44
Chapter 1 GENERAL DESCRIPTION 1.1 IMPORTANCE OF THE LEAK TEST ・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・ 3 1.2 THE INFLUENCE OF LEAKAGE VOLUME AND THE LEVELS OF ALLOWABLE LEAKAGE RATES ・・・3 1.3 A VARIETY OF LEAK TEST METHODSLEAKAGE OF GAS OR FLUID ・・・・・・・ 4 1.4 LEAKAGE OF GAS OR FLUID ・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・ 4 1.4.1 States Of Leak Flow Through A Hole 1.4.2 Factors Influencing Flow State 1.4.3 Leakage of Gas or Fluid (Turbulent Flow) 1.4.4 Leakage of Gas or Fluid (Viscous Flow) 1.4.5 Molecular Flow Chapter 2 LEAK TEST 2.1 DIFFERENTIAL PRESSURE TYPE LEAK TEST ・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・ 13 2.1.1 Operating Principle 2.1.2 Basic Operation 2.1.3 Calculation Equation of Leak Volume 2.1.4 Applied Measuring Method 2.1.5 For Performance Enhancement 2.2 DIRECT PRESSURE DETECTION AIR LEAK TEST ・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・ 22 2.2.1 Operating Principle 2.2.2 Basic Operation 2.2.3 Calculation Equation OF Leak Volume 2.3 HYDROGEN GAS DETECTION LEAK TEST ・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・ 23 2.3.1 Operating Principle 2.3.2 Basic Operation 2.3.3 Calculation Equation OF Leak Volume 2.3.4 Applied Measurement Method 2.4 HELIUM GAS DETECTION LEAK TEST ・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・ 27 2.4.1 Operating Principle 2.4.2 Basic Operation 2.4.3 Calculation Equation OF Leak Volume 2.4.4 Applied Measurement Method 2.5 FLOW RATE LEAK TEST ・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・ 31 2.5.1 Operating Principle 2.5.2 Basic Operation 2.5.3 Calculation Equation of Leak Volume 2.5.4 APPLIED MEASUREMENT METHOD Chapter 3 PERIPHERAL EQUIPMENT 3.1 EXHAUST BYPASS UNIT ・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・ 35 3.2 PRESSURIZATION & EXHAUST BYPASS UNIT ・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・ 35 3.3 FLOW MASTER ・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・ 36 3.4 CALIBRATOR ・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・ 36 3.5 SUPER ELECTROPNEUMATIC REGULATOR (APU) ・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・ 37 Chapter 4 INFORMATION ABOUT EQUIPMENT 4.1 MANAGEMET OF WATER, OIL AND FOREIGN MATERIALS ・・・・・・・・・・・・・・・・・・・・・・・・・・ 39 4.2 MANAGEMENT OF SOURCE PRESSURE AND FLOW RATE ・・・・・・・・・・・・・・・・・・・・・・・・・・ 40 4.3 CLAMPING AND SEALING JIG ・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・ 41 4.3.1 CLAMPING 4.3.2 SEALING JIG 4.4 CHANGE IN TEMPERATURE ・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・ 43 CONTENTS

Fukuda Technical Manual en 00

Embed Size (px)

Citation preview

Page 1: Fukuda Technical Manual en 00

Chapter 1 GENERAL DESCRIPTION 1.1 IMPORTANCE OF THE LEAK TEST ・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・ 3 1.2 THE INFLUENCE OF LEAKAGE VOLUME AND THE LEVELS OF ALLOWABLE LEAKAGE RATES ・・・ 3 1.3 A VARIETY OF LEAK TEST METHODSLEAKAGE OF GAS OR FLUID ・・・・・・・ 4 1.4 LEAKAGE OF GAS OR FLUID ・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・ 4

1.4.1 States Of Leak Flow Through A Hole 1.4.2 Factors Influencing Flow State 1.4.3 Leakage of Gas or Fluid (Turbulent Flow) 1.4.4 Leakage of Gas or Fluid (Viscous Flow) 1.4.5 Molecular Flow

Chapter 2 LEAK TEST 2.1 DIFFERENTIAL PRESSURE TYPE LEAK TEST ・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・ 13

2.1.1 Operating Principle 2.1.2 Basic Operation 2.1.3 Calculation Equation of Leak Volume 2.1.4 Applied Measuring Method 2.1.5 For Performance Enhancement

2.2 DIRECT PRESSURE DETECTION AIR LEAK TEST ・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・ 22 2.2.1 Operating Principle 2.2.2 Basic Operation 2.2.3 Calculation Equation OF Leak Volume

2.3 HYDROGEN GAS DETECTION LEAK TEST ・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・ 23 2.3.1 Operating Principle 2.3.2 Basic Operation 2.3.3 Calculation Equation OF Leak Volume 2.3.4 Applied Measurement Method

2.4 HELIUM GAS DETECTION LEAK TEST ・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・ 27 2.4.1 Operating Principle 2.4.2 Basic Operation 2.4.3 Calculation Equation OF Leak Volume 2.4.4 Applied Measurement Method

2.5 FLOW RATE LEAK TEST ・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・ 31 2.5.1 Operating Principle 2.5.2 Basic Operation 2.5.3 Calculation Equation of Leak Volume 2.5.4 APPLIED MEASUREMENT METHOD

Chapter 3 PERIPHERAL EQUIPMENT 3.1 EXHAUST BYPASS UNIT ・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・ 35 3.2 PRESSURIZATION & EXHAUST BYPASS UNIT ・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・ 35 3.3 FLOW MASTER ・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・ 36 3.4 CALIBRATOR ・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・ 36 3.5 SUPER ELECTROPNEUMATIC REGULATOR (APU) ・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・ 37

Chapter 4 INFORMATION ABOUT EQUIPMENT 4.1 MANAGEMET OF WATER, OIL AND FOREIGN MATERIALS ・・・・・・・・・・・・・・・・・・・・・・・・・・ 39 4.2 MANAGEMENT OF SOURCE PRESSURE AND FLOW RATE ・・・・・・・・・・・・・・・・・・・・・・・・・・ 40 4.3 CLAMPING AND SEALING JIG ・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・ 41

4.3.1 CLAMPING 4.3.2 SEALING JIG

4.4 CHANGE IN TEMPERATURE ・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・・ 43

CONTENTS

Page 2: Fukuda Technical Manual en 00

1

A VARIETY OF UNITS FOR PRESSURE AND THEIR CONVERSIONS

Pa atm Torr kg/c ㎡ bar mmH2O lb/in2 inchHg Note

1Pa = 1N/m2 = 1 9.869 23 ×10??

7.500 62 ×10?3

1.019 72 ×10?5

1 ×10?5

1.019 72 ×10?1

1.450 38 ×10?4

2.953 00 ×10?4 SI unit

1atm = 1ata = 1.013 25

×10? 1 7.6 ×102

1.033 23 ? *2

1.013 25 ?

1.033 23 ×104

1.469 60 ×101 *2

2.992 13 ×101

To be differentiated from gauge pressure (atg)

1Torr =1mmHg *6 = 1.333 22 ×102

1.315 79 ×10?3 1 1.359 51

×10?3 1.333 22

×10?3 1.359 51

×101 1.933 68

×10?2 3.937 01

×10?2

1kg/cm2 = 1at *7 9.806 65 ×104

9.678 41 ×10-1

7.355 59 ×102 1 9.806 65

×10?1 1

×104 1.422 34

×101 2.895 90

×101 The unit "at" is seldom used.

1bar *3 1

×105 9.869 23

×10?1 7.500 62

×102 1.019 72

? 1 1.019 72 ×104

1.450 38 ×101

2.953 00 ×101

May be used as an SI unit.

1mmH2O = 1mmAq *4 9.806 65 ?

9.678 41 ×10?5

7.355 59 ×10?2

1 ×10?4

9.806 65 ×10?5 1 1.422 34

×10?3 2.895 90

×10?3

1lb/in2 = 1psia *5 6.894 76 ×103

8.804 60 ×10?2

5.171 50 ×101

7.030 69 ×10?2

6.894 76 ×10?2

7.030 72 ×102 1 2.036 02

? To be differentiated from psig.

1inchHg = 3.386 39 ×103

3.342 10 ×10?2

2.54 ×101

3.453 15 ×10?2

3.386 39 ×10?2

3.453 15 ×102

4.911 55 ×10?1 1 Often used to indicate

negative pressure. *1: Density of water is assumed as 1000 kg/m 3 at 4 ℃. For water of 15℃ (Density: 0.9991), 1 atm = 1.03416 × 104mmH2O. *2: To avoid misprinting, ×1 is shown by ? , and ×10 by 101. *3: 1 bar = 106 dyne/cm2 = 0.1 MPa ≒1 atm *4: Often used are 1 mH2O = 1 mAq = 103mmH2O *5: Stands for pound per square inch, absolute. *6: 1μHg = 10-3mmHg = 10-3Torr , 1 Torr = 1 mmHg = 1000 μHg *7: This kg means kg forces (kgf) or kg weight (kgw). *8: This is pronounced as “pi’: e’i” until the Measurement Law is enacted.

Hydraulic Submerge Method

Flow Rate Measurement Method

Direct Pressure TypeAir Leak Detection

Differential Pressure TypeAir Leak Detection

Special TypeAir Leak Detection

Hydrogen Gas Leak Detection

Helium Gas Leak Detection

10-6 10-7 10-8 10-910-2 10-3 10-4 10-5 10-10

Relationship between Measurement Hardware and Detectable Leak Rate

Detectable Range ml/sec

103 102 101 100 10-1

Mea

sure

men

t H

ardw

are

Page 3: Fukuda Technical Manual en 00

2

GENERAL DESCRIPTION

Chapter 1

Page 4: Fukuda Technical Manual en 00

3

1.1 IMPORTANCE OF THE LEAK TEST In this leaflet, the term “leakage” defines the leakage of gas or fluid from a product, which will

affect the overall quality of the product. The problem of leakage includes a variety of meanings, depending upon the industrial fields

where the term is used. For example, with fuel handling products, leakage can introduce a danger of a fire, explosion, runaway, etc., while with other products it refers to low performance and economic loss. In the food and drug industries it can cause a change in properties and/or shade, or even rot in case of food. A small leak can grow into an unexpected size then creates product liability problems, which at times, can become a crucial damage to a company.

1.2 THE INFLUENCE OF LEAKAGE VOLUME AND THE LEVELS OF ALLOWABLE LEAKAGE RATES

For example, in the automotive components and parts, a comparison is made between the

air-conditioner, gasoline filter and silencing muffler. The air-conditioner contains a fixed volume of a coolant medium that circulates to cool the air

passed into the car. Leakage of the coolant gas directly means a lower cooling effect, and the allowable leakage rate during the life of the air-conditioner (several years) must be maintained within a range that can keep the cooling effect satisfactory.

The gasoline filter is positioned close to the hot engine, and gasoline is a dangerous fuel with a high- ignition point. Leakage of gasoline is obviously dangerous, for it can cause a fire or an explosion. However a gasoline leak will evaporate by its self, therefore a leak which does not evoke a danger must be set at the allowable leakage rate.

The primary role of an automotive silencing muffler is to reduce exhaust noise. After ignition, the exhaust gas is expelled from the engine and diffused into the atmosphere through the muffler. The muffler is a tube which is placed in the path to expel the exhaust gas in to the atmosphere. A leakage from the wall of the muffler cylinder which does not affect the performance must be set at the allowable leakage rate.

As you can see from the above examples, the allowable leakage rates differ from the character to the character of the products and they must be set at leakage rates which don’t cause any performance and safety problems.

Page 5: Fukuda Technical Manual en 00

4

1.3 A VARIETY OF LEAK TEST METHODS

Table 1.3 Leak Testing Methods

1.4 LEAKAGE OF GAS OR FLUID

1.4.1 States of Leak Flow Through a Hole When considering a situation where gas flows through a leak hole, the state of the flow depends

on various conditions such as the diameter of the hole, the length of the hole (or the wall thickness of the work) and the pressure within the work. These states can be roughly classified as below and suitable leak measuring methods exist for each state.

Fig. 1.1 Leak Test Methods for Various Flow Statuses

Large Flow

Small Flow

Turbulent Flow

Viscous Flow

Molecular Flow

Measured by Flow Rate or with Direct Pressure Method

Measured with Differential Pressure Method

Measured with Helium Gas Method

Measured with Hydrogen Gas Method

Page 6: Fukuda Technical Manual en 00

5

1.4.2 Factors Influencing Flow State When considering the flow in a large pipe, the flow velocity of fluid at the wall of the pipe is 0

because the flow stays with the wall due to the viscosity of the fluid. The velocity increases as the position of the flow is measured away from the wall because the influence of viscosity reduces and the flow becomes dominated by inertia. The layer which is influenced by both viscosity and inertia is called a boundary layer and the layer becomes thin when the flow velocity of fluid is high or the viscosity of fluid is low. The flow state which is dominated by inertia is turbulent flow. By contrast, when considering the flow in a small pipe, a boundary flow develops near the wall of the pipe due to viscosity and it merges at the center area of the pipe. The flow has a velocity distribution with a parabolic shape and this flow state which is dominated by viscosity is viscous flow (or laminar flow). The Reynolds number (Re) indicates the boundary of the 2 types of flows.

Re = U・D/ν

U: Mean Flow Velocity within Pipe

D: Inner Diameter of Pipe

ν: Kinematic Viscosity Coefficient

When the Reynolds number is larger then 2200 (Re>2200), the flow is turbulent flow, when the

number is smaller than 1200 (Re<1200), the flow is viscous flow, and when the number lies between the 2 values, the flow is intermediate flow in which viscosity and inertia influence each other. If the diameter of the pipe decreases further, the flow can be described as follows. Viscosity represents a state in which the molecules of fluid collide with each other and influence each other, but the flow within a pipe becomes independent of its viscosity when the diameter of the pipe is smaller than the distances of the collision of molecules (mean free path). This state is molecular flow and the Knudsen number (Kn), a dimensionless number, determines the state of flow.

Kn = λ/D

λ: Mean Free Path of Gas

D: Inner Diameter of Pipe

Flow state can be classified as follows according to the Knudsen number:

Kn>1 Molecular Flow

Kn<0.01 (Re<1200) Viscous Flow

0.01<Kn<1 Intermediate Flow

In this leaflet, differences in the state of a pipe are considered because leak flow is being

considered. However, laminar flow or molecular flow will occur even in a large pipe when the pressure in a pipe is reduced, because it causes a smaller kinematic viscosity coefficient or a longer mean free path.

Page 7: Fukuda Technical Manual en 00

6

1.4.3 Leakage of Gas or Fluid (Turbulent Flow)

Calculation Equation of Leak Volume

Q: Air Flow Volume (L) S: Effective Sectional Area (mm2) P1: Upstream Pressure (MPa) P2: Downstream Pressure (MPa) t: Temperature (℃)

1.4.4 Leakage of Gas or Fluid (Viscous Flow) Leakage is a phenomenon which occurs when a fluid such as air, water, or oil passes through an

unintended opening such as a small hole. A resultant leak volume differs according to the difference of pressure across the opening; and relates with the ease of fluid flow through the opening; conductance.

This can be expressed by the following equation. Q = C (P1 – P2) ---------------------------------------------------(1.1)

Where Q is leak volume, P1 – P2 represents the difference between two pressures, and C is conductance. When leakage is the subjective problem, the ease of fluid flow (C) depends on a variety of factors, including the configuration of the opening, length, etc. It is therefore difficult to apply one type of equation to all cases. In this section, an explanation is given using general application equations. Theoretical Equation of Leak Volume

As a representative theoretical equation to explain the behavior of fluids passing through a very

narrow opening, the Hagan-Poiseuille Law is often used. According to this law if the opening is so small that the flow of fluid is within a range of viscous flow (laminar flow), and the ratio of the hole length vs. the hole diameter is large enough, the following equation can be applied;

-----------------------------------------(1.2) Where Qa is the volumetric flow of outlet side pressure (atmospheric pressure) converted from

compressible fluid such as air. However, if P1 is negative, the leak volume is expressed in terms of the state of atmospheric

pressure. The P2 in the denominator in equation 1.2 is replaced with P1. With a volumetric flow rate QW representing non-compressive fluid such as water, oil, etc. the following equation is applied;

-----------------------------------------(1.3)

293 Q=120×S(P1+0.1)

273+t

Q=240×S (P2+0.1)(P1-P2)293

273+t

>0.5 then the flow is subsonic flow. P2+0.1

P1+0.1

≦0.5 then the flow is chocked flow. P2+0.1

P1+0.1

2

22

21

4

a P16?6)P(PpR

Ql−

=

lW

214

W 8?)P(PpR

Q−

=

Page 8: Fukuda Technical Manual en 00

7

aQ : Volumetric flow rate of compressive fluid (air) under pressure P2

WQ : Volumetric flow rate of non-compressive fluid

P1: Primary (test) pressure

(when negative pressure, atmospheric pressure)

P2: Secondary (test) pressure

(when negative pressure, test pressure)

R: Radius of the opening

l : Length of the opening

a? : Viscosity of compressive fluid

w? : Viscosity of non-compressive fluid

The relationship of volumetric flow rate influenced by the difference between gaseous fluid and liquid, and between two pressures with the same test piece is shown in Table 1.2. Refer to equations 1.1 and 1.2.

Table 1.2 Relationship between Leak Volume vs. Test Pressure and Viscosity Condition Relative Equation

The ratio of leak volume to the different test pressures towards the same gas .

y2

2y2

1

x2

2x2

1

ay

ax

PPPP

QQ

−−

=

The ratio of leak volume to the fixed test pressures towards different gases.

ax

ay

ay

ax

??

QQ

= Gas

The ratio of leak volume to the different test pressures towards different gases.

y2

2y2

1

x2

2x2

1

ax

ay

ay

ax

PPPP

?

?

QQ

−−

×=

The ratio of leak volume to the different test pressures towards the same liquid.

2y1y

2x1x

wy

wx

PPPP

QQ

−−

=

The ratio of leak volume to the fixed test pressure towards different liquids.

wx

wy

wy

wx

??

QQ

= Fluid

The ratio of leak volume to the different test pressures towards different liquids.

wy

wx

QQ

= yy

xx

PPPP

21

21

wx

wy

?

?

−−

×

The ratio of leak volume to the different test pressures towards different gas es and liquids. )P(PP

)P(P2?

?

QQ

2y1y2x

x2

2x2

1

xa

wy

wy

ax

−−

×=

Gas/ Liquid

The ratio of leak volume to the fixed test pressure towards different gases and liquids.

=wy

ax

QQ

2

21

xa

wy

P)P(P

2?? +

×

Fig. 1.2 Theoretic Model of Leakage

?

P1

P2

Pin-hole of Radius R

Page 9: Fukuda Technical Manual en 00

8

Viscosity (Viscosity Coefficient) η Viscosity is one of the important factors when handling fluid. Different units are used in

different fields. Here are some examples; 1P = 1 dyne・sec・cm-2 = 1gr・sec-1・cm-1 = 10-1N・sec・m-2 = 10-1Pa・sec = 1.02×10-6kgf・sec・cm-2 P = Poise N = Newton kgf・sec・cm-2: Engineering Unit

(kgf second per square centimeter) Absolute Unit 1gr・sec-1・cm-1

= Engineering Unit 1.02×10-6kgf・sec・cm-2×980.4cm・ sec-2

Pa: Pascal Also, to represent kinematical viscosity of the fluid, use the equation;

?= ?/? But ?is the density of the fluid. Some examples of the viscosity coefficients for air, water, brake oil and gasoline are shown in Table 1.3.

Leak Rate Calculation Using a Variety of Units of the Viscosity Coefficient (Pressure: Absolute Pressure)

(1) Calculation of Leak Rate in Compressive Fluid

Using equation 1.2, volumetric flow rates of compressive fluid based on a variety of calculations units are summarized in Table 1.4.

Table 1.4 Leak Rate of Compressive Fluid

a

R: (m); l : (m); ? a: (Pa, sec); P1, P2: (Pa)

2

2121

a

413

PPP

2PP

?R

103.925sec)/ (m Q−

×+

××= −

l

b

R: (cm); l : (cm); ? a: (Pa, sec); P1, P2: (Pa)

2

2121

a

41

PPP

2PP

?R

103.925sec)(cc/ Q−

×+

××= −

l

c

R: (cm); l : (cm); ? a: kg・sec/ cm2; P1, P2: (kg/ cm2)

2

2121

a

41

PPP

2PP

?R

103.925sec)(cc/ Q−

×+

××= −

l

d

R: (cm); l : (cm); ? a: ( 25 sec/cm kg

109.806651

×); P1, P2: (kg/ cm2)

2

2121

a

45

PPP

2PP

?R

103.879sec)(cc/ Q−

×+

××=l

Table 1.3 Viscosity Coefficient of Gases and Liquids

Note: R: Radius of Pipe, l : Length of Pipe ,?a: Viscosity of Fluid , P1: Primary Absolute Pressure, P2: Secondary Absolute Pressure

Fluid Temperatur Viscosity0℃ 1.71×10-4P20℃ 1.81×10-4P50℃ 1.95×10-4P70℃ 2.04×10-4P0℃ 1.79×10-2P20℃ 1.00×10-2P50℃ 0.55×10-2P70℃ 0.40×10-2P20℃ 26×10-2P50℃ 10×10-2P70℃ 7×10-2P

Gasoline 20℃ 6×10-3P

Air

Water

BrakeOil

Temperature

Page 10: Fukuda Technical Manual en 00

9

(2) Calculation of Leak Rate in Non-Compressive Fluid Using equation 1.3, volumetric flow rates of non-compressive fluid based on a variety of

calculation units are summarized in Table 1.5

Table 1.5 Leak Rate of Non-compressive Fluid

a

R: (m); l : (m); ?w: (Pa, sec); P1, P2: (Pa)

)P(PR

103.925/sec)Q(m 21?

413

w

−××= −

l

b

R: (cm); l : (cm); ?w: (Pa, sec); P1, P2: (Pa)

)P(PR

103.925Q(ml/sec) 21?

41

w

−××= −

l

c

R: (m); l : (cm); ?w: kg・sec/cm2; P1, P2: (kg/cm2)

)P(PR

103.925Q(ml/sec) 21?

41

w

−××= −

l

d

R: (cm); l : (cm); ?a: )sec/cm kg109.80665

1(P 2

5×= ; P1, P2: (kg/cm2)

)P(PR

103.849Q(ml/sec) 21?

45

w

−××=l

Leak Rate Conversion from Air to Liquid

R: Radius of Pipe, l : Length of Pipe, ?w: Viscosity of Fluid, P1: Primary Absolute Pressure, P2: Secondary Absolute Pressure

Using Table 1.2 and Table 1.3 the leak rates of water, gasoline, and brake oil in reference to air are calculated, provided that the same test piece is used. The temperature of fluid is kept fixed at 20℃, and the test pressure is kept at the same level. The results are as shown Fig. 1.3.

Fig. 1.3 Volumetric Flow Rate of Fluid Referred to Air

Volum

e - Flow

Ratio

Gasoline

Water

Primary Pressure kg/ cm2

Measuring Temperature : Constant at 20℃

Page 11: Fukuda Technical Manual en 00

10

1.4.5 Molecular Flow When the mean free path ? of gas is larger than the inner diameter D of a pipe, the molecules of

gas only collide with the wall of the pipe and rarely collide with each other.

In this state, the interaction among molecules (viscosity) is weak and the flow becomes molecule flow in which collisions of the molecules and the wall are dominant. By defining the Knudsen number (Kn), a dimensionless number, as Kn = λ/D, the state of flow can be classified as follows. ・Kn<0.01 Re<1200 Viscous Flow

・Kn>1 Molecular Flow

When gas at normal temperature flows in a straight circular pipe it is called molecular flow, it can be expressed by the following equation.

Q = ×(P1-P2)

Q: Flow Rate, Pa・m3/sec

a: Radius of Pipe, cm L: Length of Pipe, cm M: Molecular Weight of Gas P1: Inlet Pressure, Pa P2: Outlet Pressure, Pa

D

○ ○

○ ○

○ ○ ○

○ ○

○ λ

Fig. 1.4 State in the Pipe (with Large Diameter)

Molecules often collide with each other in the flow.

λ

D

0.523a3

M1/2・L

Fig. 1.5 State in the Pipe (with Small Diameter)

Molecules rarely collide with each other in the flow.

Page 12: Fukuda Technical Manual en 00

11

MEMO

Page 13: Fukuda Technical Manual en 00

12

LEAK TEST

Chapter 2

Page 14: Fukuda Technical Manual en 00

13

2.1 DIFFERENTIAL PRESSURE TYPE LEAK TEST

2.1.1 Operating Principle In the field of medicine the use of a balancing scale is often used as a precise method for

calculating arbitrarily small amounts of substance. The instrument is used to precisely measure a weight close to the balancing weight (a reference value) by increasing the weight value on the opposite side (Fig.2.1). The FUKUDA Air Leak Testers use the same principle as the balancing scale. The same air pressure is charged to both the work to be tested and the master work and the change in pressure balance within a fixed time is checked for the presence of a leak. If there is no leakage, the charged air pressure shows no change and the balanced status is sustained for a long time (Fig. 2.2). However, if leakage occurs, the pressure in the tested work decreases with time and this leads to an unbalanced status (Fig. 2.3). The leak rate (ml/sec) determines how fast the unbalanced status occurs.

The basic circuit of the FUKUDA Air

Leak Testers is shown in Fig. 2.4 and air pressure from the source is charged into both the tested work and the master work through the valves (A) and (B). After closing the valves (A) and (B), the pressure in the tested work is measured in reference with that in the master work using the differential pressure sensor.

Fig. 2.5 shows the basic pneumatic circuit actually installed in the FUKUDA Air Leak Testers.

The solenoid valves SV1, SV2, and SV3 are opened to charge air into both the tested work and the master work at a certain pressure. Then SV2 and SV3 are closed. If the tested work has a leakage, the inside pressure of it gradually decreases according to its leak rate and a pressure difference against that in the master work occurs. This differential pressure is detected by the differential pressure sensor. The sensor converts the pressure into an electrical signal and it is displayed on the tester. The FUKUDA Air Leak Testers utilize the differential pressure value measured after a preset elapsed time to make an OK/NG judgment automatically.

Fig. 2.2 Charge Same Pressure and Check for Pressure Balance

Fig. 2.3 If Leakage Occurs, the Balance will be lost

Fig. 2.1 Precise Weight Measurement of Medicine with Balancing Scale

Master Work

No Leak With Leak (Needle Moves)

Valve (A) Valve (B)

* The master work should be the same as the tested work without leakage

Pressure Source

Fig.2.4 Basic Circuit of Differential Pressure Type Leak Test

Page 15: Fukuda Technical Manual en 00

14

Fig. 2.5 Basic Pneumatic Circuit of Differential Pressure Type Air Leak Tester

2.1.2 Basic Operation The FUKUDA Air Leak Testers perform the following 4 processes sequentially and makes an

existence/nonexistence judgment of the leakage. 1) Pressurizing Process (Fig. 2.6)

The compressed air is adjusted to the test pressure and is charged into both the tested work and the master work through the solenoid valves SV1, SV2 and SV3. 2) Balancing Process (Fig. 2.7)

The solenoid valves SV2 and SV3 are closed to seal the charged air in the tested work and the master work, and allow the pressure to stabilize. If a large leak is detected, the tester makes an “NG” judgment, and the test proceeds to the exhaust process.

3) Detection Process (Fig. 2.8) The presence or absence of a trace leak is detected in this process. If a leak is detected, the

tester makes an “NG” judgment, and the test proceeds to the exhaust process. 4) Exhaust Process (Fig. 2.9)

After an OK or NG judgment is made, the pressures in the master and the tested work is exhausted into the atmosphere.

Air Pressure Source

Filter

Pressure Gauge

3 Way Valve, SV1 Solenoid Valve, SV2

Solenoid Valve, SV3

Differential Pressure Sensor

Calibrator

Filter

Valve, V2

Valve, V1

Start OK NG Controller Amplifier

マスター

◎ ***Pa

減圧弁 電磁弁

 SV1

電磁弁

  SV2

電磁弁

 SV3

ワーク

差圧センサー メーター

圧力計

マスター

メーター

◎ ***Pa

減圧弁 電磁弁

 SV1

電磁弁

 SV2

電磁弁

 SV3

ワーク

差圧センサー

圧力計

Work Work

Master Master

Meter Meter

Differential Pressure Sensor Differential Pressure Sensor Pressure Reducing Valve

SV1

SV2

SV3

SV1

SV2

SV3

Pressure Reducing Valve

Fig. 2.6 Pressurizing Process Fig. 2.7 Balancing Process

Page 16: Fukuda Technical Manual en 00

15

The differential pressure sensor which converts a pressure value into an electrical signal is the

heart of the FUKUDA Air Leak Testers. Therefore, FUKUDA doesn’t rely on other manufacturers for this sensor, and has originally developed and manufactures it by ourselves.

2.1.3 Calculation Equation of Leak Volume

(1) Calculation Equation of Leak Test

VL : Leak Volume (ml/min)

T : Detection Time (sec) ⊿P : Resulting Differential Pressure (Pa) Po : Atmospheric Pressure (101325Pa) Vw : Inside Volume of Tested Work (ml) [including the inside volume of the leak tester and its piping] Vs : Inside Volume of Master Work (ml) [including the inside volume of the leak tester and its piping]

P : Test Pressure (Pa)

: Sensor Coefficient (0.4×10-5 ml/Pa)

This calculation equation is available for download at website of FUKUDA.

(2) Calculation Equation of Internal Volume of the Work

Vw : Volume to be Calculated (ml)

⊿VL : Volume Change in Calibrator (ml)

Po : Atmospheric Pressure (101325Pa)

⊿P : Resulting Differential Pressure (Pa)

Constant : 0.4 (The value calculated from the sensor coefficient)

⊿P

⊿V

Vw = - Constant ⊿VL×Po

⊿P

Fig. 2.8 Detection Process Fig. 2.9 Exhaust Process

マスター

メーター

◎ 500Pa

減圧弁 電磁弁

 SV1

電磁弁

 SV2

電磁弁

 SV3

ワーク

差圧センサー

圧力計

リーク

マスター

◎メーター

***Pa

減圧弁 電磁弁

 SV1

電磁弁  SV2

電磁弁  SV3

ワーク

差圧センサー

圧力計

Differential Pressure Sensor Differential Pressure Sensor

SV3

SV3

Work

Master

Work

Leak

SV2

SV2

SV1 SV1

Pressure Reducing Valve

Pressure Reducing Valve

⊿P

T・Po { VW+ ( 1 + ) (P+Po) }⊿P ⊿V Vw

Vs VL =

Page 17: Fukuda Technical Manual en 00

16

2.1.4 Applied Measuring Method

(1)Leak Side Measurement

This method is applied when the measurement becomes unstable due to the substance within the tested work.

(2) Sealed Work Measurement with Pressurized Tank This method is applied when the tested work cannot be pressurized.

(3) Measurement with Differential Pressure Sensor of 10 kPa Scale

This method is applied for slightly rough leak tests.

Chamber

Work

Examples: Activated Carbon, Diaphragm, Oil Seal, and Hollow Fiber

Fig. 2.10 Basic Circuit of Leak Side Measurement

Tank

Work

Examples: Semiconductor, Camera, Crystal, SAW Filter, Chemical Bag, Food Package, and Molded Parts.

Fig. 2.11 Basic Circuit of Sealed Work Measurement with Pressurized Tank

Page 18: Fukuda Technical Manual en 00

17

2.1.5 For Performance Enhancement

(1) Mastering Measurement and Fitting Measurement In a leak test the gas in the work generates heat due to adiabatic compression during

pressurization of the work and this causes a momentary temperature rise in the gas within the work. Therefore, a temperature difference between the gas inside the work and the outside air occurs and the heat of the inside gas dissipates through the wall of the work over a long time. The heat dissipation continues until the temperature inside the work becomes equal to the ambient temperature. The time required for the heat dissipation depends on the material and shape of the work, the equipment used for clamp sealing, etc. For example, in cases of aluminum work and an iron work with a same shape, the aluminum work dissipates its heat quickly, because the heat conductivity of aluminum is higher than that of iron.

Since the heat conduc tivity of metals is higher than that of air, the work with internal structure and larger internal surface area dissipates its heat quickly. However, the heat remains inside and it dissipates slowly when the outside structure of the work and its internal structure have no contact. For example, a sintered metal filter has small metal contact area between its case and the sintered metal part. The speed of heat dissipation depends on works like this. In a leak test, a leak is detected according to the pressure change in the work which is pressurized at a test pressure. Therefore, it is difficult to perform accurate leak detection when the state of the gas within the work changes due to its heat dissipation. The master- less method and the fitting method that are the measuring methods employed in our

leak tester enable measurements with correction for the influence of the heat.

Work with a few contact points of the internal structure

Work with many contact points of the internal structure Hollow Work

Page 19: Fukuda Technical Manual en 00

18

(1)-1 Mastering Measurement (PAT. No.: 3461973, 3502687)

When mastering measurement is used:

Mastering Data (Noise) Subtract the leak area from the measured data area and store the resultant value as the mastering data.

Leak (equals to leak volume)

θActual Measured Data

Characteristics ・ Reduced pressurization time and detection time when compared to the existing leak test

methods. (A detection time less than 5 sec can be set) ・ Repeatability of measurements depends on the mastering data

Time sec DET

2sec

M.DET

Same Angle

Resulting D

ifferential Pressure

Pa

When the normal measurement is used with the work to which the mastering process is performed:

Time sec DET

θMastering Data

Leak

When a leak exists in the mass production work;

Time sec DET

θ′ Mastering Data

Leak

Fig. 2.12 Mastering Measurement

Fig. 2.13 Normal Measurement

Fig. 2.14 Measurement with the Work with Leak

Stable Zone (Linearity)

Actual Measured Data

Resulting D

ifferential Pressure

Pa

Resulting D

ifferential Pressure

Pa

Actual Measured Data

Page 20: Fukuda Technical Manual en 00

19

(1)-2 Fitting Measuring (PAT. PEND.) The heat dissipation of the work is realized as follows.

In Fig. 2.15, the temperature within the work increases when the work is pressurized to the test pressure through the work port. Then the pressurization valve is closed and the pressure change within the work is detected using the differential pressure sensor. This differential pressure at the start of the detection process becomes b1 (a pressure at a different temperature from b2). After that the internal temperature of the work approaches the ambient temperature at the heat dissipation rate which is determined by the heat conductivity, C, of the work and the inside temperature of the work becomes equal to the ambient temperature (b2) over a long time. The pressure b which is the convergent point of the differential pressure due to heat dissipation

is obtained by the equation; b = |b1 – b2|. The heat dissipation curves are as shown in Fig. 2.17 when the value b varies in actual use environments.

The curves are also as shown in Fig. 2.18 when the heat conductivity varies in actual use environments.

In fitting measurement, these 2 kinds of pressure variation can be corrected at the same time. These corrections are made using a calculation method to determine constants (a, b, c) by fitting the pressure curve to the equation of heat dissipation characteristic.

y = ax + b( 1-ecx ) ・・・・・・・・・・・・・・(2.2) The constants a, b and c are calculated from the differential data at 0.1 second intervals in the detection process, thus they approximate equation 2.2 with the least error. The calculated value (of a leak) is displayed as the measurement result. (FittingSet4) In some cases, however, the error due to the approximation calculation may become large when the measurement condition is unstable and these 2 kinds of pressure variations are small. In such cases, measure the values of b and c with a long detection time and obtain their accurate values. The performance will be improved by substituting them into the equation 2.2 as constants and calculating the value with approximation in the OK/NG measurement of an actual work. (FittingSet1) The FittingSet1 is a correction function similar to that in the master-less measurement. It is also conceivable that the perfo rmance will be improved by only fixing the heat conductivity C of the work in some environments. It enhances the following ability to environmental variation. (FittingSet0)

Heat Conductivity of Work ; c Pressurization

Leak Detection Port (Work Port)

Leak from Work ; a

Inside Temperature of Work b1

Ambient Temperature of Work b2 b1, b2, and b: Constants of pressure converted from temperature

Fig. 2.15 Temperature Inside of Work Fig. 2.16 Leak Characteristic Curve

Leak Curve with a leak present

Curve without Leak

c

b

Detection Process Time

Differential Pressure

Fig. 2.17 When Resultant Differential Pressure Varies Due to Heat Dissipation

Fig. 2.18 When Heat Conductivity Varies

Differential Pressure

Differential Pressure

Detection Process Time Detection Process Time

Page 21: Fukuda Technical Manual en 00

20

FittingSet2 is a measuring mode in which fixed corrections of leak values are added to FittinSet1. FittingSet3 is similar to the existing leak testers, such as the work-master comparison method, and corresponds to the measuring method which generates the heat dissipation characteristics on both sides of the differential pressure sensor, the tested work and the master work, and counterbalances them with each other. (2) APU (Super Electropneumatic Regulator) (PAT No.: 2618368)

An APU is an electropneumatic regulator which controls an air pressure and a flow rate by changing its input voltage. The APU realizes control reproducibility of ±0.1% by employing the differential type nozzle flapper system for the precise electropneumatic proportional valve used in it. It can stably provide pressure with a large flow rate even at very low pressure or negative pressure, this has been considered difficult to realize, and it can also perform the control of a coupled pressure from negative to positive pressure. Its feedback sensor is attached externally, so the pressure control at an arbitrary point can be performed. The differential pressure between both works can be set to a constant value by using a differential pressure sensor. It is also capable of performing a constant flow rate control.

・ Stable pressure control is possible at an arbitrary setting. ・ Measurements with stable pressurization can be performed as the pressure switches

automatically by inputting the settings of the lead pressure and the test pressure for the air leak tester. (Available only for FL-600, FL-601 and FM-1061)

・ Rapid pressurization is possible thanks to the feedback sensor.

(3) (Turbo-Type) Air Leak Tester (PAT No.: 173234)

The rapid stabilizing air leak tester stabilizes the internal pressures of the works after charging the preset compressed air into the works. This air leak tester is aimed to significantly reduce the pressurization time by these operations. In addition to the test pressure source, the tester uses a compressed air source whose pressure is higher than the test pressure. The works are pressurized with the lead pressure during the pressurizing process and then the temperature rise which occurs during pressurization of air is rapidly stabilized by switching the pressure to the test pressure. The pressurization time is reduced when compared to the case in which the pressurized air is charged at one test pressure. In addition to this, the influence of air temperature due to adiabatic compression and the change in the inside volume of the tested work are cancelled by the pressurizing and depressurizing operations. As a result, the pressurized air rapidly stabilizes and this can reduce the pressurization stabilizing time to one-half or one-third when compared to the normal time. However, the setting valves for the lead pressure and the pressurization time must be determined based on preliminary investigations for the condition of the test work. (For FL-3XXX Series)

Pressure/Tem

perature

Internal Temperature Change Test Pressure

Pressure Change Due to Internal Temperature

Sealed Sealed

Lead Pressure

Stable Zone

Test Pressure

Stable Zone

Pressure/Tem

perature

Pressurization Time Pressurization Time

Internal Temperature Stabilizes Rapidly

Fig. 2.20 Pressure/Temperature Stabilizing Curve of Turbo-type Air Leal Tester

Page 22: Fukuda Technical Manual en 00

21

(4) Exhaust Bypass The exhaust bypass unit is used when a foreign material such as water, oil or swarf attaches to

or mixes into the works. The foreign material may contaminate the inside of the air leak tester because the pressurized air passes through the tester when the air is exhausted. If the foreign material mixes into the inside of the air leak tester, it may cause the malfunction of the internal solenoid valve or the failure of the differential pressure sensor. Therefore, these troubles shall be prevented by letting the pressurized air within the works exhaust through the exhaust bypass, without passing through the inside of the tester.

・ The use of the exhaust bypass is recommended when foreign materials attach to or mixes into the works.

・ By the use of the exhaust bypass unit, the causes which may lead to the failure of the air leak tester will be reduced and stable measurement results can be obtained.

(5) Pressurization & Exhaust Bypass The pressurization & exhaust bypass unit is the product to which a bypass function for

pressurization is added in addition to the exhaust bypass function. Since the pressurization flow rate increases due to the additional pressurization bypass, effects such as the reduction of pressurizing process time and the improvement of measurement stability can be expected for large works with an internal volume of 1000 ml or more.

・ By the use of the pressurization bypass unit, the reduction of test time and the improvement of measurement stability for large works can be achieved.

・ With the exhaust bypass function, the causes which may lead to the failure of the air leak tester such as foreign materials attached to the work can be reduced (see “Exhaust Bypass”).

(6) LPU-300

This is the air leak tester dedicated for the works with a small internal volume (30 ml or less). It has high detection sensitivity because its capacity is very small (0.7 ml).

Page 23: Fukuda Technical Manual en 00

22

VwTPo

P=VL

2. 2 DIRECT PRESSURE DETECTION AIR LEAK TEST

2.2.1 Operating Principle In this test method, the test pressure is charged only to the tested work and a leak is detected by measuring the pressure drop (or increase) in the work after a given time. This method is suitable for the measurement of a large leak but is not suitable for a trace leak, because the influences of temperature, deformation, etc. directly result in a pressure change. 2.2.2 Basic Operation (1) Pressurizing: The test pressure is charged into the test work through the solenoid valves SV1 and SV2 (2) Balancing: The solenoid valve SV2 is closed and allow the test pressure charged into the work to stabilize. (3) Detection: Detect the leak from the work. At the start of the detection process, the zero

correction circuit is automatically activated and the detected pressure value starts from zero at all times.

2.2.3 Calculation Equation of Leak Volume

VL : Volume Leakage to Atmosphere (ml/sec)

T : Detection Time (sec)

⊿P : Pressure Drop during Detection Time (Pa)

Po : Atmospheric Pressure (101325 x 104Pa)

Vw : Volume of Tested Work (ml)

startOKNG

Work

Fig. 21 Basic Circuit of Direct Pressure Detection

SV1 SV2

Page 24: Fukuda Technical Manual en 00

23

2.3 HYDROGEN GAS DETECTION LEAK TEST 2.3.1 Operating Principle In this test method, hydrogen molecules which infiltrate through a ceramic filter are detected using a semiconductor concentration sensor. Hydrogen molecules will diffuse from high to low concentration. By taking advantage of this nature of hydrogen molecules, the leaked hydrogen molecules which diffuse within the sensor are measured. The advantages of using hydrogen gas ・ Cheap gas price ・ Easy to leak due to low viscosity and less persistent ・ The lightest molecule. ・ Easy on the environment (no pollution) ・ Nonflammable and safe because the gas is used as a mixed gas of 5% hydrogen and 95%

nitrogen. ・ Non-corrosive inert gas (harmless to humans) ・ Low background level in the atmosphere (0.5ppm)

As described above, there are many advantages when hydrogen is used.

2.3.2 Basic Operation

① Replace hydrogen gas for air. ② Perform a leak test.

2.3.3 Calculation Equation of Leak Volume If the leaking hydrogen gas is enclosed within a closed chamber, the concentration of hydrogen

within the chamber increases consistently. This phenomenon is used to increase sensitivity in the deposition method. The mean concentration of hydrogen within the chamber is describes by the following equation.

C : Hydrogen Concentration within Chamber (ppm)

Ctracer : Hydrogen Concentration of Tracer Gas (ppm)

F Leak : Tracer Gas Leak Volume (atm ml/sec)

tAce : Deposition Time (sec)

Vchamber : Internal Volume of Chamber (ml)

The deposition time can be reduced in two ways. That is, to increase the test pressure or to

decrease the chamber internal volume. For example, if the charging pressure is tripled, the deposition time is reduced to one tenth. This is effective for the leak of a viscous gas and most gases have some viscosity.

The deposit time when the hydrogen concentration of the sample gas reaches 5ppm is shown in Table 2.1. The values in the table are reduced by half if the alarm level is set to 2.5ppm.

C=CtrancerφLeak

Vchamber・tAce

Page 25: Fukuda Technical Manual en 00

24

Table 2.1 Deposit Time Leak Volume

10-2 atm ml/sec

10-3 atm ml/sec

10-4 atm ml/sec

10-5 atm ml/sec

10-6 atm ml/sec

1ml 0sec 0.1sec 1sec 10sec 100sec

10ml 0.1sec 1sec 10sec 100sec 16min

100ml 1sec 10sec 100sec 16min 2.8h

1L 10sec 100sec 16min 2.8h 28h

Work V

olume

10L 100sec 16min 2.8h 28h 278h

2.3.4 Applied Measurement Method (1) Diffusion Scanning

In this method, a porous plastic plate or cloth-like material is applied to the leak point and retain the leaked gas, and then the leak can be detected by scanning along the surface of the plate or material with a sensor.

Hydrogen molecules will diffuse from high to low concentration. By taking advantage of this the leaked molecules which diffuse can be measured with the sensor. (2) Sniffer (Suction)

The sensor has a suction function and can ‘suck ‘ in air or gas at a constant flow rate. The leak volume from the leak point can be calculated from the suction volume, exhaust volume and detected gas concentration. In this method, the gas that leaks from a leak point at a narrow range (10 ml or so) is inhaled.

Fig. 2.22 Sniffer Method

C : Hydrogen Concentration of Sample (ppm)

CTracer : Hydrogen Concentration of Tracer Gas (ppm)

Fleak : Tracer Gas Leak Volume (atm ml/sec)

Fsample : Suction Sample Volume (atm ml/sec)

Pipe Filled with Tracer Gas Air Supply Port

Exhaust Flow for Sensor including Leaked Tracer Gas

The tracer gas leaks into the sniffer channel within the cover.

C=CtrancerφLeak

Vchamber

Page 26: Fukuda Technical Manual en 00

25

Example: If the units of Fleak and Fsample are the same, other units can be used. When the mixed gas is 5% hydrogen and the sample flow rate is 1 atm ml/sec, the equation above is expressed as follows. C = 5000 Fleak Assume that a normal sample flow rate is 1 atm ml/sec The suction probe used is the AP-55. (3) Deposition Test

In this method, the work is enclosed by a chamber or hood when its volume or surface area is large and the leaked gas is measured by depositing the gas inside the chamber or hood. With this method, even a trace leak can be measured by taking plenty of time and increasing the concentration of the leaked gas.

There are 2 measuring methods for the deposition test. 3-1 Natural Elevation Deposition

The leaked hydrogen gas diffuses and elevates within the chamber or hood and it accumulates at the upper portion. Though the concentration of hydrogen gas depends on the sampling position even with a same leak rate, this method is used when the leak should be detected quickly. Since the shape of the chamber is an important factor, its effect must be checked with experiments.

3-2 Forced Circulation The air and leaked gas in the chamber are forcibly circulated. This method is used when the same detection result should be obtained if a same leak exists at any position. Therefore, the deposition time is longer than that in the natural elevation method. If the circulation is insufficient, the use of a diaphragm pump or piston pump with an AC motor is recommended (these pumps are also used for fish tanks). Mixture can be improved by attaching a circulation fan. But, in general, a fan can only generate a very low pressure. If the diameter of its flow path is very small (30 mm or less), it is necessary to use a pump instead. A circulation system must be designed to circulate the air within the chamber at least 10 times during the deposition time. The mixing ability of the system must be fully checked by placing a standard leak device at different points on the tested work.

Work with Tracer Gas

Leaked Gas

Chamber or Hood

Fig. 2.23 Natural Elevation Deposition Method

Fan or Pump for Air Circulation

Work with Tracer Gas

Chamber or Hood

Fig. 2.24 Forced Circulation

Minimize the Dead Volume (Air)

Page 27: Fukuda Technical Manual en 00

26

3-3 On Decompression Chamber Method (An Evolved Version of Decomposition Methods) The sensor of ‘Sensistor Technologies’ leak detector which uses a mixed gas of 5% hydrogen

and 95% nitrogen as a tracer gas is necessary to conduct this measurement at atmospheric pressure. In automatic leak tests, some works can not be tested in practice because the space within the chamber has large influences on the detection capability and cycle time. In this method, the apparent space can be reduced and the pressure at the sampling position of the automatic probe can be set to atmospheric pressure by reducing the pressure within the camber to 1 kPaabs and by sucking in a certain amount of air and exhausting the same amount. As a result of this, the detection capability can be enhanced and the cycle time can also be shortened.

Exhaust Air

Detector

Work

Chamber

Charging and Exhaust System of Tracer Gas

Automatic Probe

Exhaust Compressed Air

Normal Gas

Decompression Pump

Sampling Pump

Sampling Hose

Air

Page 28: Fukuda Technical Manual en 00

27

2.4 Helium Gas Detection Leak Test 2.4.1 Operating Principle

In this method, a trace amount of helium gas which is leaked from the tested work is removed and placed in a detector (mass spectrometer), to detect the gas. The helium gas is ionized by the electron beam from the filament within the ion chamber of the analyzing tube. The ions are accelerated and move out through a slit and then pass through the magnetic field generated by the analyzer. Since the circular trajectories of the ions depend on their mass, the collector can catch the helium ions. There are several types of analyzer. The typical types include the magnetic field modification type (magnetic sector), omegatron type, and quadrupole type.

2.4.2 Basic Operation (1) Suction Method (Sniffer Method) 1-1 Charge the pressurized helium gas into the work. (Fig. 2.26) 1-2 Collect the helium gas which is leaked from the work with the a helium detector. (Fig. 2.27) 1-3 Exhaust the gas with the vacuum pump. (Fig. 2.28)

Fig. 2.25 Principle of Helium Gas Detection Leak Test

×

Slit Collector

○Molecule

A

Light ion

Heavy ion

Helium ion

Analyzer (generates magnetic field)

Heボンベ

Work

H eディテクター

VP

Heボンベ

Work

H eディテクター

VP

Fig. 2.27 Detection with Helium Detector Fig. 2.26 Pressurization with Helium Gas

He Gas Cylinder He Gas Cylinder

He Detector He Detector

Heボンベ

Work

H eディテクター

VP

Fig. 2.28 Exhaust

He Gas Cylinder

He Detector

Page 29: Fukuda Technical Manual en 00

28

(2) Vacuum Vessel Method (Bell Jar Method) 2-1 Depressurize the work and chamber. (Fig. 2.29)

Depressurize the work so that the concentration of helium gas is not decreased by atmospheric pressure. 2-2 Pressurize with helium gas. (Fig. 2.30)

Discharge the pressurized helium gas into the work. 2-3 Conduct a leak test. (Fig. 2.31)

Detect and calculate with the detector. 2-4 Clean the helium gas with nitrogen gas. (Fig. 2.32)

Exhaust the gases with the vacuum pump. 2-5 Open the work and chamber to atmospheric pressure. (Fig. 2.33)

Fig. 2.29 Depressurization of Work Fig. 2.30 Pressurization with Helium Gas

Fig. 2.31 Leak Test Fig. 2.32 Cleaning of Helium Gas

N2

ボンベ

Heボンベ

VP

Work Heディテクター

VP

N2

ボンベ

Heボンベ

VP

Work Heディテクター

VP

N2

ボンベ

Heボンベ

VP

Work Heディテクター

VP

N2

ボンベ

Heボンベ

VP

Work Heディテクター

VP

N2

ボンベ

Heボンベ

VP

Work Heディテクター

VP

He Gas Cylinder

N2 Gas Cylinder

He Gas Cylinder

N2 Gas Cylinder

He Gas Cylinder

N2 Gas Cylinder

He Gas Cylinder

N2 Gas Cylinder

He Gas Cylinder

N2 Gas Cylinder

He Detector He Detector

He Detector He Detector

He Detector

Fig. 2.33 Exhaust

Page 30: Fukuda Technical Manual en 00

29

2.4.3 Calculation Equation of Leak Volume Measurement of Detection Sensitivity (a) While opening the valve for the calibration leak, record the elapsed time and the meter reading

of the helium leak detector, and take the reading when it is stabilized as X2. In this step, the reading can be obtained from either the chart of the recorder or the time history of the reading.

(b) Close the valve of the calibration leak and take the reading when it is stabilized as X1. If the

background level varies, measure the maximum value (X1max) and minimum value (X1min) and take their average as the background X1.

(c) Calculate the detection sensitivity (S) using the equation below.

Qc : Leak volume from the calibration leak after compensated for the temperature at (Pa・m3/sec) X1 : Reading of the helium leak detector with the background level. (Output) X2 : Reading of the helium leak detector for the calibration leak. (Output)

(d) Calculate the minimum detectable leak volume (L) using the equation below.

(e) Measurement of time constant Measurement (a) Charge helium gas into the tested work. In this step, it is preferable to replace with helium gas by depressurizing

the work with a vacuum pump in advance. (b) Measure the background level and record it. (c) Record the reading on the helium gas detector when at least, the time constant (t) has elapsed, after the start of

the helium gas charge. (d) Calculate the leak volume (QT) using the equation below.

X1 = (X1max+X1min )

2

S = [1/(Pa・m3/sec)] (X2-X1)

Qc

L= S 2 (X1max-X1min ) =

2Qc(X1max-X1min )

X2-X1 (Pa・m3/sec)

X2-X

1

37% of (X2-X1) τ

X2 X

1

Background Background Level

Calibration Leak Valve, Close Calibration Leak Valve, Open

Time

Output

Fig. 2.34 How to Obtain Time Constant t

Page 31: Fukuda Technical Manual en 00

30

Xp : Value of the helium leak detector at the test

X1 : Value of the helium leak detector with the background level (Output)

S : Detection sensitivity

Open Discharge the helium gas within the work and open the vacuum vessel (bell jar) to the atmosphere.

Post-processing Perform the necessary post-processing.

Judgment Judge the result according to the inspection standard or specification. Use 1 x 10-7Pa・m3/sec unless other

wise noted. 2.4.4 Applied Measurement Method

Bombing Method (for Measurement of Sealed Work) If the work can not be pressurized, charge helium gas into it beforehand and then measure its

leak with the detector. With a defective work, the

helium gas infiltrates into the work. And then the helium gas, which had infiltrated into the defective work, is detected when it leaks to the chamber. In this method, the waiting time

between the completion of gas charge and the measurement affects the accuracy of measurement. Therefore, the waiting time must be controlled.

By combining the differential pressure detection leak test and the helium gas detection leak test, an ideal measurement environment can be configured.

QT = (XP-X1)

S

良品不良品

Heガス

Heディテクター

He

He

不良品

Fig. 2.35 Bombing Method

He Gas

Defective Work Good Work

Defective Work He Detector

Page 32: Fukuda Technical Manual en 00

31

2.5 FLOW RATE LEAK TEST

2.5.1 Operating Principle There are two major flow meter types for air (gas), a volume flow meter and mass flow meter. The volume flow meter includes a differential pressure flow meter (laminar flow meter), variable area flow meter, and volumetric flow meter, and the mass flow meter includes a thermal mass flow meter (mass flow meter) and vortex flow meter. The volume flow rate is expressed by volumetric displacement per unit time (l/min) and the mass flow rate is expressed by mass displacement per unit time (g/min). However, they are generally expressed by a volume flow rate unit at a certain condition where reference pressure and temperature are specified in pursuant to typical flow rate measurements nowadays. Among volume flow meters, the laminar flow meter has a characteristic that the pressure loss ⊿P is proportional to the flow rate Q when the flow in the pipe is a laminar flow. 128µL 32µL

pd4 d2

⊿P : Pressure Difference µ : Viscosity Coefficient L : Length of Pipe d : Diameter of Pipe Q : Flow Rate U : Average Flow Velocity of Fluid

A laminar element has a structure such that many capillary tubes are bundled along a flow path, and fluid passes through each capillary tube. Since the d of it is small, the Reynolds number is extremely small. The pressure difference ⊿P (pressure drop) of an actual laminar element is expressed as the sum of the pressure drops, ⊿P1 and ⊿P2, when the flow in it is assumed as a Poiseuille flow over the entire length. 32µL V2

d2 2g

Kr: Loss Coefficient at the End of Capillary Tube

In order to keep the linearity of the pressure difference ⊿P against the flow rate, the value of the following equation must be small enough. ⊿P1 k d

⊿P2 64 L

The FUKUDA Laminar Flow meter is designed to keep the linearity up to twice the rated flow.

The typical flow rate sensor used in a mass flow meter is a thermal mass flow rate sensor. On the capillary tubes that work as a sensor, resistive elements with a high resistance temperature coefficient are attached at the upstream (Rus) and downstream (Rds) of the tube. When electric currents are applied to the elements, they generate heats. The temperatures of these elements are same when there is no flow in the pipe. If the fluid begins to flow in the pipe under this condition, the upstream element is deprived of heat by the fluid and the down stream element receives the heat. A temperature difference ⊿T occurs between the upstream element and downstream element, and the temperature difference ⊿T has a functional relationship with the mass flow rate of the fluid. Therefore, the mass flow rate is measured by the changes of the resistances of electric signals, and then by amplifying and correcting them.

⊿P= Q= U

Fig. 2.36 Principle of Differential Pressure Flow meter

L

Pressure Difference ⊿P

⊿P=⊿P1+⊿P2= U+Kr ⊿P1 ⊿P2

Capillary Tube

⊿P

Fig. 2.37 Principle of Laminar Flow meter = Re =

Fig. 2.38 Principle of Thermal Mass Flow meter

Rus Rds

Flow

Page 33: Fukuda Technical Manual en 00

32

2.5.2 Basic Operation (1) Flow meter (at Atmospheric Pressure)

In the atmospheric pressure method, the work is positioned at the air pressure source side and the laminar at the atmospheric pressure side. The test pressure is supplied to the work and the leaked air passes through the laminar and it is exhausted into the atmosphere. The flow rate can be obtained by multiplying the pressure difference at the laminar by the laminar coefficient.The same laminar coefficient can be used when the test pressure is changed.

(2) Flow meter (at Line Pressure) In the line pressure method, the laminar is positioned between the air pressure source side and the work at the atmospheric pressure side. The test pressure passes through the laminar and is supplied to the work and is exhausted into the atmosphere. The flow rate is calculated by multiplying the pressure difference at the laminar by the laminar coefficient. The laminar coefficient in this method depends on the test pressure.

(1) Supply the test pressure. (2) Wait until the flow rate stabilizes. (3) Measure the flow rate. (4) Exhaust the pressure.

2.5.3 Calculation Equation of Leak Volume

Q = ⊿P ×k

Q : Flow Rate (ml/min)

⊿P : Pressure Difference (Pa)

k : Laminar Coefficient

Fig. 2.40 Mass Flow meter at Line Pressure

Work

Work

Fig. 2.39 Mass Flow meter at Atmospheric Pressure

Page 34: Fukuda Technical Manual en 00

33

2.5.4 Applied Measurement Method The basic methods of gas flow rate have been described in the BASIC OPERATION section above, the methods to improve the measurement performance are stated here.

When a flow rate measurement is conducted using a volume flow meter (laminar flow meter, * hereinafter referred to as laminar) or mass flow meter, the flow velocity in the pipe is extremely low if the estimated effective sectional area is extremely small. In such a case, it is necessary to take a substantially long measurement time to perform a proper flow rate measurement, because the flow velocity in the pipe is extremely low. There is a method to improve the measurement performance by using a pressurization bypass to reduce the measurement time as short as possible.

As shown in Fig. 2.41, by using a method to achieve the specified pressure in the work with holes whose flow rate is very low by the use of a circuit without the work, the area with a slow flow rate is practically limited to the holes of the work and this leads to the reduction of measurement time. When a flow rate is measured with a laminar flow meter or mass flow meter, a regulator is used to adjust a pressure to the specified pressure. Since the regulator adjusts the pressure while supplying a flow to compensate the pressure loss of the work in the secondary side, a normal regulator can not maintain the specified pressure because of its structure. When the specified pressure can not be maintained, this causes the failure to obtain stable flow rate measurement results. In such a case, the above mentioned problem can be solved by the use of an electropneumatic regulator (we call this an APU) which always supplies the specified pressure even if the pressure loss of the work varies. (The APU theory will be explained in Chapter 3.)

Work

Fig 2.41 Flow Rate Measurement Using Pressurization Bypass

Page 35: Fukuda Technical Manual en 00

34

Peripheral Equipment

Chapter 3

Page 36: Fukuda Technical Manual en 00

35

3.1 EXHAUST BYPASS UNIT

OUTLINE The exhaust bypass unit is used when foreign materials such as water, oil, or swarf attaches to

or mixes into the works. Since the pressurized air charged within the pass through the air leak tester (here after referred to as ALT) when it is exhausted, these materials may enter into the inside of the ALT. If they enter into the inside of the ALT, it may cause the malfunction of the internal solenoid valve or the failure of the differential pressure sensor. Therefore, these troubles shall be prevented by using the exhaust bypass as it lets the pressurized air within the work directly exhaust through it, without passing through the inside of the ALT. ・ The use of the exhaust bypass is recommended when foreign materials attach to or mix

within the works. ・ By using the exhaust bypass unit, the causes which may lead to the failure of the ALT will

be reduced and stable measurement results can be obtained. There is also the pressurization & exhaust bypass unit which has the pressurizing bypass function in addition to the exhaust bypass function.

3.2 PRESSURIZATION & EXHAUST BYPASS UNIT

OUTLINE The pressur ization & exhaust bypass unit is that to

which the bypass function for pressurization is added in addition to the exhaust bypass function. Since the pressurization flow rate increases due to the additional pressurization bypass, effects such as the reduction of pressurizing process time and the improvement of measurement stability can be expected for large works with an internal volume of 1000 ml or more. ・ By using the pressurization bypass unit, the

reduction of test time and the improvement of measurement stability for large works can be achieved.

・ With the exhaust bypass function, the causes which may lead to the failure of the ALT such as foreign materials attached to the work can be reduced (see “Exhaust Bypass” section).

CBU-600

FE-20

Page 37: Fukuda Technical Manual en 00

36

3.3 FLOW MASTER

OUTLINE The flow master is a flow rate device which can maintain

a constant flow rate by applying a specified pressure. It has a glass capillary tube in the body. It generates a constant flow rate at a specified pressure by adjusting the length and diameter of the glass capillary tube. By using the flow master, the daily checks of the ALT’s function and accuracy with actual leaks can be conducted easily. The OK/NG judgment condition for leak testing can be set by using the flow master in combination with a work as a leak master. It is also capable of measuring the internal volume of a work when used with a differential pressure type ALT. ・ The traceability certificates can be issued.

3.4 CALIBRATOR

OUTLINE The calibrator changes its internal volume when the scale

dial is rotated. By using it in combination with a differential pressure type ALT, the sensitivity check of ALTs and the internal volume measurement of works can be conducted by measuring the resultant pressure difference due to a volume change. ・ By using the calibrator, the sensitivity check of

differential pressure type ALTs and the internal volume measurement of works can be conducted easily.

・ The traceability certificates can be issued. ・ There are 3 types: 0.1 ml, 1.0 ml and 5 ml.

FFM-100

Page 38: Fukuda Technical Manual en 00

37

3.5 SUPER ELECTROPNEUMATIC REGULATOR (APU)

OUTLINE

The super electropneumatic regulator (here after referred to as APU) is an electropneumatic regulator which controls an air pressure and a flow rate by changing its input voltage.

The APU realizes control reproducibility of ±0.1% by employing the differential type nozzle flapper system for the precise electropneumatic proportional valve used in it. It can stably provide pressure with a large flow rate even at very low or negative pressure, this has been considered difficult to realize, and it can also perform a control of a coupled pressure from a negative pressure to a positive pressure. Its feedback sensor is attached externally, so the pressure control at an arbitrary point can be performed. The APU is also capable of performing a constant flow rate control by using a differential pressure sensor to control the pressure difference at both sides of the work. ・ Stable pressure control is possible at an arbitrary setting by using the APU. ・ Measurements with stable pressurization can be performed as the pressure switches

automatically by inputting the settings of the lead pressure and test pressure for the ALT.

APU-120

APU-300

Page 39: Fukuda Technical Manual en 00

38

Information about

Equipment

Chapter 4

Page 40: Fukuda Technical Manual en 00

39

4.1 MANAGEMET OF WATER, OIL AND FOREIGN MATERIALS

When using the air leak testers, it is most important to correctly manage the water, oil and foreign materials mixed into the air used for testing. This is because they may have various harmful effects and cause a failure, if they enter into the inside of air leak testers.

As shown in Fig. 4.1, 66% of the failures caused are due to foreign materials. According to the survey on the detail of the main cause, 74% of mix is due to primary pressure and 26% due to works as shown in Fig. 4.2. Installation of a filter set (air filter and mist separator) just before an air leak tester is recommended to prevent the mix of foreign materials from the primary pressure.

Fig. 4.3 Filter Set Installed at Primary Pressure Side

Capacity of Filter Set to be Installed (Flow rate shall be selected according to the product)

Table 4.1 Filter Capacity

Filtration Size Ambient and Gas Temp. Secondary Side Oil Mist

Concentration.

Air Filter 5µm ‐5 – 60℃ -

Mist Separator 0.01µm ‐5 - 60℃ 0.1mg/m3 *1

*1: at compressor discharge concentration of oil mist of 30mg/m3 (ANR).

Failure Causes of Repaired Products

Breakdown of the Source of Foreign Materials

Fig. 4.1 Failure Causes Fig. 4.2 Details of the Source of Foreign

Air Filter Mainly removes moisture

Mist Separator

Removes oil mist.

Air Leak Tester Compressed Air Source

Request to Customers Regarding Long and Stable Use With a filter installed, the maintenance of the filter shall not be neglected, otherwise its performance will be lowered and foreign materials will enter into the air leak tester. Make sure to conduct “Drainage at daily inspection before operation” and “Exchange of filter element at every periodic inspection”

(For the proper exchange interval of the filter element in use, refer to the instruction manual.)

Repair Causes of Leak Testers Details of the Source

Mix of Foreign Materials Customer’s Reason Aging Others

Primary Pressure (Oil) Primary Pressure (Others) Work (Oil) Work (Others)

Page 41: Fukuda Technical Manual en 00

40

One of the failure causes of the products is the suction of foreign materials from the work during exhaust. Installing an external exhaust unit (charge external exhaust unit) is recommended to prevent the suction of foreign materials.

Fig. 4.4 Basic Circuit with External Exhaust Unit

4.2 MANAGEMENT OF SOURCE PRESSURE AND FLOW RATE Since test pressures of the leak test depend on each work, the management of the primary pressure is important. The variation of the source pressure and insufficient flow rate will result in reduced detection accuracy and longer tact time

Table 4.2 Primary Pressure Management Range for Each Measurement Range

Measurement

Range Control Range Primary Pressure

V -5 - -90kPa Direct supply from vacuum pump

UL 0 - 20kPa 50kPa

L 10 – 99kPa 100 – 200kPa

M 50 – 700kPa 100 – 900kPa

H 300 – 990MPa Set Pressure + 30%

H1 0.5 - 2.0MPa Set Pressure + 30%

H2 1.0 - 4.0MPa Set Pressure + 30%

As shown in Table 4.2, the management of the primary pressure is necessary. A sufficient discharge flow rate of the regulator must also be secured for the measurement of large works.

* As the management of the primary pressure and flow rate is the key point for the equipment, please contact our sales representative.

Leak Tester

External Exhaust Unit

Work

Page 42: Fukuda Technical Manual en 00

41

4.3 CLAMPING AND SEALING JIG Clamping and sealing are the absolute requirements for leak testing. Since shape and test

pressure depend on each work, appropriate clamping pressure and seal method are required for the work. 4.3.1 Clamping Clamping pressure is expressed as follows: Cylinder diameter x Air (Oil) pressure = (Seal area x test pressure + Crush force of rubber seal) x Factor for stabilization. The factor for the stabilization of clamping depends on the seal conditions such as surface roughness, rubber hardness, etc (the normal ratio of maximum value to minimum is 2). 4.3.2 Sealing Jig (1)Reduce the volume change The measurement becomes unstable if the internal volume changes when a force is applied to the work by clamping. The typical seal methods are O-ring sealing and urethane rubber sealing. We strongly recommend the O-ring searing in consideration of the characteristics of air leak testers. The urethane rubber sealing shall be used only when it is not applicable due to the shape of the work. With the use of a core cylinder, the volume change decreases during a test using a urethane rubber seal and this leads to the improvement of detection sensitivity and accuracy.

* The O-ring may come away during unclamping, so take an appropriate preventive measure such as dovetail groove.

Fig. 4.5 Seal Method (O-ring Sealing)

Fig. 4.6 Seal Method (Urethane Rubber Sealing)

Work

Sealing Jig O-ring

Work

Sealing Jig O-ring ゙

There is almost no volume change due to a metal contact condition between the work and sealing jig.

The rubber deforms when it contacts all over the work and the change of internal volume occurs easily, and this leads to a false measurement.

* By covering the side surface of the rubber, the deformation during sealing can be reduced.

O-ring Sealing

Urethane Rubber Sealing

Sealing Jig

Urethane Rubber Urethane Rubber Sealing Jig

Page 43: Fukuda Technical Manual en 00

42

Fig. 4.7 Seal Method (Urethane Rubber Sealing with Core Cylinder)

1) Select parts with little deformation due to pressurization ・ Select hard piping parts such as those of high-pressure type even if their specifications are sufficient in terms of pressure. ・ Do not use soft materials such as polyurethane. ・ Do not include devices with a diaphragm or spring in the measurement circuit. 2) In cases of deformable works

When a measurement is done for the work of a soft material such as resin, a structure to prevent deformation shall be employed. (2) Improvement of detection sensitivity With a leak, the change of pressure becomes large when the volume of a work is small. Leak detection of high sensitivity can be realized by reducing the volume of a work. 1) Utilization of core

2) Selection of piping In leak testing, the volume of the piping is added to the volume of a work. Therefore, use piping

as short as possible and increase the detection sensitivity.

Since the deformation of the seal is restrained and the change of internal volume decreases, the measurement accuracy is improved.

Urethane Rubber Seal with Core Cylinder

Urethane Rubber

Core

Urethane Rubber

Since the volume decreases due to the volume of the core, the detection sensitivity increases.

Urethane Rubber Urethane Rubber

Sealing Jig Sealing Jig

Core Cylinder Core Cylinder

Page 44: Fukuda Technical Manual en 00

43

4.4 CHANGE IN TEMPERATURE When the temperature change of the sealed work occurs, the pressure of it also changes and the

measurement becomes unstable. Sources of heat

・ Solenoid valves ・ Change of ambient temperature (air conditioning etc.)

Measures

・ Use pneumatic valves instead of solenoid valves. ・ Enclose the test equipment with a acrylic box ・ Wrap the piping with spiral tubes etc.