Commissioning of a Test Stand for Turbocharger ...isromac-isimet.univ-lille1.fr/upload_dir/finalpaper17/50.finalpaper.pdf · Commissioning of a Test Stand for Turbocharger Investigations

Commissioning of a Test Stand for TurbochargerInvestigations at Constant Turbine Inlet TemperaturesAlexander Esper1*, Christian Lenzen1, Manfred Wirsum1

SYM

POSI

A

ON ROTATING MACHIN

ERY

ISROMAC 2017

InternationalSymposium on

Transport Phenomenaand

Dynamics of RotatingMachinery

Maui, Hawaii

December 16-21, 2017

AbstractIn this paper, the developed turbocharger test stand design at the IKDG facility is evaluated. �e spe-ci�c characteristics of the test stand as well as operational test data is discussed. A test turbochargeris thermodynamically examined in regard of operation at 220◦C turbine inlet temperature. Anoverview of the measurement instrumentation is given and the thermodynamic parameters of theturbocharger necessary for generating a power map, are presented. �e key aspect of providing aconstant turbine inlet temperature nessesary for the acquisition of reliable perfomance measure-ments is vary�ed by means of an operation point stability analysis at di�erent rotational speedsacross the whole operating range of the test turbocharger. Additionally, an uncertainty analysisof the temperature and pressure measurement instrumentation according to DIN V ENV 13005and [1] is conducted. Based on these results, a performance map of the compressor as well as theturbine of the turbocharger is generated and discussed. �e paper is concluded by providing anoutlook regarding the possibilities of upgrading the performance measurement instrumentation inthe exisitng test stand design as well as upgrading certain parts of the test stand.KeywordsTurbocharger — Test stand — Performance measurement1Institute for Power Plant Technology, Steam and Gas Turbines, RWTH Aachen University, Germany*Corresponding author: [email protected]

INTRODUCTION

Design parameters of state of the art turbochargers evolvetowards slender rotating components due to cost and opera-tion mode e�ciencies. Slender rotating components increasetransient response characteristics due to lower inertia andtherefore, improving performance and e�ciency. To evaluatethe operational characteristics of a turbocharger at di�erentturbine inlet temperatures, a test stand was developed at theInstitute for Power Plant Technology, Steam and Gas Tur-bines (IKDG) to enable investigations of performance maps atselectable constant turbine inlet temperatures. To reproducereal time applications where turbochargers are implementeddownstream the exhaust outlet of an engine, a lot of teststands use loop operation where the compressed air of theturbocharger compressor passes a combustion chamber up-stream of the turbocharger turbine [2, 3]. By decoupling themass �ows of turbocharger compressor and turbine, a widerrange of the performance map can be investigated. Further-more, this simpli�es operating the turbine at constant inlettemperatures. �erefore turbocharger test stands have beendeveloped on which the turbocharger turbine is operatedwithout using the compressed air mass �ow from the tur-bocharger compressor [1, 4]. A signi�cant di�erence betweenthese test stands and the test stand introduced in this paperis the use of two hot gas generators that run sequentially en-abling the adjustment of the turbine inlet temperature with amaximal deviation of about +/- 1.0K during steady state op-eration. �is allows an operation of the turbocharger close toa thermodynamic equillibrium operation to maximize perfor-

mance measurement accuracy compared to conventional teststands using combution chambers as hot gas generators. Fur-ther advantages in the use of an adjustable electrical heateris the simpler control of the heater, the uniform warm-upprocess as well as the avoidance of thermo shock stress onthe measurement instrumentation due to combustion cham-ber ignition. On the other hand, the implementation of anelectrical heater leads to slower warm-up ramps and negatesa real time operation with exhaust gas. A unique featureof the IKDG test stand is the possible mass �ow range forturbocharger testing since the turbine of the turbocharger issupplied by a six-stage radial compressor with a maximumair mass �ow of 12 kg/s. While test stands like [5], [6], and [7]also use sequential hot gas generators, the mass �ow range islimited compared to the IKDG test stand setup. �e commis-sioning of the developed test stand and its thermodynamicboundary conditions as well as the chosen operation route,lay the basis for future experimental investigations regardingtransient pressure and blade vibration phenomena.

NOMENCLATUREc Flow velocity (absolute)cp Speci�c heat capacityMa Mach numbern Mechanical exponentp PressurePr Prandtl numberr RadiusR Set radius of pipe

Commissioning of a Test Stand for Turbocharger Investigations at Constant Turbine Inlet Temperatures — 2/8

Re Reynolds numberrec Recovery factors SecondS SampleT Temperatureu UncertaintyV Voltage

SymbolsI Output signalm Mass �owκ Isentropic coe�cientσ Input signal

SubscriptsCbl CableDIN Deviation compared to DINHC Heat conductionInh Inherent resistanceLin Treatment and linearizationLts Long term stabilitym Meridional / MeasuredMa In�uence of Mach numberMt Transient e�ectsMv Viscosity impactMw Wall in�uenceRef Reference temperature of MASs StaticShs Slef heating of sensorStab Stabilization periodt TotalTam Ambient temperature of MAS�str �ermo-voltageVs Voltage supply of MAS

AbbreviationsCFD Computational �uid dynamicsDPT Di�erential pressure transducerFSO Full scale outputMAS Measuring acquisition systemMP Measuring planeTC �ermocouple

1. EXPERIMENTAL SETUP

1.1 Test stand setup�e test turbocharger is designed for usage in marine engineapplications and is provided by Kompressorenbau BannewitzGmbH (KBB). �e turbocharger provides a total compressorpressure ratio of up to 5.5 with a maximum turbine mass�ow rate of 5.7 kg/s at a certain inlet temperature of 220◦C.�is allows the determination of the turbocharger behaviourand e�ciency at below usual real time application. �e testturbocharger is designed for turbine inlet temperature op-eration up to 650◦C during real time application. �e teststand consists of di�erent subsystems composed of air sup-

ply infrastructure, air heater components, test turbocharger,data acquisition system and control unit. Figure 1 shows asimpli�ed �ow diagram of the turbocharger test stand at theIKDG facility with the instrumented measurement planes(MP). �e design parameters are presented in Table 1.

Table 1. Design parameters of the turbocharger test standat IKDG

Turbine inlet mass �ow ≤ 6kg/sTurbine inlet pressure ≤ 6barTurbine inlet temperature currently 220◦CTurbine inlet temperature possible 500◦CTurbine wheel diameter ∼ 250mm

1.1.1 Compressor facility�e test rig is powered by a high pressure air supply systemdriving the turbocharger turbine at the IKDG test facilitywith compressed air. �e �ow medium is provided by a 6-stage radial compressor (2) which provides a maximum airmass �ow of 12kg/s at 30bara. �e compressor is controlledby variable inlet guide vanes (1) at the �rst three stages. Asecurity bypass valve (3) at the outlet of the compressor unitis used during the start-up procedure, for emergency shut-down as well as to adjust the required test stand mass �ow.To further increase the adjustment precision a control valve(4) is installed upstream the air preheater.

1.1.2 Air preheater�e air exits the 6th stage of the compressor with a temper-ature of approximately 150◦C. �en, the air temperature isfurther increased by a 10.5MWnatural gas �red air preheater(5)–(7) to a temperature of about 180◦C. For investigationsrequeiring higher turbine inlet temperatures the air temper-ature can be increased up to 500◦C.

1.1.3 Turbine sectionBefore the �ow medium enters the turbocharger, a �lter (8)cleans out any particles which may be carried along the �owpath. �e mass �ow entering the turbocharger turbine ismeasured by means of an ori�ce plate (9) that is designedaccording to DIN(5167-2, 2004) . A control valve (10) is in-tegrated into the turbine piping as a safety measure in casethat the lubrication system, providing the turbocharger withoil, fails and therefore prevent damage of any rotating com-ponents. Since the NG �red heater provides temperaturegradients too wide for steady turbocharger operation, anelectrical preheater (11) is directly integrated into the teststand design thus enabling steady state parameters with amaximum turbine inlet temperature deviation of about +/-1.0K during steady state operation. �e �uid enters the tur-bocharger turbine (12) and exits into the stack. �e positionof the ori�ce plate upstream of the electrical preheater waschosen in order to minimize mass �ow measurement inac-curacies due to hot gas streaks. �e true �ow temperatureentering the turbine is measured at measurement plane 1.


Figure 1. Overview of the test stand at IKDG with instrumented measurement planes

Due to the below real time application no limits are placedon the turbine outlet temperature during test operation.

1.1.4 Compressor section

At the IKDG test stand, the compressor and the turbine side ofthe turbocharger are operated independently of one another.On the induction side of the compressor the intake mass �owis measured using an ori�ce plate (13). To acquire all datarequired to develop a performance map for the turbocharger,a thro�le valve (15) is installed at the pressure side of theturbocharger compressor (14) allowing the adjustment of thecompressor operation point from stall to surge (dischargevolume: 0.008m3). �e compressor is supplied with air atambient condition and discharges the compressed air into astack (16).

1.2 Measurement setup�e measurement instrumentation comprises the acquisi-tion of the thermodynamic operating conditions of the tur-bocharger as well as the rotational speed and the lubricatingoil inlet and outlet temperatures. �e static pressures arerecorded by means of three holes equally distributed over thecircumference of the in- and outlet planes of the turbocharger(MP1-MP4). �e total inlet and outlet pressure for the tur-bocharger are calculated from the measured static pressureand the Mach number. Kummel [8] states, the pressure cal-culation relies on the average �uid velocity and the average

Mach number (see Equation 1)

ptps=

(1 +

κ − 12

Ma2) κ

κ−1(1)

�e temperature at the turbine side is measured by threetype K thermocouples for each measurement plane. �e mea-surement planes are displayed in 1 labeled MP where the�ow temperature is measured. �e equipped thermocou-ples feature a sensor diameter of 3 mm and are distributedequally over the circumference of the measuring plane. Onthe compressor side the temperature at the in- and outlet aremeasured by means of three Pt100 resistance thermometerswith the same circumferential distribution. Since the temper-ature is not measured by means of total temperature probes,the measured temperature does neither represent the staticnor the total temperature of the �uid. �e static temperatureis calculated using Equation 2.

T s = TProbe − recc(r)22cp

(2)

�e recovery factor was approximated for turbulent �owconditions in reference to [8]. Since the temperature depen-dency of the Prandtl number is negligible in the examinedtemperature interval from ambient temperature to 220◦C,the static temperature is calculated using a constant recoveryfactor.

rec = 3√Pr = 3√0.7 = 0.8879 (3)


�e total temperature is calculated using the velocity atthe radial position of the thermocouple sensing head withinthe �uid �ow path in contrast to the total pressure of the �uidthat is derived by means of the average �uid velocity. For allmeasurement planes the temperature was measured at theequivalent same radial position for all three probes. �ereforethe insertion depth’s in�uence on measurement uncertaintyis not considered throughout the following investigations.�e radial velocity pro�le is a�ained using Equation 4.

c(r)cm=(1 + n)(2 + n)

2

(1 − r

R

)n(4)

Kummel [8] suggests to set n=1/7 as in turbulent �owpro�les this value is proven to work. Performed comparativemeasurements with a 6 point pressure measuring rake provedthe suggested value [9].

T t

T s= 1 +

κ − 12

Ma(r)2 (5)

2. RESULTS AND DISCUSSIONFor the acquisition of turbocharger performance map datait is essential to validate the operating point stability to var-ify the assumption of steady state operation. Furthermorethe operating point stability and the obtained thermody-namic equilibrium state of the turbocharger during measure-ment procedure ensures comparability of di�erent conductedmeasurements. �erefore, the tolerances of the steady statede�nition are analyzed. During data acquisition the follow-ing criteria for steady state have been formulated to ensurerepresentative thermodynamic data parameters. �e dataacquisition procedure entails a sample rate of 1S/s for themeasurement of temperature, pressure, mass �ow and rota-tional speed. A total number of 60 sample points is measuredand arithmetically averaged to a�ain the thermodynamicparameters of the current operation point. �e analysis ofthe operation point stability compares each sample point tothe arithmetically averaged operation point thus displayingthe �uctuation over the data acquisition interval of 60 samplepoints equivalent with 60 second measurements.

2.1 Operation point stabilityDuring steady state data acquisition, it is essential that nei-ther increasing nor decreasing tendencies in turbine outlet aswell as compressor outlet temperature occurs. A maximumdeviation of about +/- 1.0K of the turbine inlet temperatureduring the acquisition time interval at steady state opera-tion was aimed for. �is allows a more steadier operationof the turbocharger compared to test stands using combus-tion chambers as hot gas generators. A stability in rotationalspeed of 2Hz as well as the stability of the oil inlet temper-ature of +/- 0.5K during acquisition procedure were alsode�ned as required steady state criteria since �uctuationsin the oil inlet temperature directly a�ect the stability in

rotational speed. �e operation point stability was analyzedfor three representative rotational speeds of around 25%, 60%and 89% of the turbocharger’s maximum speed. Figure 2 andFigure 3 show the stability of the turbine inlet parameters(temperature and pressure) during the acquisition intervalof 60 sample points. Maximum peak to peak variation inturbine inlet temperature shows values of about 0.3K, whichtranslates into a realtive variation of 0.061%. �e maximumturbine inlet pressure peak to peak �uctuation is 4 mbar(0.098%).

-0.2

-0.15

-0.1

-0.05

0

0.05

0.1

0.15

0 10 20 30 40 50 60

Te

mp

era

ture

flu

ctu

atio

ntu

rbin

ein

let

[K]

25.4% rpm_max 60.1% rpm_max 89.3% rpm_max

Measuring point [-]

Figure 2. Temperature �uctuation turbine inlet

Measuring point [-]

-2.5

-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

2.5

0 10 20 30 40 50 60

Pre

ssu

re flu

ctu

atio

n tu

rbin

e in

let

[mb

ar]


Measuring point [-]

Figure 3. Pressure �uctuation turbine inlet

When adjusting the operation point using the thro�levalve at the compressor outlet side, the temperature mea-sured at the compressor outlet indicates, whether thermody-namic equilibrium of the operation point is achieved. �e�uctuation in temperature a�er adjusting the thro�le valveand reaching steady state operation is shown in Figure 4.A�er the turbine and compressor �uid temperatures havereached thermodynamic equillibrium, the acquisition is onlystarted when the oil inlet temperature �uctuation is withinthe de�ned limits as shown for three exemplary rotationalspeeds in Figure 5.

2.2 Uncertainty analysisIn order to evaluate uncertainties of the acquired data, the fullmeasuring chain has to be analyzed, since a certain amountof disturbing in�uences may cause deviations and errorsthroughout the measuring interval. In accordance with DINV ENV 13005 the installed measurement acquisition system


Measuring point [-]

-0.2

-0.1

0

0.1

0.2

0 10 20 30 40 50 60

Tem

pera

ture

flu

ctu

ation

com

pre

ssor

outlet

[K]


Figure 4. Temperature �uctuation compressor outlet

Measuring point [-]-0.4

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

0 10 20 30 40 50 60

Tem

pera

ture

fluc

tuat

ion

oil i

nlet

[K]


Measuring point [-]

Figure 5. Temperature �uctuation oil inlet

(MAS) and all implemented sensors are assessed. �e tem-perature on the turbine side is measured with thermocouples(TC) of Type K, which are connected to modules with inte-grated “cold junction compensation”. All TC are connected tothe same module and therefore have got the same referencetemperature. �e temperature at the compressor side is mea-sured with four-wire system Pt100 resistance thermometers.�e measurement chain is modeled as suggested in [1].

2.2.1 Uncertainty of temperature measurement�e temperature measurement chain is modeled by meansof Equation 6 for the Type K thermocouples implemented onthe turbine side. �e temperature measurement chain of thecompressor side is modeled by means of Equation 7 basedon the use of Pt100 resistance thermometers.

lT = Tm + σHC + σDIN + σStab + σRef + σCbl + σVs

+σTam + σLin + σInh + σLts(6)

lT = Tm + σHC + σDIN + σStab + σShs + σ�str + σVs

+σTam + σLin + σInh + σLts(7)

�e acquired output signal lT depends on the measuredtemperature Tm and a number of uncertainties which areexplained and quanti�ed in Figure 6 and Figure 7.

�e thermocouples Type K are located directly in the �uid�ow path of the turbine by means of compression ��ings.

Input

parameter

Source of error Distribution Uncertainty

σHC Heat conduction Normal Unattended because TC is

insulated to stem

σDIN Deviation to DIN Rectangular 1.0K/ 3=0.577K

σStab Stabilization

period

Cancelled because turbine inlet temperature is

kept constant and all operating points are

measured in steady state

σRef Reference

temperature

Considered in σDIN

σCbl Cables Considered in σDIN

σVs Voltage supply Rectangular Measured temperatures

0°C: 0.49K

100°C: 0.46K

300°C: 0.58K

0.58K/ 3=0.335K

σTam Ambient

temperature of

MAS

Rectangular

σLin Treatment and

linearization

Rectangular

σInh Inherent

resistance

Rectangular

σLts Long term

stability of

electronic

components

Rectangular Cancelled

Combined standard uncertainty with 68.3% confidence interval

𝒖𝑻𝑺 = 𝟎. 𝟗𝟑𝟐𝑲

Figure 6. In�uencing variables for temperature measure-ment TC Type K

�e sensing element is insulated inside the compression �t-ting thus heat conduction is not considered. Deviations fromthe characteristic curves are delivered by the manufacturerof the TC. For Type K thermocouples the limiting deviation isgiven as 1.0K for temperature intervals from 0◦C to 1100◦C.During data acquisition the test stand runs in steady statekeeping the in- and outlet temperatures of the turbochargerconstant. Because the TC reacts much faster to transientsthan the test stand itself, stabilization periods of the TC donot have to be considered. Errors caused by reference tem-perature, extension cables and linearization are accountedfor by σDIN. Errors due to a change of the ambient temper-ature of the MAS, inherent resistance and linearization isconsidered using the MAS manufactures spec sheet. For ameasured sensor temperature of 300◦C an error of 0.58K hasto be considered. �e con�dence interval of 95.5% is not usedbecause the assumption is made that all errors would occur atthe same time. �erefore, the present uncertainty analysis isconducted with a con�dence interval of 68.3%. �is leads to acombined standard uncertainty of uTS=0.932K for the turbinetemperature measurement by means of thermocouples TypeK.

�e Pt100 resistance thermometers are positioned in the�uid �ow path in the same way as the Type K TC thus heatconduction is not considered. Deviations from the charac-teristic curves are again delivered by the manufacturer. ForPt100 Class A the limiting deviation is set at 0.75K for themaximum occurring temperature of 300◦C at the compres-sor outlet. Although internal calibration studies have shownthat the error is below 0.75K, the standard manufacture valueof Pt100 class A is used to consider the maximum error in-�uence. In contrast to the Type K TC, errors for the Pt100may occur due to thermo-voltage and self-heating of the sen-sor. Since the supply current of 1mA is too low to in�uencethe temperature acquisition, the self-heating of the sensoris neglected. �e thermo-voltage is considered according toScheller. �is leads to a combined standard uncertainty of


Input

parameter


σHC Heat conduction Normal Unattended because Pt 100 is

insulated to stem

σDIN Deviation to DIN Rectangular 0.75K/ 3=0.433K

σStab Stabilization period Cancelled because all operating points are

measured in steady state

σShS Self-heating of the

sensor

Rectangular Unattended since supply current of

1mA is used

σThstr Thermo-voltage Rectangular <60mK within temperature interval 0

to 500°C [1]

0.06K/ 3=0.035K

σVs Voltage supply Rectangular Measured temperatures

-200 to 150°C: 0.15K

150 to 850°C: 0.2K

Max:

0.2K/ 3=0.115K

σTam Ambient

temperature of MAS

Rectangular

σLin Treatment and

linearization

Rectangular

σInh Inherent resistance Rectangular

σLts Long term stability

of electronic

components

Rectangular Cancelled

Combined standard uncertainty with 68.3% confidence interval

𝒖𝑻𝑺 = 𝟎. 𝟓𝟖𝟑𝑲

Figure 7. In�uencing variables for temperature measure-ment Pt100

uTS=0.583K.

2.2.2 Uncertainty of pressure measurementDue to the di�erent pressure levels during the operationmode of the turbocharger, a selection of pressure transduc-ers was installed at the in- and outlet measuring planes ofthe turbocharger. �e pressure measurement chain is mod-eled by means of Equation 8. Likewise to the temperaturemeasurement model, the output signal lPi is composed ofthe acquired pressure VPi and the accumulated uncertaintyvariables, which are explained in Figure 8 as a function ofthe full scale output (FSO) of the sensor measuring interval.

lPi = VPi +σMt +σMw +σMa +σMv +σDPT +σMAS (8)

Input

parameter


σMt Transient effects Cancelled

σMw Wall influence Unattended

σMa Influence of Ma-

number

Unattended, since Ma < 0.3

σMv Viscosity impact Unattended, since Ma < 0.3

σDPT DPT Zero pressure

offset

Unattended due to calibration

Linearity/Hystere

sis

Rectangular 0.1% FSO

Temperature

offset

Rectangular 0.1mV

Temperature

span

Rectangular 0.2% FSO

Long-term

stability

Unattended since < 1 year

σMAS

MAS

Gain Error Rectangular 160ppm of reading

Offset Error Rectangular 156ppm of reading

Noise

Uncertainty

Rectangular 0.62μV

Figure 8. In�uencing variables for pressure measurement

Likewise to the temperature measurement, transient ef-fects can be ignored. �e wall in�uence is neglected sinceonly the static pressure is acquired using wall pressure holes.�e in�uence of the Mach number has not been taken into

account because occurring Ma-numbers are less than 0.3for the measured rotational speeds. �is assumption is alsoapplicable for the viscosity impact, since it only has to beconsidered for Re ≤ 1000. �e accuracies of the di�erentialpressure transducers (DPT) are speci�ed by the manufacturerand reviewed. In addition to the measurement instrumenta-tion uncertainty, the static pressure wall tap diameters needto be included in the pressure unvertainty analysis as well.Bore diameters of 5mm were chosen for all measurementplanes. �e uncertainty value was calculated from the ra-tio of manufactured pressure hole diameter to the �ow areaat each measurement plane based on [10]. Absolute uncer-tainty values for each measuring plane are given in Table 2using the diameter ratio of 0.4 to consider the hole diameteruncertainty .

Table 2. Absolute uncertainty values for all turbochargermeasuring planes

Measuring plane Combined uncertainty 68.3%

MP 1 3.10 mbarMP 2 6.05 mbarMP 3 6.05 mbarMP 4 6.40 mbar

2.2.3 Performance map data�e following section is dedicated to show the acquired datafrom which the performance maps of the turbine and thecompressor are generated. �e acquired turbocharger turbineperformance map is shown in Figure 9. �e correspondingcompressor performance map is shown in Figure 10. �e per-formance map is acquired by successively closing the thro�levalve at the compressor outlet while maintaining a constantturbine mass �ow and turbine inlet temperature. As soon asan increase in rpm occurs, the mass �ow on the turbine sideis readjusted. �us providing the data for a characteristiccurve for a de�ned turbine speed. �e performance datais acquired until reaching the surge line of the compressorlimiting the width of the acquired rpm curve. �e expectedtrend of the characteristic turbine rpm curves can be seen inFigure 9.

Due to compressibility e�ects the reduced mass �owbends towards an asymptote and decreases a�er peakingat a turbine pressure ratio of approximately 3.5. Since thecompressor side of the turbocharger is operated at ambientconditions, the inlet pressure varies depending on the rota-tional speed due to the ori�ce plate mass �ow measurementupstream of the compressor inlet. �e variation of the inletpressure at compressor inlet in reference to the rotationalspeed is shown in Figure 11. To further improve performanceinvestigation, the ambient operation should be switched to aclosed-loop operation, which can be integrated into the teststand. Since in turbocharger operation the focus lies on theacquisition of the turbine performance where the compressorside is merely used to adjust the load, the closed-loop teststand version will o�er a much wider range for obtaining per-


0.55

0.60

0.65

0.70

0.75

0.80

0.85

0.90

0.95

1.00

1.05

1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0

Norm

aliz

ed

red

uce

d tu

rbin

e m

ass flo

w [-]

Turbine pressure ratio (total-static) [-]

25.5% rpm_max 27.3% rpm_max





Figure 9. Acquired turbocharger turbine performancemap

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1

Co

mpre

ssor

pre

ssu

re r

atio (

tota

l-to

tal) [-]

Normalized reduced compressor mass flow [-]






Figure 10. Acquired turbocharger compressor perfor-mance map

formance map data. Repeatability investigations were con-ducted for an individual reference operation point, de�ned byturbocharger rotational speed and total-static turbine pres-sure ratio, for certain rotational speeds at two di�erent testdays. �e results are listed in 3 and show an excellent overallrepeatability for turbine perfomance operation points.

Table 3. Repeatability of turbine performance map data attwo di�erent test days

Rotational speed Error mred [%]

44.2% rpmmax 0.0751.9% rpmmax 0.2756.8% rpmmax 0.1960.3% rpmmax 0.0377.4% rpmmax 0.0989.5% rpmmax 0.32

When discussing thermal stability and the in�uence ofmeasurement uncertainty on performance map data, the tur-

0

50

100

150

200

250

300

20 30 40 50 60 70 80 90 100

Pre

ssu

re d

iffe

ren

ce

pC

I -

pam

bie

nt [m

ba

r]

Rotational speed in reference to maximum rotational speed [%]

Figure 11. Pressure variation of the compressor inlet inreference to the rotational speed

bine e�ciency needs to be assesed. Due to the di�culty inthe assesment of turbine e�ciency based on measurementsat the in- and outlet piping of the turbocharger turbine, it hasbecome common, to introduce the turbine thermomechanicale�ciency [11].

ηe� =Ppol,compressor

Pis,turbine(9)

where Ppol,compressor is the polytropic power output ofthe turbocharger compressor and Pis,turbine is the isentropicturbine power output calculated on the basis of an isentropicturbine outlet temperature.

∆his,turbine = cpturbine(Tbt,tot − Tat,is) (10)

�e calculated normalized e�ective e�ciency is displayedin Figure 12 in regard to the total-static turbine pressure ratio.�e analysis of the e�ciency curves supports the assesment,that good thermal stability was achieved during steady stateoperation.

0.80

0.82

0.84

0.86

0.88

0.90

0.92

0.94

0.96

0.98

1.00

1.02

1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0

No

rmaliz

ed turb

ine e

ffic

iency [-]

Turbine pressure ratio (total-static) [-]

n = 20435 n = 20669 n = 22591

n = 23987 n = 30790 n = 35618

44.2% rpmmax 51.9% 56.8%

89.5% 77.4% 60.3% rpmmax

Figure 12. Normalized e�ective e�ciency of the tur-bocharger turbine at 220◦C turbine inlet temperature


3. CONCLUSION AND OUTLOOK�e developed turbocharger test stand provides an oppor-tunity to operate the test turbocharger at stable conditions.Testing has shown, that the requirements of the thermody-namic parameters, like the turbine inlet temperature, arefull�lled. Hence the operation point is stable enough thatdata for performance map purposes can be acquired at thepresented test stand. �e operation of the test stand at con-stant turbine inlet temperatures, especially at below real-time application, can be used for investigations where theimplemented measurement instrumentation is limited bymaximum operation temperature. �e required equilibriumstate between compressor and turbine of the turbochargerwas proven to be obtained during the performance map dataacquisition by comparing the outlet temperatures of bothsides. A stability interval of the rotational speed of 2 Hz wasaccomplished. �e uncertainty analysis for the applied mea-surement instrumentation shows, that the acquired data lieswithin an uncertainty interval which su�ces the requiredmeasurement accuracy for obtaining the needed thermody-namic parameters for generating the displayed performancemaps as well as CFD validation. A constant oil inlet temper-ature within tolerances of +/- 0.5K was also achieved duringtesting operation of the turbocharger.

In regard of the acquired test data the developed test standcan be modi�ed for further investigations on turbochargerperformance and e�ciency. �e displayed measurementrange of the turbocharger compressor can be identi�ed asthe limiting factor on the possible measurement range. Sincethe electrical preheater is less sensitive to low �ow turbineconditions, the compressor side needs to be adjusted towardslower mass �ow. Internal investigations have shown, that byadjusting the compressor side pressure losses (usage of twoparallel pipes equiped with ori�ce plates) the measureableperformance range can be increased. Furthermore, the men-tioned variation in compressor inlet pressure can be adressedby implementing a closed-loop compressor cycle with inte-grated intercooler. �us providing similar compressor inletconditions independent from the rotational speed. In a �rststep Also the installed lubracating oil ciruit o�ers the opor-tunity to adjust the oil inlet temperature in a wider scopewhich allows a large spectrum of test turbochargers thatcan be investigated. Furthermore the compressor facilityat IKDG provides enough mass �ow to enable testing in aturbocharger scope of application from small automotive upto marine turbocharger applications.

ACKNOWLEDGMENTS�e authors would like to thank the research associationForschungsvereinigung Verbrennungskra�maschinen e. V.and the Association Industrial Research (AIF) for the �nan-cial support of this research project. �e authors would alsolike to gratefully acknowledge Kompressorenbau BannewitzGmbH, whose support was essential for the present investi-gations.

REFERENCES[1] G. Scheller. Messunsicherheit einer Temperaturmesske�e

mit Beispielrechnungen. 2003.[2] Barros J. Venson, G. and J. Pereira. Development of an

automotive turbocharger test stand using hot gas. SAETechnical Paper Series 2006-01-2680, 2006, 2006.

[3] J. Byron and E. Figge. Turbocharger performance evalua-tion on a closed loop test stand. National Gas MachineryLaboratory, Kansas State University.

[4] M.Idzior et al. A concept of a turbocharger test stand.Combustion Engines, 2014.

[5] M.Y. Young and D.A. Penz. �e Design of a New Tur-bocharger Test Facility. 1990.

[6] Marmorato G. Capobianco M. Marelli, S. andJ. Boulanger. Towards the direct evaluation ofturbine isentropic e�ciency in turbocharger testing.SAE Technical Paper 2016-01-1033, 2016, 2016.

[7] Seume J. Shaaban, S. Impact of turbocharger non-adiabatic operation on engine volumetric e�ciency andturbo lag. International Journal of Rotating Machinery,2012.

[8] W. Kummel. Technische Stromungsmechanik. �eorie undPraxis. 3.Au�age. 2007.

[9] Rakut C. Diefenthal, M. andH. Tadesse. Abschlussbericht:Transientes Temperaturfeld in einem Radialturbinenrad.2014.

[10] AnasRahman A. Yeong T.K. Choon, T.W. and L.E. Aik.Static hole error analysis within laminar and turbulentregime using cfd approach. IEEE Colloquium on Human-ities, Science ans Engineering Research, 2012.

[11] K. Zinner. Supercharging of internal combustion engines.Springer Verlag, Berlin, 1978.

Documents

Commissioning of a Test Stand for Turbocharger ...isromac-isimet.univ-lille1.fr/upload_dir/finalpaper17/50.finalpaper.pdf · Commissioning of a Test Stand for Turbocharger Investigations