

Prog switchE38-E31*

Kick downSwitch

Input speed

Manual gate

STEPTRONIC

+

SpeedSignals

Wheel

CAN

Output speed

Oil temp

WITHOUTSTEPTRONIC

HALL EFFECTBRAKE PEDAL

SENSOR

CANterminatingresistor

87a

ECMmainrelay

EDS 1

Power

EDS 4

EDS 3

EDS 2

EDS 5

MV 1

MV 2

MV 3

8.55 (CAN 60)8,60.2

DISplusGT-1

Figure 5.68 Electronic system overview

[647] Chapter 05 25/8/06 16:35 Page 234

Transmission summary 235

possible a pressure gauge should be used to takereadings under operating conditions, so that thesecan be compared with manufacturer specifications.If there is a throttle cable, adjustment of this shouldbe checked. Incorrect adjustment can causeproblems with shifting (shift points).

Check resistances of sensors and actuatorsDepending upon the nature of the problem, it may benecessary to check various components for correctresistance readings with a multimeter. Manufacturerinformation is essential for this to compare themeasured value with the correct value. Changes inambient temperature can affect resistance readings inelectromagnetic coils, so a realistic tolerance must beallowed for this. To prevent damage to sensitiveelectronic components and false readings, generalmultimeter etiquette must be followed, withcomponents completely disconnected from the circuitwhen they are checked.

Check signals (waveforms) and voltages atsensors and actuatorsHowever, just because a sensor or actuator has thecorrect resistance doesn’t mean that it is workingproperly when installed. Intermittent faults areparticularly difficult to find as they usually occur onlywhen the vehicle is being operated. Therefore, ifpossible, measure the signal at the sensor or actuatorunder running conditions. In most cases, anoscilloscope will give much more information than amultimeter. The voltage signals at some sensors andmost actuators are quite dynamic in nature and amultimeter is not able to process the signalappropriately. There are many portable, cheaposcilloscopes on the market ideal for this purpose androbust enough for workshop use.

Check system electronics via diagnosticinterfaceNearly all vehicle electronic systems have a diagnosticinterface. In newer vehicles (2001 on), this interface isa standardised on-board diagnostic (OBD) connectorand there are a number of generic scan tools availablethat can access fault codes from the transmission ECU(as well as from the engine ECU). The fault codes have

a standardised protocol, which aids diagnostics quiteconsiderably. Older vehicles tend to have diagnosticinterfaces specific to their manufacturer. Generally,specific equipment protocols exist, which must be usedto access the information. On some vehicles(particularly Japanese ones) there is a blink codeindication, which, although limited in the information itprovides, can be a useful aid.

Diagnostics summaryIrrespective of its type (CVT, DSG, etc.) an electronictransmission is a complex unit and a systematicapproach is needed when dealing with any technicalproblems. An electronic transmission is, though, simplya system (Figure 5.68). It has inputs which areprocessed, giving outputs with the desired reactions orresponses. From first principles, therefore, the inputs tothe system must be confirmed as correct. If this is so,then the outputs can be checked for correct action. Ifactions are incorrect, then the actuators should bechecked or tested. If they are OK, then the connectionbetween the actuators and ECU should be checked, etc.,etc. This approach will solve most problems that mightoccur.

Another point to consider when diagnosing faults isthat a good understanding of the system and its inputsand outputs is especially important, with so manyvariables changing at the same time in such a complexway. Make your life easy! Get as much information aspossible about the system. Try to get a goodunderstanding of the system in overview and how allthe components fit together and work together innormal operation. This information can be obtainedfrom the manufacturer through training documentationor from workshop manuals – as well as from goodtextbooks of course!

A logical approach is essential for diagnostics

Consider systems as black boxes and check allinputs and outputs

Understand the system operation as well as thediagnostic procedures

Check sensor signals and supply signals using anoscilloscope as well as a multimeter

Key

Poin

ts

5.14 TRANSMISSION SUMMARY

5.14.1 Outline of electronic controlIt is clear that the development of electronic control fortransmission systems will continue to move at a swiftpace. Technical developments in mechatronics andmicroelectronics now allow great freedom andflexibility in terms of packaging of control units andinterfacing between vehicle control systems. These

developments are making electronic control of thetransmission, even automatic shifting, more attractivein markets where an automatic gearbox would oncehave been shunned, for example, in high performancecars.

Highly integrated vehicle systems can providesophisticated control algorithms to improve driver

[647] Chapter 05 25/8/06 16:35 Page 235


comfort and safety. Together with adaptive strategiesthat take into account additional operating parametersand boundary conditions, systems can optimise vehicleperformance.

In summary, the main types of transmissioncurrently available are:

● manual constant mesh transmissions● automatic transmissions, with a planetary gearbox

and torque converter● continuously variable transmissions, with a belt

driven variator.

Electronically controlled transmissions, in conjunctionwith advanced powertrain control systems, can provideimproved vehicle performance and efficiency. Thefollowing powertrain systems have all seen the benefitsof the implementation of electronic controls:

● manual transmission – electronic clutch and shiftingfor improved economy and/or vehicle performance

● automatic transmission – sophisticated shiftstrategies and control systems to improve efficiencyand driver interaction

● continuously variable transmission – control ofshifting to improve driver feel, and more controlover hydraulics for greater efficiency

● four-wheel drive – improved traction and extendedfunctionalities through integration with othercontrol systems

● axle/traction control – better tractive forcemanagement on surfaces with limited or no grip.

Current developments show that integrated powertraincontrol will be the basis of future developments. Toimprove performance, as well as to meet legislativerequirements on harmful vehicle emissions, it is nolonger appropriate to treat the powertrain system asseparate, mechanically connected units (the combustionengine, gearbox, axle, etc.). The powertrain system,consisting of these elements, must be considered as asingle unit and calibrated and optimised duringdevelopment for significant improvements to be made.

5.14.2 Future developmentsVehicle manufacturers are under increasing pressure toreduce vehicle emissions. Advanced powertrains willplay a major part in achieving future goals of emissionreduction and improved efficiency to conserve naturalresources. A critical factor to be overcome is the reducedefficiency of the combustion engine at low speeds, andits polluting effect at idle where no power is required atall by the vehicle for motion.

This is an area of great interest, with large potentialfor improvement, and the current trend is towardshybrid powertrains. The hybrid vehicle combines thepositive attributes of an electric drive (high efficiency,quiet, full torque at low/zero speed) with those of a

combustion engine (high specific power to size ratio,long driving range between refuelling). This technologyneeds a sophisticated control system for the primemovers to work together in a harmonious and efficientway. This can easily be achieved with an electroniccontrol system but there are a number of factors thatimpede the acceptance of this technology in the marketplace:

● battery technology – current units are expensive andheavy, and this technical hurdle must be overcometo move the technology forward significantly

● the existing electrical system voltage – the standardvoltage for vehicle systems is 12 volts, which is fartoo low to transmit power efficiently at the levelrequired for tractive force; higher voltage systemswill be developed to cope with increasing powerdemands, which will be essential for the adoption offull hybrid systems

● motor size – although highly efficient, modernelectric motors are still quite large relative to theirpower output; as technology develops, motors ofthe appropriate power rating will become smallerand will be easier to integrate into a combinedvehicle propulsion system.

Manufacturers are currently working hard to overcomethese issues, as well as introducing light hybrid systemsthat can be integrated within current vehicle ranges asan intermediate step. This allows them to introduce thetechnology into the market place and gauge customerreaction to it.

Transmission management is used primarily toimprove economy and vehicle performance, aswell as driver interaction

Hybrid solutions are likely to be the future …

Web links

Transmission systems informationwww.zf.comwww.porsche.comwww.luk.dewww.bosch.co.ukwww.haldex-traction.comwww.siemensvdo.comwww.audi.comwww.vw.comwww.sae.org

Teaching/learning resourcesOnline learning material relating to transmissionsystems:


Key

Poin

ts

[647] Chapter 05 25/8/06 16:35 Page 236

absolute pressure sensors 19AC motors, ISG systems 227acceleration

pedal position sensors 177sensing 12stoichiometric air:fuel ratios 84

accelerometers, knock pressuresensors 25

accumulator converters 135,137–8

actuators 26–33assemblies 197–8common rail diesel systems 175communication between ECUs

27control signals 32–5electrohydraulic assemblies

197–8, 209Haldex coupling 233injector 100, 106non-mechanical 26–7outputs 4–5resistances, checking 235simulation tests 156single-point injection 113–14status monitoring 197transmission systems 196–201wiring, M1.5 system 108

adaptive strategies 154–5adaptive transmission control

214, 215advance mechanisms 40, 47–8,

61–2after treatment systems 128–30,

133–4air bypass ports 94–5, 96air conditioning control 3, 12

status signals 196air:fuel ratios 21–2, 39

and dwell period 45–6and pollutant levels 128,

143–4and power control 119and spark ignition 44control 138–43converter processes 134–5diesel systems 163fuel mixture maps 84

see also lambda values; mixturesair injection 114, 146–7air mass control

diesel systems 163, 177, 185sensors 12, 20, 177, 185

hot wire/film 110–11variables 83, 86

air temperature sensors 5, 12, 88,97–9, 104–5, 114, 155

air valves 100–1, 102airbag systems 2–3, 12airflow, cylinder volumetric

efficiency 119airflow sensors 12, 19–21, 88,

97–8airflow/temperature sensors

97–9, 104–5analogue 23direct injection systems 121live data systems 155M1.5 system 108power supply 101

alternating current (AC)analogues 51

amplifiersECUs 7–8ignition 56–9

analogue sensor signals 9–10, 19,22–5

MAP sensors 23, 111–12rotational speed 14–15speed/position 66temperature 13–14, 23throttle position 23

analogue to digital converters 10,23–4, 192

aneroid capsules, pressure sensors12

angular displacement torsionsensing 195

angular movement/positionsensors 11–12

anti-lock braking systems (ABS)electronics 2–3interface, VW DSG 209sensors and 12

arcing, contact faces 42atmospheric pressure sensors 19

Audi Multitronic transmission223–4

automatic brake differential (ABD)229, 231

automatic gearboxeselectronic control 188transmission management

212–16automatic injection advance units

167, 168automatic shifting hydraulic

transmissions 187automatic transmission 200, 207,

212–16sensors and 12

auxiliary air valves 100–1, 102

ballast resistor ignition modules56

basic fuel program, injectionsystems 83–4

batterieshybrid vehicles 236live data systems 155memory backup 9

belt/chain drive transmissions187

belt driven starter–generators 228blink codes 150–1, 157–9BMW

Drivelogic feature 205SMG system 205

bob weight and spring systems61–2

boost pressure control,turbochargers 185

boost pressure sensors 5, 12Borg Warner 55 gearbox 213Bosch

in-line diesel injection pump164

K & KE Jetronic injectionsystems 77, 79

LE/LE2 systems 97–103M1.5 system 103–9Motronic engine management

systems 148, 149VE distributor pump 165

INDEX

[647] Index.qxp 25/8/06 15:40 Page 237

brakeson/off signals 196pedal position sensors 12

braking downshift, Tiptronicgearbox 216

broadband lambda sensors 142–3burn off function, hotwire air mass

sensors 110burn times, ignition timing 46–7butterflies, and cylinder

volumetric efficiency 119bypass port, mixture adjustment

98–9, 100

cam and rocker system pumpdrives 175

camshaft position sensors 5, 12,16, 23–4, 25, 177, 181

camshaft speed/position sensors23–4

capacitor discharge ignition (CD)60

capsule type pressure sensors 18carbon dioxide emission 125–6,

130carbon monoxide emission 127,

129diesel systems 163

carburettors 77, 78alternatives to 40–2and fuel efficiency 38–9vs single-point injection 112–13

casing mounted ignition modules58–9

catalysts 133, 136catalytic converters 134–8

and mixture formation 117operating temperatures 135–6oxygen monitoring 5, 21–2pre-cat control 22post-cat monitoring 22sensors 135–6

CD systems 60centrifugal cancellation, hydraulic

clutches 218–19centrifugal governors, diesel

pumps 167–9chain drive, Multitronic

transmission 223charge time see dwell periodchemical treatment, pollutants

128–30circuits

ECUs 7–8fault recognition 153monitoring 153power transistors 8switching 8, 32–4

closed crankcase ventilationsystems 130

closed loop systems 57–8operation 138–9

clutches 201actuators 197, 198–9electronic control 201–3Haldex coupling 232piston mechanisms 219pressure control VW DSG MTM

209slave cylinders 206

clutch-by-wire (CBW)transmission systems 201–3

clutch-to-clutch automatictransmission systems 217–19

code readers 150, 157, 159–60coil on plug ignition systems

72–3coils

electronic switching 48energy build up time 68output improvement 57wasted spark systems 71, 72–3

cold runningauxiliary air valve 100–1emission levels 131, 146stoichiometric air:fuel ratios 84

cold startingdiesel 165, 172–4emission levels 131, 146

combustionefficiency 41, 116, 130NOx formation 144pilot injection 184

combustion chambersinjection timing 118diesel systems 164pollutant reduction 129, 130

combustion knock 41, 116, 174diesel knock 164, 174sensors 25, 67

common rail diesel fuel systems175–85

communication, standardisation161

commutators 28complex systems 5components, ECU control of 6–7compression ratios 41, 116compression stroke injection

timing 118–19compressors, air conditioning 7computer controlled systems 4computers see electronic control

units (ECUs)concentric clutch release systems

198–9

238 Index Fundamentals of Motor Vehicle Technology: Book 2

constant energy ignition systems58–9

contact breakerscoil primary circuit 48–50dwell angles 45–6elimination 40ignition systems 38–9wear 42

contact brushes 28contact-based temperature sensors

13contaminants, catalytic converters

137continuous fuel delivery systems

122continuous rotation motors 31continuously variable

transmissions (CVTs) 187,220–6

control signals 32–4altering 34–5M1.5 system 107sequential systems 86

control spools, EDC systems170–2

control unit temperature sensors,VW DSG MTM 209

controlled pressure difference 78controller area network (CAN)

systems 189–90coolant temperature sensors 5,

12, 13, 88, 170–1, 177, 196LE2 systems 99, 101live data systems 155M1.5 system 105, 108single-point injection 114

cooling systems 42cornering detection 215crank position sensors 196crankcase emissions, pollutant

125crankshafts

position sensors 5, 12, 109position triggers 89–90rotational sensors 14, 15–16,

25speed/acceleration monitoring

162speed/position sensors 11–12,

23–4, 65–6, 106speed sensors 177, 181

creep regulation, VW DSG 207current control systems

feedback 57–8contact breaker systems 42

cylindersinjection systems 77, 78, 79,

116–19

[647] Index.qxp 25/8/06 15:40 Page 238

isolation 162position reference points 15–16pressure requirements 44recognition sensors 72–3, 85,

181valves, diesel systems 164volumetric efficiency 119

damping chambers 98dashboard warning lights 157–8,

161DC/DC converters, ISG systems

227deceleration, stoichiometric

air:fuel ratios 84delivery valves, diesel pumps 167diagnostics

capabilities, VW DSG 207connector plugs, standardisation

161EDC systems 172electronic transmission control

216on-board 150–1transmissions 233–5

diaphragm type pressure sensors18

diesel enginesemissions legislation 38injection systems 42, 77,

163–85knock 164, 174

differential control, electronic229–33

differentials, 4WD systems 231digital pulses, Hall effect 54digital signals 9–10, 19, 22, 25–6

MAP sensors 112optical ignition triggers 54–6

digital timing control 63–4digital to analogue converters 10,

192diode control, wasted spark

systems 72direct acting clutch actuators 198direct fuel injection (DI) 115,

116–24diesel 174–5engine management system 149pre-heating systems 172

direct ignition systems 68–73direct shift gearboxes (DSG) 191,

199–201, 207–10disc valves 81, 82distance travelled information 14distributor-based systems 68–9distributor body located sensors

66

distributor caps 52distributor plunger drives 167distributorless ignition systems

68–73distributors, Hall effect pulse

generators 25–6double shift actuators 200–1downshifting

Tiptronic gearbox 216uncontrolled 215

drive-by-wire systems 204Drivelogic feature (BMW) 205driver information 12, 190–1driver input 5, 214driver stage amplifiers, ECUs 7–8driver’s lever position sensors 194drivetrain torque sensors 195driving members 201driving modes, selectable, VW

DSG 207dual clutch systems 187, 203

actuation 198–9dual-bed catalysts 134duty cycles 33–5dwell angles 45–6, 56dwell control, ignition modules

57dwell period 45–6, 68, 70–1dynamic control program (DRP),

Multitronic transmission 224dynamic stability control (DSC)

205dynamic vehicle control systems 3

EGR see exhaust gas recirculationelectric motor type actuators 4,

28–9, 31–2electric vehicles, hybrid 226–9,

236electrical actuation vs servo

operation 196–7electrical control, transmissions

188–9electrical/electronic injection

systems 77, 80, 81–90electrical systems

LE2 system 101–3M1.5 system 107–9

electro-hydraulic shift mechanisms205

electroactuators 198–201electrodes, spark plugs 74, 75–6electrohydraulic actuator

assemblies 197–8electrohydraulic transmissions

187electromechanical actuators,

clutch systems 202–3

Index 239

electronic airflow sensors 20–1electronic and mechanical sensing

devices 12electronic clutch management

201–5electronic control units (ECUs)

4–5, 6–11accumulator converters 138actuator control signals 32–4common rail systems 174–85communication between 27diesel systems 177–85glow plugs 173–4Haldex coupling 232–3injection systems 80, 82–9,

177–85integrated 148–50, 188–91,

202–3lookup tables 63–4, 67M1.5 system 109monitoring, EDC systems 170–2power supply, LE systems 103powertrain control systems

202–3sensor signals 192shift control 205–6system fault recognition 152–5transmissions 188–91, 197–8,

214, 235–6VW DSG 209

electronic diesel control (EDC)systems 170–2

electronic differential control229–33

electronic differential lock (EDL)interface, VW DSG 209

electronic ignition systems 40–8electronic manual gearboxes

204–10electronic petrol injection systems

77–96electronic shift control, manual

gearboxes 204–5electronic switching, coil primary

circuit 48electronic system overview 234electronically assisted ignition

48–50emergency operation 153–4emission control 2, 6–7, 37–9,

125–8, 132–4, 236after treatment 128–30cold running 146CVT systems 221diesel systems 163EDC systems 170engine operation 131–2(EVAP) systems 130

[647] Index.qxp 25/8/06 15:40 Page 239

injection systems 78integrated 148–50legislation 37–8, 160petrol engines 124–47sensors 12, 90system actuators 5

energycontact breaker systems 42production, stratified mixture

formation 117–18regeneration, ISG systems 227waste, distributor based systems

68engine breather systems

maintenance 39engine intake pressure 12, 18,

93–4engine knock pressure sensors 25engine maintenance, and

emissions 38–9engine management systems 6,

11, 12and fuel injection 78load sensing 5, 12interface, VW DSG 209integrated 148–50, 188–91,

202–3engine oil

changing 39temperature sensors 12

engine output, EDC systems 170engine speed

contact breaker systems 42EDC systems 170live data systems 155sensors 5, 12, 105–6, 177, 181,

196timing advance 61–2trigger references 99–100

engine stop/start function, ISGsystems 227

engine torque control 215engine tuning, pollutant reduction

129environmental pollutants 125–7

see also emission controlepicyclic gears, automatic

transmission 212EU emissions legislation 126–7,

160, 163European on-board diagnostics

(EOBD) 160–2evaporative emission (EVAP)

control 134, 147live data systems 155M1.5 system 108

excess air factor, lambda scale124–5

excess fuel return pipes 93exciter coils, rotational speed

sensors 15exhaust gas

composition 125–6oxygen sensors 5, 11–12, 21–2pressure, and EGR 144–6temperature sensors 11–12

exhaust gas recirculation (EGR)systems 130, 133, 143–6,171–2

common rail systems 185control systems 145–6sensors 12valves 5, 145

external drives, double shift drums200

fault codes 150–1accessing 157–60memory, OBD system 216standardisation 160–1

fault handling, transmissions 214fault recognition 152–5feedback current control systems

57–8firing end configuration, spark

plugs 74firing voltage requirements 43–4fixed dwell angle systems 56fixed gear transmissions 187flame speed 130flap type airflow meters 19–20flow type pumping 90–1fluid temperature sensors 11–12flyweights, centrifugal governors

167–9flywheels

as driving members 201sensor locatedion 65–6starter–generators 228

Ford CTX transmission 221four-speed all clutch-to-clutch

systems 217–19four-wheel drive control 229–33

Haldex coupling 232friction clutches 201friction wheel variators 187front-wheel drive vehicles

CVT systems 220gearboxes 217

fuel consumption/economyand CO2 emission 130CVT systems 221diesel systems 164direct injection systems 116–19electronic manual gearbox 205,

207


powertrain automation 204–5stratified mixture formation

118fuel delivery systems 77–8, 90–4

common rail diesel systems175–6

control, precise 80cut-offs, solenoid operated 165,

167diesel 163–4filters 80, 91–2flow 90–1gauges 10pollutant reduction 125, 128,

129quantity control actuators 5sealed 130sensors 12, 18, 19see also fuel injection systems

fuel emissions see emissioncontrol

fuel evaporative emission control(EVAP) systems 130

fuel injection systems 6–7, 30, 42,77–8, 80, 83–4, 97–112

comparison 116–19control 5, 177–85

driver control modules 181duration control 86EDC systems 181live data systems 155timing, direct injection

systems 118–19diesel 42, 163–85direct fuel injection (DI) 115,

116–24, 149diesel 174–5, 177–85GDI systems 115–24pre-heating systems 172

electrical/electronic 77, 80,81–90

engine intake pressure 93–4fuel/ignition control

combination 148–9fuel maps 83–4fuel sprays 81–2indirect, pre-heating 173intermittent 78multi-point 80, 97–112, 116

nozzle injectors 169–70pilot/main 184–5sensors 5, 11, 80, 87–90,

110–12, 121, 177–85simplified 112–13single-point 114–15solenoid actuators 5valves 80vaporisation 116, 120

[647] Index.qxp 25/8/06 15:40 Page 240

volume control, EDC systems170

see also injection pumps;injectors

fuel mixtures see air:fuel ratio;mixtures

fuel pressureregulators 80, 93–4, 183sensors 12, 19systems, diesel 176–7

fuel pumps 90–1continuous rotation motors 31diesel rotary pumps 165–7power supply, LE systems 101wiring, M1.5 system 108

fuel rails 80, 93–4common rail diesel systems

175–6injector solenoid valves 81rail sensors 121

fuelling references 84full rotation motors 31, 32

gas temperature sensors 11–12gasoline direct injection (GDI)

systems 115–24gear actuator valves, VW DSG

MTM 209gear ratios

automatic transmission 214Multitronic transmission 223

gear selectionelectronic transmission control

214Honda system 217–18Porsche PTM system 231VW DSG 208–9

gear shiftactuators 199–201, 221high-performance vehicles 205

gear trainsBorg Warner 55 gearbox 213Honda system 217–18

gearboxesautomatic, electronic control

188electronic manual 204–10information sharing 190–1lightening 205mode selection 205–6RPM sensors 193sensors 194, 196sequential manual (SMG) 205Tiptronic 216–17transmission management

212–16VW DSG 208VW DSG MTM 209

glow plugs 133, 170, 173–4governors, rotary diesel injection

pumps 167–9gradients detection, adaptive

control 215

Haldex coupling 231–3Hall effect 16

digital pulses 54ignition triggers 16–17, 53–4phase sensors 85pulse generators 25–6, 53–4rotational speed sensors 15,

16–17, 192, 193speed/position sensors 65, 67switches 53–4Tiptronic gearbox operation 196

headlight circuits 3–4heat ranges, spark plugs 75heated exhaust gas oxygen

(HEGO) sensors 22heated step type lambda sensors

140–1heater plugs 170, 173–4heating, broadband sensors 143high energy coils, ignition

modules 57–8high pressure pumps 181, 182

diesel 176, 177direct injection systems 122–3

high speed switching, circuits 8high tension cables, ignition 68high voltage rapid opening

injectors 120hill-holder function, VW DSG 207homogeneous charge compression

ignition (HCCI) 189homogenous mixture formation

117Honda clutch-to-clutch automatic

transmission system 217–19hot film sensors 111, 121hot wire sensors 20–1, 110–11hybrid electric vehicle 236

classification 226powertrain technology 226–9transmissions 189

hydraulic clutches 218–19Haldex coupling 232

hydraulic pump and pressureaccumulators 205

hydraulic shift transmissions 187hydraulic system checks 233–4hydraulics, Haldex coupling

231–3hydrocarbon emissions 127

diesel systems 163reduction methods 129–30

Index 241

idle speedand pollutants 131–2control 5, 94–6, 113

EDC systems 172LE2 system 100–1live data systems 155M1.5 system 106–7, 108

crankshaft sensor 106disconnection, torque converters

212motors

partial rotation 32stepper 5, 32

rotary diesel injection pumps169

stoichiometric air:fuel ratios 84idle valves 94–5ignition advance curves 61ignition coils 29

as actuators 5, 27M1.5 system 107, 108monitoring 162output voltage 44

ignition systems 40computer controlled 61–7constant energy 58electronically assisted 48–50fuel control combination 148–9and fuel efficiency 38–9modules/amplifiers 56–9pollutant reduction 128sensors 5, 12

ignition timing 45–8actuators 5advance/retard 57as after treatment 134and cylinder pressure

requirements 44dwell time 42pressure sensors 18and engine intake depression

Hall effect pulse generators54

live data systems 155pollutant reduction 129, 130

ignition triggers 89–90Hall effect 16–17inductive 50–3single-point injection 114

immobiliser systems, EDC systems172

improbable/implausible values154

in-car entertainment systems 3in-line injection pumps, diesel

164in-line transmissions 187inductive sensors

[647] Index.qxp 25/8/06 15:40 Page 241

analogue rotation 192–3crankshaft speed/position 66ignition systems 40phase sensors 85pulse generators 50–3self-generating 52–3

injection pumps 116high pressure 176rotary diesel 167, 169

injection systems see fuel injectionsystems

injectors 80–2actuator

LE2 system 100M1.5 system 106single-point injection 113

common rail systems 175–6,183

direct injection systems 120ECU switching 84LE systems power supply 101motion sensors, EDC systems

171, 172nozzles 81–2operation speed 82

common rail diesel systems175

rotary diesel injection pumps169–70

solenoid valves 81–2timing 85–7

input shaft speed sensors, VWDSG MTM 209

input signals, ECUs 4–5shift control 206

insulators, spark plugs 75intake air temperature

NOx reduction 130sensors 12, 170–1, 177

intake manifold pressure sensors5

map sensors 111and EGR 144–6

intake portsdiesel systems 164injection systems 77–80pollutant reduction 129

intake system, auxiliary air valve100–1

integrated controlemission systems 148–50ignition modules 58–9powertrain systems 189–90,

202–3starter–generator (ISG) systems

226–9torque

converter/starter–alternator

212intelligent powertrain

management 229interfaces

Haldex coupling 232–3VW DSG 209

interference suppression 48intermittent injection 79internal drives, double shift drums

200

Jaguar 6 speed automatic gearbox213

knock see combustion knock

lambda sensors 5, 138–43live data systems 155M1.5 system 108single-point injection 114step type 141–2see also oxygen sensors

lambda valuesdiesel systems 163excess air factor 124–5fuel mixture maps 84lambda 1 air:fuel ratio 21lambda window 124–5and pollutant levels 128see also air:fuel ratios

LE2 systems (Bosch) 97–103lead emissions, origin/effects 127leaded fuel, and catalytic

converters 137lean burn technology

broadband sensors 142emissions control 132–3pollutant reduction 129, 130

LED optical ignition triggers 55LED probes, blink codes 158–9light hybrid powertrain

technology 226–9lighting systems 3–4, 9, 26–7limited slip differentials (LSDs)

229limp home mode 153–4, 215, 207linear movement sensors 11–12linear solenoids 96linear speed information sharing

190–1link plate chain drives 223live data systems 155–7live rear axle systems 230load

and pollutants 131sensing 5, 12stoichiometric air:fuel ratios 84timing advance/retard 47


vacuum retard 62locations, ignition modules 58lock-up clutches 211–12, 215lookup tables, ECU 63–4, 67Los Angeles smog 38low pressure pumping systems

121lubricant viscosity, and system

pressure 194

M1.5 system 103–9magnetic fields

actuators and 27–30inductive pulse generators 50–1

magnetic sensorsignition systems 40rotational speed sensors 14–15

magnetoelastic torque sensors195

main injection 184–5malfunction indicator lamps (MIL)

161manifold absolute pressure (MAP)

sensors 5, 19, 88, 170analogue 23, 111–12diesel fuel injection systems

177live data systems 155

manifoldsheating 173

single-point injection systems115

pressure 93–4direct injection systems 121signals 196

manual checks, transmissiondiagnostics 233

manual idle speed adjustment100

manual transmissions 187, 207clutches 201–2gear shift actuators 200gearbox electronic control

204–10selection 200

MAP sensors see manifold absolutepressure sensors

mass airflow measurement 88direct injection systems 121sensors 20–1signals 196

master cylinder assemblies 198master references

single coil systems 72–3wasted spark systems 71

mechanical advance mechanisms61–2

mechanical concentric clutch

[647] Index.qxp 25/8/06 15:40 Page 242

release systems 198–9mechanical friction clutches 201mechanical ignition systems 42–3mechanical injection systems 77mechanical system checks 233–4mechanical timing systems 61–3mechanical type sensors 12

airflow sensors 19–20pressure 18

mechatronic control units,integrated, VW DSG 207

mechatronic transmissionmanagement units 197–8, 209

memory, ECUs 8–9, 83, 216Mercedes-Benz, fuel injection

systems 77, 78micro/mild hybrid systems 227misfires

and catalytic converters 136–7monitoring 162

mixturesadjustment, LE2 system 98–9control

pollutant reduction 129, 130power/torque regulation 119

enrichment, cold running 146formation, direct injection

systems 117–18lean burn technology 132rich, and cold-start 172see also air:fuel ratios

mode selection, gearboxes 205–6modulating valves, VW DSG MTM

209monoliths, catalytic converters

136Motronic engine management

system (Bosch) 103–9, 148movement sensors 11–12multi-plate clutches 203, 223multi-point fuel injection (MPI)

97–112, 116components 80hotwire air mass sensors

110–11MAP sensors 111–12nozzle injectors 169–70

multi-stage transmissions 187multimeters, duty cycle readings

34multiple ignition coil systems

68–9multiplexing

control valves, VW DSG MTM209

transmissions 189–91Multitronic transmission (Audi)

223–4

narrow band sensors 139–42needle valves 81, 82negative sparks 70negative temperature coefficient

(NTC) sensors 13–14, 196Nissan CVT system 224–5nitrogen, in exhaust gas 125nitrogen oxides (NOx) 127

accumulator converters 135,137–8

emissions, diesel systems 163formation 143–4reduction methods 130

noise, direct injection systems174

non-mechanical actuators 26–7

OBD system, fault memory 216off-road vehicles, differential

control 229–33oil pressure control valves, VW

DSG MTM 209on/off actuators 197on-board diagnostics (OBD) 22,

150–1open circuits, fault recognition

153open loop operation 138–9opening time, injectors 82operating conditions 62–3, 131optical ignition triggers 54–6optical speed/position sensors 66oscillation frequency, step type

lambda sensors 142oscilloscope readings 34, 86, 87outlet port valves, common rail

diesel systems 175output shaft speed sensors, VW

DSG MTM 209output signals

amplifiers, ECUs 7–8broadband sensors 143D/A converters 10ECU systems 4–5sensors 192

overdrive systems 188, 189oxidation 133

catalysts 134oxides of nitrogen see nitrogen

oxidesoxygen reduction, exhausts 130oxygen sensors 11–12, 21–2,

138–43analogue signals 25single-point injection 114see also lambda sensors; lambda

values

Index 243

parallel shift gearboxes 210partial rotation motors 31, 95–6passwords, code readers 159phase sensors 85photochemical smog formation

38phototransistors 55pilot injection 184

phase control 174, 176Pintaux injectors 173pintle nozzle injectors 169–70,

173planetary gears 187, 212playback mode, data systems 156plug gaps 43–4, 76points ignition systems, and fuel

efficiency 38–9polarity, spark plug electrodes 76pollutants see emission controlPorsche

Tiptronic gearbox operation191, 196, 216–17

traction management (PTM)230–1

port type injection systems 80,117, 119–20

position reference points 15–16position sensors 11–12, 17–18,

23–4, 195positive crankcase ventilation

(PCV) systems 130positive displacement pumping

90–1positive temperature coefficient

(PTC) sensors 13–14potentiometers

airflow sensors 19, 97–8, 104–5position sensing 25throttle position 17–18, 88–9

power amplifiers, ECUs 7–8power supply

boost function, ISG systems227

LE systems 101loss, CVT systems 221–2M1.5 system 108regulation, mixture/throttle

control 119power transistors 8, 48–50powertrains

control units 190–1light hybrid 226–9transmission types 187

pre-heating systems, diesel 172–4pre-ignition detection 41pre-programmed driving modes,

selectable, VW DSG 207pressure accumulators 205

[647] Index.qxp 25/8/06 15:40 Page 243

pressure controlcarburettors 78gearboxes 215fuel 90, 93–4VW DSG MTM 209

pressure sensors 12, 18–19, 67analogue 25direct injection systems 121information sharing 190–1pressure/vacuum sensors 11–12transmission system 194–5

pressure systems, diesel 176–7pressure valves, diesel pumps 167programming, ECUs 6pulley assemblies, CVT systems

221–2pulse air systems 146–7pulse generators

Hall effect 16–17, 53–4ignition systems 40inductive 50–3signals 9–10, 33

pulse width modulated (PWM)signals 33, 193

pump based air injection systems146

pump-turbine unit, torqueconverters 210–11

pumpscommon rail diesel systems

175–6direct injection systems 121–3elements 90–1high pressure, diesel 176, 177,

181, 182single high pressure 178–80

purge valve wiring, M1.5 system108

racing enginescapacitor discharge ignition 60fuel injection systems 77, 78

radio interference, spark plugs 74reference points

crankshaft sensors 65–6ECU monitoring 9–10ignition timing 51–2rotational angular position

sensors 15reference signal systems 65–7

sequential injection systems85–6

reference voltagesLE systems 101ECU systems 13

references, mastersingle coil systems 72–3wasted spark systems 71

regenerative retarders, ISGsystems 228

relay operationheadlight circuits 3LE systems 102M1.5 system 108

reliability/durability issues 39–40reluctors

crankshaft sensors 65–6inductive pulse generators 50–1rotational speed sensors 14–15

remote ignition modules 58–9resistance

requirements 44throttle position sensors 17

resistorsballast 56spark plugs 74temperature sensors 13variable 9

retarder function, ISG systems227

retrofit gear shift actuators 199returnless systems 93reverse torque converters 212road speed information 14road tests, transmission

diagnostics 233road wheel rotational speed

sensors 14roller cell pumps 92roller ring assemblies 167rotary diesel injection pumps

165–72rotary idle valves 32, 95–6

M1.5 system 106–7rotary injection pumps, diesel

165–72rotational sensors 11–12

angular position 15–18rotational movement 11–12speed sensors 14–18, 190–1,

192–3rotor arms 52

elimination 68–9energy waste 68

rotor discs, Hall effect sensors16–17

safety systemselectronic transmission control

215–16integration 190, 230VW DSG MTM 209

satellite navigation systems 3scan tools 150, 157, 159–60seat sealing, spark plugs 74secondary air injection 146–7


selectorsassembly, electronic manual

gearbox 207control unit interface, VW DSG

209sensors, VW DSG MTM 209

self-bleeding fuel systems 165self-diagnosis systems 150–1, 154self-generating inductive sensors

52–3self-learning capability 205semiconductor pressure sensing

elements 195sensors 4, 11–13

applications 11–12ECU input 4–5, 6failure substitute values 215–6inductive, self-generating 52–3information sharing 190MAP 111–12M1.5 system 103–4resistance checking 235signals 22–6

analogue to digital converters10

types 13–22VW DSG MTM 209

sequential injection 85sequential manual gearbox (SMG)

205serial data systems 155–7series resistance circuits 13service adjust mode, data systems

156servos 196–7shaft layout, VW DSG 208shaft torque sensing 195sheathed-element glow plugs 174shift control

automatic transmission 214–15electro-hydraulic 205mode selection switches 194quality 214sensors 206uncontrolled 215

shift curve adaptation, Tiptronicsystem 217

short circuits, fault recognition153

shut-off valves, solenoid 167sideways movement sensing 12signals 5, 9–10, 22–5, 192

analogue to digital converters10

broadband sensors 143checking 154, 197, 235ECUs communication 27stepped 23, 141–2

[647] Index.qxp 25/8/06 15:40 Page 244

see also sensors; specificapplications

silver spark plug electrodes 75–6simple multi-point injection with

airflow sensor 97–103simplified injection systems

112–13simulation tests, actuator 156simultaneous injection 85

LE2 system 100M1.5 system 106

single barrel pumps 122–3single-bed converters 134single coil systems 68, 72–3single high pressure pump system

components 178–80single hole nozzle injectors

169–70single-point petrol injection

112–15sliding mesh gearboxes 204slip control, lock-up clutches

211–12slip differentials, limited (LSDs)

229smart sensor technology 192smog 37–8snapshot mode, data systems 156solenoid actuators 4, 5, 27–8,

30–1, 196–7solenoid injectors, single-point

injection 113solenoid operated fuel cut-offs

165solenoid valves

air 30–1idle valves 95–6injector 81–2pulse air 146–7

solid state components, pressuresensors 18–19

soot emissions, diesel systems163

sparkadvance maps 63–4detection 162duration 44–5plugs 39, 73–6positive and negative 70quality 43timing 46–8

speed, injector operation 82speed/position sensors 192–3

analogue 23–5crankshaft 65–6single-point injection 114

speed-related timing advance46–7

spin control, sensors and 12

sports map shift, Tiptronic gearbox216

sports utility vehicles (SUVs)229–33

spring pressure, clutch operation201

stability control, sensors and 12stalling, at idle speed 94standardisation, on-board

diagnostics 160–1starter integration

starter–alternator torqueconverters 212

starter–generator systems (ISG)226–9

startingdirect injection systems 124electronic manual gearbox 207rotary diesel injection pumps

169voltages see reference voltages

status monitoring, actuators 197steering angle sensing 12steering column control unit

interface, VW DSG 209step type sensors 139–42stepped signals 9–10, 23stepped transmission, tractive

effort curves 220stepper motors 29, 32

as actuators 5idle valves 94–5

stoichiometric air:fuel ratios 21,84, 117–18, 124–5

stratified mixture formation117–18

lean burn technology 132–3pollutant reduction 129

substitute values, sensors 154,215–6

substrates, catalytic converters136

supercharger control sensors 12switch-based temperature sensors

13switches/switching

and relays 3and sensors 4circuits, high speed 8coil primary circuit 48digitally controlled 9ECU functioning as 32–4Hall effect 53–4power transistors 8, 48–50throttle 88–9wasted spark systems 71

synchromesh gearboxes 204six speed, VW DSG 207

synchromesh transmissions 187

Index 245

system components, M1.5 system103, 105–9

system fault handling, VW DSG207

system fault recognition 152–5system faults, self-diagnosis

150–1system layout, M1.5 system 103,

104system LED, blink codes 158system pressure, automatic

transmission 214

temperature control, ECUs 6temperature sensors 11–12, 13–14

analogue signals 23coolant 99EDC systems 170–1fault recognition in 152–3information sharing 190–1

temperaturesand catalytic converters 135–7extremes, actuator operation

196spark plugs and 73

terminology, code readingequipment 157

testsactuator simulation 156emission regulations 126–7fault codes access 157–60routines 156–7

thermal afterburning 133thermistors 13thick film pressure sensing

elements 195three barrel pumps 122three-phase AC motors, ISG

systems 227three-way catalytic converters

134–5thresholds, digital signals 9–10throttle body fuel injection (TBI)

112–15, 116throttle butterflies 17throttle control

direct injection systems 120–1power/torque regulation 119

throttle linkages, stepper motoraction 94–5

throttle position sensors 5, 12, 17,88–9

information sharing 190–1live data systems 155M1.5 system 105, 109pedal sensors 205signals 23, 196single-point injection 114variable resistors 9

[647] Index.qxp 25/8/06 15:40 Page 245

throttle positioner systems 129throttle switches 88–9

LE systems 99, 102timing advance

advance/retard 47characteristics 63–4EDC systems 172rotary diesel injection pumps

167timing control 61–4

ECU lookup tables 63–4, 67ignition systems 40injection systems 79

timing trigger systems see triggersTiptronic gearbox operation 191,

196, 216–17top dead centre (TDC) 15, 16, 25toroidal variator drive (Torotrack)

transmissions 187torque 186

and emissions, diesel systems163

boosters, ISG systems 228control

diesel rotary pumps 165mixture/throttle control 119

converters 187, 212electronic control 210–12lock-up control 196, 211,

215, 224turbine speed information

193distribution, Porsche PTM

system 231sensors

drivetrain 195Multitronic transmission 223

transfer, CVT systems 223torque/inertia ratio, electric

motors 228torsion angular displacement

sensing 195traction control

adaptive control 215ASR/ABD 229(ESP) system interface, VW DSG

209efficiency, 4WD systems 231tractive effort curves 220

transient operation, clutches 206transistors, switching 48–50transmission systems 186–236

diagnostics 233–5electronic switching 205integration 188–91, 205sensors 192–6, 195–6splitting principle 209–10

temperature monitoring 193–4transverse drive vehicles,

gearboxes 217triggers

circuits 50–3discs 16–17, 65–6Hall effect 16–17, 53–4injection systems 89–90optical 54–6timing 65–7

LE2 system 99–100M1.5 system 105–6

turbine/pump unit, torqueconverters 210–11

turbochargersboost pressure control 5, 185sensors 12

twin clutch systems 187, 203actuation 198–9

two-speed governors, rotary dieselinjection pumps 167–9

two-stage current control process86–7

unit injectors, common rail dieselsystems 175–6

USA, emissions legislation 37, 38,160

vacuum advance/retardmechanisms 62

vacuum sensors 11–12valve timing

electronic 42NOx reduction 130, 134variable control 149

valves 39actuator, VW DSG MTM 5, 209rotary idle 32

vanesairflow sensors 19–20, 97–8Hall effect pulse generators 54

variable position bleed actuators197–8

variable reluctance sensors14–15, 51

variable resistors 9variable-ratio pulley assemblies

221–2variables, injection ECUs 83Variomatic CVT system 220, 221VE distributor pumps 165–72vehicle application software 159vehicle speed monitoring 196

information sharing 190–1speed sensors 12

vehicle stability control sensors 12


venturi 78visual checks, transmission

diagnostics 233Volkswagen

direct shift gearbox (DSG) 199,207–10

electronic manual gearbox205–7

voltagesanalogue pressure sensors 19checking, sensors/actuators

235ECUs 7–8feedback control 57–8ignition systems 29–30, 43–4oscillation, rotational speed

sensors 14–15rapid opening injectors 120threshold points, temperature

sensors 23throttle position sensors 17

volumetric efficiency, cylinders119

Volvo Variomatic CVT system 220

warm-up maps, Tiptronic gearbox216

washcoats, aluminium oxide 136wasted spark ignition systems 68,

69–72water in exhaust gas 125waveform checking 235weak mixtures, and CO2 emission

130wet clutches

twin clutch arrangement 203VW DSG 209–10

wheel slip detection 215wheel speed sensors 12, 24–5

information sharing 190–1wheel spin control 12windings, secondary 29windscreen wipers 4wipers, potentiometer 17, 97–8,

104–5wiring circuits

Hall effect pulse generators 54inductive pulse generators 52–3LE2 system 101–3M1.5 system 107–9optical ignition triggers 55–6

worm gear drive actuators 198Wright Brothers 77

zero voltage points 51–2zirconium oxide oxygen sensors

21

[647] Index.qxp 25/8/06 15:40 Page 246

[647] Index.qxp 25/8/06 15:40 Page 247

[647] Index.qxp 25/8/06 15:40 Page 248

Documents

Hillier's Fundamentals of Motor Vehicle Technology: Powertrain Electronics (Book 2), 5th Edition