VTECH LEAK DETECTION

Leak Detection Theory and Practise

Comparison among leak testing techniques

Rev. 1 – 29 Mar. 06

VTECH_leak_detection -0- Rev. 2 – 23 nov. 2004

INDEX

1. INTRODUCTION ............................................................................ 3
2. LEAK TYPES.................................................................................. 4
3. LEAKAGE MEASUREMENT UNIT.................................................. 4
4. LEAK SIZES ................................................................................... 4
5. SOME USEFUL FORMULAE .......................................................... 6
5.1 Ideal Gas Law and Real Gases ........................................................................ 6
5.2 Gas Flow and Leak Rate ................................................................................. 7
5.2.1 Turbulent flow ........................................................................................ 7
5.2.2 Laminar flow ........................................................................................... 8
5.2.3 Molecular flow ........................................................................................ 9
5.2.4 Transitional flow ................................................................................... 10
5.3 Gas Dependency.......................................................................................... 11
5.3.1 Laminar flow ......................................................................................... 11
5.3.2 Molecular flow ...................................................................................... 11
5.4 Conversion from Mass Leak Rate to Volume Leak Rate ................................ 11
5.5 Diffusion of Gases ....................................................................................... 12
5.6 Permeation .................................................................................................. 12
6. LEAK TESTING IN REFRIGERATION INDUSTRY ....................... 13
7. LEAK TESTING METHODS.......................................................... 13
7.1 Ultrasonic Leak Detector.............................................................................. 15
7.2 Water Immersion Bubble Test Method ......................................................... 15
7.3 Soap Solution Bubble Test ........................................................................... 17
7.4 Pressure Decay Test .................................................................................... 17
7.5 Vacuum Decay Test or Pressure Rise Test.................................................... 19
7.6 Tracer Gas Leak Testing .............................................................................. 21
7.6.1 Sniffing ................................................................................................. 22
7.6.2 Accumulation leak testing..................................................................... 22
7.6.3 Vacuum chamber inside-out leak testing ............................................. 23
7.6.4 Outside-in leak testing ......................................................................... 23
7.6.5 Halogen leak detectors ......................................................................... 24
7.6.6 Inside-out helium sniffer detectors....................................................... 24
7.6.7 Outside-in helium spraying .................................................................. 26

VTECH_leak_detection -1- Rev. 1 – 29 Mar. 06

 7.6.8 Outside-in helium leak testing ............................................................. 26
7.6.9 Inside-out helium vacuum chamber leak testing .................................. 28
7.6.10 Inside-out hydrogen sniffer detectors ............................................... 29
8. HOW TO CHOOSE THE TEST METHOD ..................................... 30
9. CONCLUSION.............................................................................. 32
ANNEX A ........................................................................................... 33
ANNEX B ........................................................................................... 34
ANNEX C ........................................................................................... 36
ANNEX D ........................................................................................... 38
ANNEX E ........................................................................................... 39

VTECH_leak_detection -2- Rev. 1 – 29 Mar. 06

1. INTRODUCTION
Leak detection procedures are becoming popular in many areas of industrial production, playing key
roles in all processes where components or equipment must be leak-free to prevent external air
penetration or internal gas loss. Applications include vacuum chambers and pressure vessels, pumps,
cathode ray tubes production, refrigeration and HVAC systems, vacuum thermal isolation (e.g. dewars),
automotive manufacturing, chemical production, aerosol containers, electronic and semiconductor
industries, beverage cans, electron microscopes, pacemakers, etc. Enclosures with greater or smaller
tightness have to assure a satisfactory isolation between external atmosphere and inside over or under-
pressure. The words "leak" and "leakage" appear in the field of hermetically sealed units and do not
involve only vacuum technologists but also engineers working with high pressures. In some industries,
such as pharmaceutical, medical and food industries, the term “package integrity” or “seal integrity” are
used instead of leak. A leak can be defined as an unintended crack, hole or porosity in an enveloping wall
or joint which must contain or exclude different fluids and gases, allowing the escape or penetration of an
undesired medium. Critical leak spots in closed systems are usually connections, gaskets, welded and
brazed joints, defects in material, etc.

In spite of modern technologies, it is practically impossible to manufacture a sealed enclosure or system

that can be guaranteed to be leak-proof without first being tested. It is sufficed to say there are no
products without some leakage; they only have leaks smaller than a specified value. A leak test procedure
is usually a quality control step to assure device integrity, and should preferably be a one-time non-
destructive test, without impact on the environment and operators. Regarding quality demands in
different production processes, production engineers very often encounter specific standards for tightness
testing. The basic functions of leak detection are:

 determining if there is leakage or not

 measurement of leak rate
 leakage location

For solving the aforementioned problems, different treatments and techniques are known but among
them there is no universal method. Each test is suitable only for a selected leak rate or for fixed forms and
technologies.

In the most common uses of leak detection, explicit leak rate measurement is not required, but the system
must be able to recognize if the leak rate is above or below a specified level, so as to meet the required
tightness. To establish this reference limit, it is necessary to determine what the maximum acceptable
leak rate is consistent with the reasonable performance life of the product. This acceptance level is the
main parameter to be considered when selecting the proper testing method or combination of methods. In
this selection, several other factors have to be taken into account. In particular, system costs,
complexities, environmental impact, reliability, influence of external conditions, operator dependence
and user friendliness. Leak testing is a challenge. On one hand products must meet stricter leak rate
standard while, on the other, leak testing processes should be less costly and less dependent on operator
skill. To meet this dilemma, all aspects of the leak testing process have to be well understood. A lot of
documentation about leak detection, leak testing methods is available in industry literature. This article
presents leak types, their sizes and various leak detection techniques with a comparison of their
performance with particular attention to refrigerant leakages.

VTECH_leak_detection -3- Rev. 1 – 29 Mar. 06

2. LEAK TYPES
Different types of leak sources can be determined.

1. Leaks caused by defects in the enclosure or materials permitting gas diffusion and permeation
through the wall. For example, a thin wall of a plastic bottle becomes cracked microscopically at
high enough pressure difference, or in the canning industry if the score mark is too deep in the
ring on a pull-tab can top or a porous cast in a machine housing metallurgy, etc.
2. Leaks in newly manufactured products are most commonly due to imperfect joints, fixed or
demountable, or seals among the various assembled parts. The most commonly used junctions
are welded, brazed and soldered joints, glass-to-metal and ceramic-to-metal seals, O-rings and
other gaskets.
3. Leaks caused by internal sources of gas or vapor creating an internal pressure rise. In this case
they are not really leaks and are known as out-gassing or as a virtual leak in vacuum technology.
This is due to surface phenomenon on the chamber walls or materials inside the enclosure, dirty
materials, or presence of low vapor pressure so that substances, such as water and oils permeate
inside the chamber.

3. LEAKAGE MEASUREMENT UNIT

The main measurable effect of a leak is the flow of a medium through it, so the generally accepted
method to define a leak is the quantifying of this gas or fluid flow in specified conditions of temperature
and pressure difference. Consequently, leak rates can be defined in terms of flow, that is:

 mass in a specified time at reference condition of pressure and temperature, e.g. grams of a
refrigerant in years at reference pressure (3 g/y R134a at 500 kPa);
 number of bubbles of a specified diameter in a certain time period and pressure (5 bubbles of 2
mm per second at 400 kPa);
 volume in a specified time at standard condition of pressure (atmospheric pressure) and
temperature (0 ºC), e.g. standard cubic meter or centimeter per second (std cm3/s or std cc/s);
 pressure variation in a specified time period, e.g. 200 Pa pressure increment in 1 hour;
 referring to vacuum leak detection method, e.g.: 3 · 10-7 Pa · m3/s helium (using He-detection
method), or 3 · 10-6 mbar · l/s.

Clearly, all these different ways to identify a leak rate are equivalent and related to the kinetic theory of
gases. For convenience, a conversion table for the most commonly used units are reported in Annex A.

4. LEAK SIZES
Since the shapes of leaks (cracks, fissures, porosity, damages, etc) are very different, unknown and non-
uniform, it is impossible to measure their sizes with any geometrical dimension. Exceptions are very big
leakages and ideal or artificial leaks, as those used for calibration.
The leak rate does not only depend on the geometric dimensions (diameter, length) of the leak but also on
the physical properties of the gas (or the liquid), such as viscosity, relative molecular mass and the
pressure difference. For example, in the same environmental conditions helium flows through orifices 2.7
times faster than air. Measuring the same leak by various medium implies getting different results, so it
must always be specified which testing procedure was used in leakage definition. This procedure may be
implicitly defined considering the specific application, for refrigerating purposes it is the grams/year of
refrigerant at the plant working conditions. For industrial applications, the acceptable leak rates are

VTECH_leak_detection -4- Rev. 1 – 29 Mar. 06

generally in the range 10-9 ÷ 1 · mbar l/s, while the majority of products require less stringent
specifications, usually 10-6 ÷ 1 · mbar l/s. Only few products, such as very long life systems and high-
quality units, and/or safety constraints, such as in hazardous substance containment, require high
sensitivity leak test, up to or better than 10-9 mbar · l/s. The maximum acceptable leak rate for a given
product depends on the nature of product. Since the cost of leak detection (and manufacturing hermetic
envelopes) increases in inverse proportion to the leak rate, it follows that testing for unnecessary small
leaks causes unnecessary rise in production costs. Some examples of acceptable leaks in different
elements and systems are reported in the following table.

Max. permissible
System Remark
leakage

chemical process equipment 10-1 ÷ 1 mbar · l/s great process flows

automotive components (i.e. steering system) 10-3 ÷ 10-4 mbar · l/s liquid containment

beverage can bottom 10-5 ÷ 10-6 mbar · l/s retention of CO2

refrigerating application 10-4 ÷ 10-7 mbar · l/s HVACR plant life and environment impact rules

dynamic pumped vacuum system 10-5 ÷ 10-7 mbar · l/s permanent pumping

IC-package 10-7 ÷ 10-8 mbar · l/s

pacemaker 10-9 mbar · l/s long time implanted in body

closed vacuum elements 10-8 ÷ 10-10 mbar · l/s e.g. TV and X-ray tubes

Table 1 - Leak rate specification for various systems

A statistical analysis of leaking units, considering leakage size, frequency, leak types, is available in
industry literature. The following figure shows the typical distribution of leakage probability against the
size as observed in industrial products.

Fig. 1 - Typical distribution of leaks in industrial products

The two relative maximums shown in this curve refer to the two different leak modes: permeation or
orifice flow. It is not unusual that a part has more small leakages rather than a large one.

VTECH_leak_detection -5- Rev. 1 – 29 Mar. 06

5. SOME USEFUL FORMULAE
Before investigating the leak detection issue, it is worthwhile to recall some basic principles of the kinetic
theory of gas. The basic concepts are well known. The measurement units are expressed in the
international metric system (SI).

5.1 Ideal Gas Law and Real Gases

The pressure, volume and temperature of a perfect gas are related by the universal gas law:

p V  n  R  T

where p is the gas pressure (Pa or N/m2), V is the volume occupied by the gas (m3), T is the gas absolute
temperature (K), n is the gas quantity in number of moles, and R is the universal gas constant:

R  8.314472 J/ (mol · K)

Real gases can behave in a different way from that expected for an ideal gas. The state equation, i.e. the
pressure, volume, temperature relationship, for a non-ideal gas can be written as:

p V  Z  n  R  T

Where Z, the gas compressibility factor, describes the difference between ideal and actual behavior of a
gas. It should be written as Z(p, T), as it is a function of both pressure and temperature:

p V V an 1 an
Z    
n  R  T V  n  b R  T V 1  n  b R  T V
V
Z is always one for ideal gases and, in general, can be either greater or less than unity, depending on the
effect of either the size of the molecules (repulsive forces), modeled by parameter b, or the attractive
forces, modeled by parameter a. While this equation (being linear) is very simple to use, obtaining values
of Z is not. There are several models to compute Z factor, the most known are the Redlich-Kwong
equation and the Peng-Robinson equation; refer to Annex C for more details. For pressures close to
atmospheric, it is common to use a value of Z set to 1.0, and similarly, when the pressure is controlled
very close to a design set-point, it is usually adequate to use the fixed design compressibility. If, however,
the pressure and temperature can vary during normal operating conditions, then there can be a significant
variation in the compressibility factor for a gas. The following plot shows the compressibility factor of
some gases typically used, varying with pressure for a number of different temperatures.

Fig. 2 - Nitrogen and other gas compressibility factor typical behavior

VTECH_leak_detection -6- Rev. 1 – 29 Mar. 06

The deviation from ideal gas behavior tends to become particularly significant (or, equivalently, the
compressibility factor strays far from unity) near the critical point, or in the case of high pressure or low
temperature.

5.2 Gas Flow and Leak Rate

The flow of a gas through pipes, holes or orifices depends on gas properties, container wall’s
characteristics and pressure. When the particles in the gas are very close each other, typically at pressure
above 2 Pa, they collide more frequently with each other than with the container walls. If pressure
difference is established, gas flows like a fluid in the direction of the pressure gradient. This condition is
known as viscous flow and a distinction in two different kinds of flow, turbulent flow and laminar flow,
is usually carried out. When the gas particles are less close, each particle travels independently of other
molecules. They collide more frequently with the walls than with each other. The mean free path is larger
than the geometric dimension of the container. This flow condition is known as molecular flow. There is
not an abrupt transition from the laminar flow to the molecular flow, but there is an intermediate
situation, identified with transitional flow, where gradually the flow passes to the molecular condition.
These flow considerations can be applied also to leak analysis. However, in leak detection technology, in
most cases it is not possible to determine accurately the kind of flow. It must be estimated using formulae
for calculating it in various conditions of pressure, pipes, etc. All these formulae contain geometry factors
(length, diameters, etc.) but with leaks, in normal practice, these factors are unknown. A commonly
reported table in industry literature that may help in immediate flow identification is:

Flow type Leak rate (Pa · m3/s)

Turbulent flow > 10-3

Laminar flow 10-2 ÷ 10-7

Transitional flow 10-5 ÷ 10-8

Molecular flow < 10-8

Table 2 - Leak rate and type of flow

The most common leaks are in the turbulent and laminar flow range, while small leaks are characterized
by a transitional flow. Molecular flow signifies only very small leaks.

5.2.1 Turbulent flow

Turbulent flow occurs only in larger leaks and at higher pressure differences. Due to the high speed of
gas, leaks with turbulent flow emit a sound (whistle) and these can be detected and located by ultrasonic
leak detectors.

Fig. 3 - Typical velocity profile of a particle in turbulent flow

VTECH_leak_detection -7- Rev. 1 – 29 Mar. 06

The formula for the leak rate at turbulent flow is not given here since leaks with turbulent flow are so
large and can be readily located and repaired. There is seldom a need for calculation. In general, in
turbulent flow increasing the internal pressure will increase the volumetric flow rate until the gas flow
velocity reaches the speed of sound and the flow becomes “chocked”. From that point, increasing the
pressure does not increase the volumetric flow rate, but the gas density will increase.

5.2.2 Laminar flow

Laminar flow is defined as a parallel flow of molecules in a pipe, whereby the transverse distribution of
the speed of the molecules is parabolic, as sketched in the following picture.

Fig. 4 - Typical velocity profile of a particle in laminar flow

The speed of the molecules in the centre of the pipe is at a maximum value and it decreases towards the
walls of the pipe. Laminar flow through a straight pipe with circular cross-section is described by the well
known Poiseuille formula.

Considering a leak like a cylindrical path (virtual leak), the leak rate can be expressed as:
Q
 r4
16   l

 pi  po
2 2

Where:
 is the dynamic viscosity of the gas (Pa · s) (e.g. 1.8·10-5 Pa · s for air 2.0·10-5 Pa · s for He)
l is the length of the round tube, that is the virtual leak path (m)
r is the radius of the pipe, that is the virtual leak radius (m)
pi is the absolute internal pressure (Pa)
po is the absolute external pressure (Pa)
Q is the flow rate, i.e. the leak rate (Pa · m3/s)

From this equation the main property of the laminar flow emerges. The volumetric leak flow rate does
not depend on gas molecular weight, but on gas viscosity. Therefore, it is not true that helium leaks four
times more than nitrogen, from a given defect at a given pressure, since it is four time lighter then
nitrogen. Instead, since helium viscosity is about ten percent higher than nitrogen at room temperature, a
product with a fixed defect and differential pressure will leak less if filled with helium rather than
nitrogen.

Furthermore, if the pressure difference across a leak changes, the leak rate changes with the square of the
pressure. Assuming that the geometrical dimensions of the leak do not change during the measurement
period, the following relation can be written:

QA QB

 p Ai  p Ao
2
 
2
pBi  pBAo
2 2

Where:
pAi and pBi are the absolute internal pressure (Pa)
pAo and pBo are the absolute external pressure (Pa)

VTECH_leak_detection -8- Rev. 1 – 29 Mar. 06

QA and QB are the corresponding leak rate (Pa · m3/s) in the two cases

From this equation it can be stated that a drastic increase of sensitivity test can be achieved by increasing
the internal pressure of the object under test. In large containers filled with a tracer gas (e.g. helium) for a
sniffer-test, cost reduction can be achieved by increasing the inner total pressure, but reducing the
concentration of the test gas. Of course the safety conditions for filling containers with pressurized gas
must be followed. The Poiseuille equation can be rearranged so to express the leak as mass flow rate:

M
 r4
16   l  k B  T

 m  pi  po
2 2

Where:
 is the dynamic viscosity of the gas (Pa · s) (e.g. 1.8·10-5 Pa · s for air 2.0·10-5 Pa · s for He)
l is the length of the round tube, that is the virtual leak path (m)
kB is the Boltzmann constant (1.38 · 10-23 J/(molecules · K)
T is the absolute temperature (K)
r is the radius of the pipe, that is the virtual leak radius (m)
m is the mass of one molecule (kg) (e.g. 4.78·10-26 kg for air, 6.6·10-27 kg for He)
pi is the absolute internal pressure (Pa)
po is the absolute external pressure (Pa)
M is the total mass flow (kg/s)

To convert mass flow to volumetric flow (m3/s) the previous equation must be divided by the density of
the gas (e.g. 1.16 kg/m3 for air 0.17 kg/m3 for He).

5.2.3 Molecular flow

Molecular flow exists in small leaks (cf. Table 2) and at low pressures. In a molecular flow condition
each molecule travels independently of other molecules. The mean free path is larger than the diameter of
the leak capillary. Therefore it is possible for a single molecule to travel against the general flow
direction, because molecules do not collide with each other, but only with the walls. Despite this fact, the
general flow is in the direction of the pressure gradient.

Fig. 5 - Typical velocity profile of a particle in molecular flow

Molecular flow through a straight pipe with circular cross-section is described by the Knudsen formula.
Considering a leak like a cylindrical path (virtual leak), the leak rate can be expressed as:

2 R T d 3
Q     pi  po 
6 M l
Where:
R is universal gas constant (8.314472 J/(mol · K))
T is absolute temperature (K)
M is relative mass of molecule
d is the diameter of the pipe, that is the virtual leak diameter (m)
l is the length of the round tube, that is the virtual leak path (m)

VTECH_leak_detection -9- Rev. 1 – 29 Mar. 06

pi is the absolute internal pressure (Pa)
po is the absolute external pressure (Pa)
Q is the flow rate, i.e. the leak rate (Pa · m3/s)

From the previous formula one can see that, to the contrary of the laminar flow, leak rate follows linear
proportion to pressure difference. To calculate a leak rate after a pressure change, the following formula
is valid:

QA QB

 p Ai  p Ao   p Bi  p BAo 
Where:
pAi and pBi are the absolute internal pressure (Pa)
pAo and pBo are the absolute external pressure (Pa)
QA and QB are the corresponding leak rate (Pa · m3/s) in the two cases

5.2.4 Transitional flow

There is a gradual transition from laminar to molecular flow, which can be interpreted as both flows are
present simultaneously or as a change of condition long the path, i.e. the flow, laminar at the entrance of
the leak, gradually changes to molecular at the end of the leak.

Fig. 6 - Typical velocity profile of a particle in molecular flow

The mathematical description of this flow condition is difficult. There are some formulae available, but
all of them have some restrictions. The simplest formula, from Burrow, combines the laminar and
molecular flow:

Q
 r4
16   l

 pi  po 
2 2 2
6


R T d 3
M

l
  pi  po 

Where:
 is the dynamic viscosity of the gas (Pa · s) (e.g. 1.8·10-5 Pa · s for air 2.0·10-5 Pa · s for He)
R is universal gas constant (8.314472 J/(mol · K))
T is absolute temperature (K)
M is relative mass of molecule
l is the length of the round tube, that is the virtual leak path (m)
r is the radius of the pipe, that is the virtual leak radius (m)
d is the diameter of the pipe (d = 2 · r) (m)
pi is the absolute internal pressure (Pa)
po is the absolute external pressure (Pa)
Q is the flow rate, i.e. the leak rate (Pa · m3/s)

VTECH_leak_detection -10- Rev. 1 – 29 Mar. 06

This formula can be used for rough calculations. For practical use it is preferable to estimate if the
laminar or molecular flow predominates and utilize the correspondent Poiseuille or Knudsen formula to
calculate the dimension of the idealized leak.

5.3 Gas Dependency

In the previous paragraphs, starting from the relations describing gas flow throughout leaks, the effects of
pressure are reported. The dependencies of leak rates on the gas properties are analyzed in the following.

5.3.1 Laminar flow

Recalling the Poiseuille equation:

Q
 r4
16   l

 pi  po
2 2

One can see that the gas flow rate is inversely proportional to its dynamical viscosity. So changing the
type of gas, in the same pressure condition, it can be written:

QA  B

QB  A
Where A and B are the dynamic viscosities (Pa · s) of the two gases and QA and QB are the leak rate (Pa
· m3/s) in the two cases.

Annex B reports a table of the viscosities of some gases. Since the viscosity values for most gases are
very closed each other (they have all the same order of magnitude), little variation in flow rates result
changing the test medium. In Annex B is reported also a conversion table of leak rates.

5.3.2 Molecular flow

From the Knudsen equation, given here again for convenience:

2 R T d 3
Q     pi  po 
6 M l

One can see that the gas flow rate is inversely proportional to the square root of its molecular relative
mass. So changing the type of gas, in the same pressure condition, it can be written:

QA MB

QB MA

Differently from the laminar flow, in the molecular flow condition, a change of gas can make larger
differences in leak rate (at constant pressure difference). In annex B, the molar masses and the leak
correction factors for some gases are reported.

5.4 Conversion from Mass Leak Rate to Volume Leak Rate

In vacuum technology, the leak rate in general is described as a volume leak rate (Pa · m3/s or mbar · l/s).
In many industrial applications, e.g. the refrigeration industry, it is common to define the leak rate in loss
of mass of a specified material per year, for example grams of refrigerant per year (g/y). Clearly the two

VTECH_leak_detection -11- Rev. 1 – 29 Mar. 06

indications are related and a conversion is possible using the kinetic theory of gas. Assuming a
simplifying hypothesis of ideal gas and known flow condition, laminar or molecular, this conversion is
easy, as well as determining the behavior of different gases. In this condition it can be written:

Fs VM
Q  T
M T0
Where:
T is the ambient absolute temperature (K)
M is molar mass of the gas (kg/mol)
T0 is the absolute temperature in standard condition (273.16 K)
VM is the molar volume in standard condition (22.414·10-3 m3/mol)
Fs is the mass leak rate (kg/s)
Q is the flow rate, i.e. the leak rate (Pa · m3/s)

Knowing the internal pressure and the type of leak flow, using the equation previously discussed the
equivalent flow rate of different gases or at different pressures can be computed.

5.5 Diffusion of Gases

Diffusion is the phenomenon of the spreading of one gas into another. Diffusion happens also with liquid,
but in leak detection only diffusion of gases are important. Under equal conditions, light gases diffuse
faster than heavy gases. The diffusion of gases is inversely proportional to their relative molecular mass.
The diffusion law, known also as Graham Law, is:

D1 M 2

D2 M 1

Where D is the diffusion coefficient and M is the molecular mass. In Annex B the diffusion coefficient
for the most commonly used gases are reported.
The diffusion of one gas into another at atmospheric pressure is relatively slow. This requires special
attention when test pieces are filled with tracer gas for a sniffer test. With long tubes with closed ends, the
tracer gas cannot bypass the air which is in the test piece as far as the end of the tube. This air cushion can
remain there for some time, before it is mixed with the other gases by diffusion. A leak, which happens to
be at this end of the tube, may not be detected by the sensor, which is only sensitive of the tracer gas.
To avoid this problem, test objects should be evacuated prior to filling with tracer gas. Diffusion speed
increases with decreasing pressure since at low pressure the mean free path increases and, therefore, the
collisions among gas particles decrease. If a complete filling with tracer gas is desired, an evacuation
down to a pressure of 10 to 50 Pa is sufficient.

5.6 Permeation
Permeation is the passage of a fluid into, through and out of a solid barrier having no holes. The process
involves diffusion through a solid and may involve many phenomena such as adsorption, dissociation,
migration and desorption.
Permeation can have negative effects on a helium leak test, especially when the leak rate specification is
low and the test time long. Some materials, e.g. PTFE, require special attention because helium is highly
permeable. Accurate results are almost impossible to get from Helium leak tests on pieces containing
seals of such materials. The following figure shows the permeation characteristics of some elastomers.

VTECH_leak_detection -12- Rev. 1 – 29 Mar. 06

 Fig. 7 - Permeation

Permeation does not affect routine leak tests, when the measurement occurs in times too short to permit
permeation to take effect.

6. LEAK TESTING IN REFRIGERATION INDUSTRY

In the refrigeration industry, components and system must be leak tested to ensure that refrigerant
leakages are below specified limits. These reference limits may depend either on the maximum
acceptable leak rate, consistent with the reasonable working life expectation for final products, and,
especially in certain countries, either on rule and regulation constraints. The basic functions of leak
testing are determining if there is leakage or not (detection), measurement of leak rate and leakage
location. There are many methods and many types of test equipment but to approach these problems but
unfortunately there is no universal solution – no one method fits every situation. Each testing method is
suitable only for a specific leak rate or for fixed forms and technologies. In the common use of leak
detection, explicit leak rate measurement is not required, but the system must be able to recognize if the
leak rate is above or below a specified level, in order to meet the required tightness. Usually the
acceptable leak rate, depending on refrigerant type and application, spans from 15 g/y of refrigerant, as
for some big air conditioning system and/or automotive application, to 0.5 g/y, as for some domestic
refrigerators.

This acceptance level is the main parameter to be considered when selecting the proper testing method or
combination of methods. Several other factors have to be taken into account for a correct choice. In
particular, system costs, complexities, environmental impact, reliability, influence of external conditions,
operator dependence and user friendliness.

A lot of documentation about leak-testing, leak detection and leak location methods is available in
industry literature. Some references are reported in Annex E. This section presents some leak detection
techniques and comparison of their performance with particular regard to refrigerant leakages.

7. LEAK TESTING METHODS

A leak can be defined as an unintended crack, hole or porosity in an enveloping wall or joint which must
contain or exclude different fluids and gases, allowing the escape of closed medium. Critical leak spots in

VTECH_leak_detection -13- Rev. 1 – 29 Mar. 06

closed systems are usually connections, gaskets, welded and brazed joints, defects in material, etc. A leak
test procedure is usually a quality control step to assure device integrity, and should preferably be a one-
time non-destructive test, without impact on the environment and operators. Several leak-testing
techniques are available, spanning from very simple approaches to systems that are more complex. The
most commonly used non-destructive leak test methods are the underwater bubble test, bubble soap
paint, pressure and vacuum decay and tracer gas detection (halogen, helium and hydrogen). The first
three techniques, due to their characteristics and sensitivity, can be used only for gross leak detection, 300
g/y or more refrigerant leakages. Tracer gas leak testing methods are much more responsive than the
previous group but, in many cases, their theoretical sensitivity is more than required. In the practical use,
however, this is limited by environmental and working conditions.

In the following, each method is briefly described and its advantages and drawbacks are reported.
In annex A, a conversion chart for the most commonly used vacuum and leak rate measurement units is
provided.

In the diagram below, the performance of various leak-test techniques are summarized:

LEAK DETECTION SENSITIVITY

Ultrasonic

Bubble test (soap painting)

Bubble test (He, alcohol)

Water immersion (bubble test)

Acustical
Leak Detection Tecniques

Pressure decay (without pressure differential)

Pressure decay (with pressure differential or

compare mode)

Vacuum decay

Thermoconductivity

Halogen sniffer

Helium sniffer (inside-out)

Hydrogen detector

Vacuum chamber helium leak test (inside-out)

Helium leak test outside-in

1.E+02 1.E+01 1.E+00 1.E-01 1.E-02 1.E-03 1.E-04 1.E-05 1.E-06 1.E-07
Leak Rate (mbar l/s or atm cc/s)

Fig. 8 - Leak detection sensitivity of some techniques

VTECH_leak_detection -14- Rev. 1 – 29 Mar. 06

It is important to note that a leak test would not be possible if the leak test personnel could not control the
test permanently against failures.

In the following, each leak detection method is analyzed in detail.

7.1 Ultrasonic Leak Detector

Acoustical leak detection uses the sonic or ultrasonic energy generated by gas as it expands through an
orifice. Pressurized gas proceeds from a tested system through leaks which are detected outside by a
sensitive microphone (typically about 40 000 Hz). Acoustical leak detection is widely used in testing
high pressure lines, ductworks etc. It requires modest instrumentation; it is simple and fast but is limited
to about 10-3 mbar l/s. This procedure can be very successful if, and only if, the area in which you’re
trying to detect the leak in is absolutely quiet area. This is usually impossible, which imposes great limits
on the ultrasonic leak detector. The higher the pressure in the system, the greater is the chance of finding
or even hearing the leak.

7.2 Water Immersion Bubble Test Method

The water immersion bubble test, called also "bubble testing" or "dunking" is a traditional and a
relatively primitive technique of leak detection. It consists in immersing a charged or pressurized part,
usually with high pressure dry air or nitrogen, in a water tank and watching for bubbles to escape from
the eventual leaks. The larger and more frequent the bubbles, the bigger the leak is. Small leaks are
possible to detect, but very difficult. The main limit of this method is sensitivity, that is, the minimum
detectable leak rate. Considering a spherical bubble of radius R, its internal volume V will be:

V = 4/3 · π · R3

Let p the pressure inside the bubble and t the time required to form the first bubble, the leak rate Q will
be:

Q = (p · V) / t

The two key parameters determining the sensitivity of this method are the smallest bubble detectable by
the operator and the waiting time for bubble generation. This time must be compatible with the
production rate and with operator attention.

It is reasonable to consider that the smallest bubble an operator could detect has 1 mm radius and that the
waiting time is 30 seconds. Assuming that the pressure inside the bubble is the atmospheric pressure,
from the previous equations, it can be stated that the bubble volume is V = 4.2·10-3 cm3 and then
minimum detectable leak rate is:

Q= (p · V) / t = 1000 * 4.2·10-6 / 30 = 1·10-4 mbar · l/s

This is a theoretical value. The real sensitivity is strongly affected by many external factors, such
illumination conditions, water turbidity, unit location and placement and water movement. All these
issues, together with operator dependency, limit the useful sensitivity to 5·10-4 mbar · l/s, but usually
1·10-3 mbar · l/s is considered. The following picture exploits the dependency of bubble formation time
and leak rate at several internal test pressures.

VTECH_leak_detection -15- Rev. 1 – 29 Mar. 06

 Fig. 9 – Bubble formation time and leak rate at different internal test pressure

In the following picture, mass leakage of refrigerant (R134a) is related to the bubble formation time.

Fig. 10 – Bubble formation time and mass leak rate at different internal test pressure

Some tricks can be introduced to get some improvements to this method. For instance, an incremental
increase in the internal pressure may yield more a probable and less time consuming leak pinpointing. A
detergent can be added to the water to decrease surface tension, which helps to prevent the leaking gas
from clinging to the side of the component. Using different gases (e.g. helium) and/or liquid may give
some advantage in system performance, at a cost disadvantage. Hot water in the tank sometimes helps to
increase the pressure inside the component or piping system. If dry nitrogen is used, this does not help
because nitrogen does not increase its pressure significantly. If refrigerant is contained in the system or
component, it may help considerably to increase the pressure and, therefore, increase the chance of
finding the leak.
VTECH_leak_detection -16- Rev. 1 – 29 Mar. 06
It can be concluded that this technique features accuracy in the 10-3 mbar · l/s range in high volume
production applications, but it is not recommended. Immersion is a very economical leak testing method.
It allows leak detection and, in most cases, also leak location. However, disadvantages range from a
relatively low sensitivity, high operator dependency, and possible part contamination, to fluid waste and
the possibility of having to dry parts after testing. Moreover, in particular for big coils, the unit handling,
to put parts in and out from tanks, may be complex and a source of part damages. Some costs may be
hidden. This approach, in fact, implies using large space and can produce a certain amount of waste -
water. Especially for big units, like in large heat exchangers, the tank could have great dimensions and
require a lot of water. The dryers are not costless machines; they require as much care and maintenance
as any in the plant.

7.3 Soap Solution Bubble Test

Instead of submersing the part, the pressurized unit to be tested can be painted with a soap solution and
again the operator can see the bubbles escaping where the leak is. Soap solutions are available in many
different types. Some have a brush applicator and others have a dabber (an absorbent ball attached to a
stiff wire inside of the cap). Some brands may even have a spray applicator to quickly cover large areas
of tubing in a short amount of time. This is an advantage but is also messy and time consuming to clean
up. Some soap solutions even have an antifreeze base to prevent them from freezing in the winter time.
Others may have a lower density to make them even more sensitive to very tiny leaks.

This method has a higher sensitivity than the water immersion. It enables detecting the leakage up to 10-5
mbar · l/s and is suitable for very large systems. This method is particularly useful if the soap solution can
be applied in the approximate area where a leak may is known to exist. In this case the soap solution can
be used in that area to test for and pinpoint a leak. It is the simplest and least expensive (material wise)
method known today. However, it may also be more expensive to use, because of labor costs, if the
operator does not have any idea where the leak could be. Increasing the gas pressure makes it more a
probable and less time-consuming way of pinpointing the leak. However, for operator safety, the pressure
must be limited to low values, up to 1700 kPa (250 psi). The use of the soap solution bubble test is
limited by some drawbacks. The surface area to be painted should be simple and easily accessible.
Otherwise, as for finned pipes or the bottom part of a large heat exchanger, it could be a hard, if not
impossible, for the operator to “paint” the part and watch for a bubble. Moreover, the application is not
well suited for high productivity lines.

7.4 Pressure Decay Test

This method consists of pressurizing the system with a high pressure gas, usually air or dry nitrogen.
Then the part is isolated from the gas supply and, after a stabilizing time, its internal pressure is controlled
over time. The pressure drop ∆p is measured in the time ∆t. If the pressure in the system drops fast, there
is a large leak present in that component or section of the system. If the system’s pressure drops slowly,
there is a small leak present. If the pressure remains the same, that component does not leak. The leak
rate Q can easily computed considering the volume V of the component. That is:

Q = (∆p · V) / ∆t

Leak detection sensitivity is related to the testing time, the pressure transducer resolution and the volume.
The most advanced system measures a pressure variation even less than 70 Pa (0.010 psig) at test
pressure. Depending on the volume of the units to be tested, the leak detection cycle can be as short as 30
seconds and guarantees high resolution. Considering V = 1.5 dm3 (0.4 gal) internal volume component
with a ∆p = 70 Pa (0.010 psig) of pressure decay at 3450 kPa (500 psig) test pressure in ∆t = 60 seconds,
the leak rate is:

VTECH_leak_detection -17- Rev. 1 – 29 Mar. 06

 Q = (∆p · V) / ∆t = 0.7 · 1.5/60=1.7·10-2 mbar · l/s

Several external factors, such as temperature variations and mechanical deformations, affect this test. The
internal pressure, in fact, depends on temperature, and thermal fluctuations may cause changes in
pressure altering the results. Fortunately, dry nitrogen experiences very little pressure changes when it is
exposed to small temperature changes.

The sensitivity of this testing technique depends on pressure measurement resolution, test time and
pressure values. The following picture shows the typical test sensitivity related to internal test pressure
and leak rate, for a reference volume (1 litre or 0.26 gal) and a fixed pressure difference (100 Pa or 0.014
psi).

Fig. 11 – Computed pressure test decay sensitivity

Long test times allow a more sensitive check but, in this way, the test can be very time-consuming
because some low-level leaks may require a very long holding period, even some hours. The unit
volume also affects test sensitivity. In fact, the same pressure variation will correspond to a larger leak if
the volume is bigger, and therefore the sensitivity will be less. The higher the pressure, the faster you can
determine if a leak is present. However, operator safety concerns limit the maximum admissible pressure
value. Components can be leak tested at low pressures, less than 2 MPa (290 psig) without protection,
higher pressures, e.g. 7 MPa (1000 psig), may be used adopting safety interlocked protection hoods.
Using the proper pressure, this test method also allows compliance with technical specifications, such as
American Underwriters Laboratory (UL) Burst test and the European EN378 rules. Burst test is designed
to test the mechanical strength of the refrigeration tubing circuit, i.e. ruptured tubing, bad brazed joints
with material separation. Pressure ranges for the test vary depending on if the test unit is a component of
the refrigeration circuit, or if it is a complete refrigeration circuit with a compressor.

The testing performance can be improved using a pressure differential. In this mode, the test unit is
pressurized together with a reference volume and the two pressure trends are compared. For this reason
this method is know also as compare mode.

VTECH_leak_detection -18- Rev. 1 – 29 Mar. 06

 Straight
pressure
decay

Differential
(compare)
pressure

pressure
decay

Filling Settle Test Exhaust

time time time time

time

Fig. 12 – Typical pressure behavior in pressure decay test

Pressure decay proof is a go-no go test. It detects the leakage, but leak location requires using other
techniques, such as soap paint, or better, tracer gas detection. Usually, the limit is in range 10-3 mbar · l/s
for pressure decay test without pressure differential and 10-4 mbar · l/s for pressure decay test with
compare mode (pressure differential).

This leak testing technique has some advantages; this method will positively identify whether or not a
leak exists by monitoring pressure drop. If any pressure drop occurs, it means a leak is definitely present.
Furthermore this method can be realized completely automatic, so as to avoid operator errors. This
procedure is a preliminary leak test, which detects large leaks before the final automatic leak test
operation using tracer gas, e.g. helium. This test will detect over 90%, up to even 98%, of the defective
parts, especially those not brazed correctly. If the test unit has a large leak, i.e. an over looked brazed
joint, without doing the pressure decay test first, large quantities of helium, would leak out of the test unit.

The helium, at this much volume and concentration, would render the system inoperable for hours.
Another advantage of the nitrogen filling, besides the mechanical stress and leakage test, is the purging of
the circuit to be tested, lowering its humidity. On a successful completion of the pressure decay test, the
component is ready for final test.

The disadvantage is that this method does not identify where the location of the leak, only if a leak is or is
not present.

7.5 Vacuum Decay Test or Pressure Rise Test

Vacuum decay test or pressure rise test works in the opposite way of the pressure decay test. This method
involves evacuating the part to suitably low pressure and, after stabilizing the pressure, measuring the
increase in pressure caused by test media entering the part. Only parts that are able to withstand external
pressure can be tested in this way (e.g. thin walled plastic parts cannot be tested due to the danger of
collapsing).

VTECH_leak_detection -19- Rev. 1 – 29 Mar. 06

Even if in a vacuum decay test is not possible to get more than one atmosphere of pressure difference
from inside to outside, using some solvents (i.e. alcohol, acetone or similar) exalts the pressure increment
due to solvent entering into the leak. This approach, however, has some shortcomings, such as the
possibility of solvent freezing causing temporary stuffing of leak, or elastomer gaskets becoming
damaged by solvents. Like the pressure decay test, the accuracy of this method is related to the volume of
the part to be tested; the bigger the volume the smaller the sensitivity will be. Respective of the pressure
decay test, however, this technique has some advantages. It is less sensitive to temperature changes since
the pressure inside the part is lower than atmospheric pressure. Furthermore, vacuum-meters usually are
very sensitive to small pressure changes, so the theoretical sensitivity might be very high, up to 1·10-5
mbar · l/s.

Fig. 13 – Theoretical pressure rise behavior in vacuum decay test

However, surface out-gassing and liquid evaporation affect and limit the real sensitivity. For instance,
small quantities of water, even a few grams, start evaporating at 2 kPa (0.3 psia) and at 1 Pa (7.5 µm Hg)
the water vapor content is so high that the consequent pressure increase is comparable to a leakage.
In refrigeration circuits, where the out-gassing form oil is so significant that could be mistaken for a leak,
the sensitivity is limited to 1·10-3 mbar · l/s

Fig. 14 – Pressure rise due to leak and out-gassing exemplification

VTECH_leak_detection -20- Rev. 1 – 29 Mar. 06

Vacuum decay method can be realized in a fully automatic procedure and, in this way, it is independent
of an operator. This technique is a “pass-no pass” test. It detects the total system leak and more than one
leak can exist, but leak location requires other techniques. In vacuum decay test, the unit to be tested is
evacuated and its internal pressure is lower than atmospheric pressure. Therefore, this leak testing method
will stress the part in the opposite way, if the working condition requires an internal pressure.

7.6 Tracer Gas Leak Testing

The words “tracer gas leak testing” point out a group of several test methods characterized to detect and
measure a tracer gas flowing through a leak. These techniques differ for the tracer gas used and for the
realization technology.

The most commonly used tracer gases are halogen gas, just CFC, HCFC and HFC refrigerant, helium
and a mixture of nitrogen 95% hydrogen 5%. Despite the simplicity of their detectors, electronic devices,
halogens are loosing their appeal as tracer gas, due to environmental protection rules following Montreal
and/or Kyoto protocols. On the other hand, helium and especially hydrogen/nitrogen mixture are gaining
ever more interest.

Helium has been used as tracer gas for long time and this success is due to its physical properties. It is
neither toxic nor flammable. It is an inert gas and does not react with other compounds. Helium has low
viscosity and relative molecular mass, so it easily passes through cracks. In the same environmental
conditions, it may flow through orifices 2.7 times faster than air. Since its concentration in air is low (5
ppm), it is easy to detect an increment of helium concentration. However, there are some shortcomings.
Helium slowly disperses into the atmosphere so, in case of big leaks, its concentration will only return to
a value suitable to continue leak testing in a long time, even hours. It is not cheap, even if less expensive
than halogen gases. The most appropriate helium detector is based on mass spectrometer, which is an
expensive and delicate apparatus requiring care and maintenance, more suitable for a laboratory than for
a manufacturing industry.

The mixture of nitrogen 95% and hydrogen 5% is a relatively new tracer gas. Hydrogen has a number of
properties that make it an excellent trace gas for leak testing, even in production environments. It is the
lightest element, with higher molecular speed and lower viscosity than any other gas, so it is easy to fill,
evacuate and dissipate, it finds and passes through a leak faster, is easier to flush out and vent away and
its molecules do not stick to surfaces as easily as helium atoms. It is environmentally friendly and
renewable. More importantly, it has the highest leak rate of any gas. Moreover, the normal background
concentration of hydrogen (0.5 ppm) is ten times less than helium. Detectors use a semiconductor sensor,
have no moving parts, making it completely maintenance free. These devices are not affected by the
presence of other gases. Pure hydrogen should never be used as trace gas, but a standard industrial grade
mix of 5% hydrogen in nitrogen is inexpensive, non-flammable (as for ISO10156 specification), easily
available from industrial gases suppliers.

It is important to remember that background concentration in air is a limiting factor for any tracer gas
detector. There are two ways to carry out leak testing with tracer gas: external detection of tracer gas
escaping from leaks of a filled unit (inside-out method), and internal detection of tracer gas entering from
leaks (outside-in method). For each of these two methods there are two realization techniques. The
inside-out methods can be executed with atmospheric sniffing or with vacuum chamber detection, while
the outside-in method is generally implemented putting the unit to be tested in a room containing the
tracer gas or, very rarely, spraying the tracer gas on the unit surface. In the following sections, each of
these methods is described.

VTECH_leak_detection -21- Rev. 1 – 29 Mar. 06

7.6.1 Sniffing
This is the simplest realization of an “inside-out” testing. The sniffing technique of leak detection utilizes
a detector probe, or sniffer, to sense leaks from a unit previously filled and pressurized with the tracer
gas. Before filling the unit with the tracer gas, it must be evacuated, so a pumping group, even a small
one, is required. This method is very operator dependent. In fact, the probe (or wand) is moved over the
part and detects the leak as it passes over that leak. The speed, distance from the part and the probe
sensitivity determine the accuracy of leak detection. However, sniffing will locate a leak on a part, unlike
the other methods described, and has the ability to sense leaks as small as 10-7 mbar · l/s, depending on
the tracer gas. Sniffing is not recommended in a high volume production environment, other than for
locating leaks for repair. Depending on the tracer gas, sniffing may involve a relatively low tooling cost
investment, representing an economical method of leak detection. Besides, the cost of the tracer gas may
be significant and, in case of a particularly expensive gas, the used of a suitable gas recovery and reclaim
system should be considered, further increasing the overall costs. Disadvantages include a high chance of
missing leaks due to operator dependency, fragile equipment in rugged environments, rejecting good
parts (because of the inability to quantify the leak) and it is not a good overall leak detection technique.
Some sniffers and the relevant detectors require a periodic maintenance to assure proper functioning,
since they are complex system composed of vacuum pumps, mass spectrometer and vacuum fittings.
Electronic detectors, without moving parts, are very profitable. Some detectors are sensitive to gases
other than the tracer gas used. Therefore, using these sensors, attention to the chemical environmental
conditions is required.

The minimum leak rate measurable by sniffer is the concentration of the tracer gas in the working area,
the value known as background level. This level may change during the production cycle and increase in
the case of leaking units. Relating to the tracer gas used, in case of big leak in the part under test, a lot of
tracer gas escapes from it and may remain for long time in the working area strongly affecting the
following tests, causing rejection of good parts. It is good practice to use a preliminary leak testing
system to reject parts with gross leakages. It is possible to integrate this preliminary test, i.e. a pressure
decay test, in the tracer gas-filling machine, so as to simplify the system.

It is important to note that sniffing techniques are local methods, allowing the testing of single points.
Each of the tested points can have a leak below the sniffing sensitivity, but the overall leakage may be
above the acceptance limit. As result, the test is successful, but the part is defective. Global tests, such as
vacuum chamber inside-out and outside-in methods, solve this problem.

7.6.2 Accumulation leak testing

This method is a variation of sniffer leak testing. The part to be tested is enclosed in a closed containment
box, and then it is pressurized with the tracer gas. The sniffer is connected to this hood where the leaked
tracer gas is accumulated during the test time. When accumulated, the tracer gas is more readily sensed
by the detector. The gas sensor will measure the total or global leak.

In addition to the characteristic limitations of the sniffer testing method, this technique has other
drawbacks. The larger the accumulation volume, the longer the time needed to detect the leakage. Leak
rate and sensibility depend on the residual volume and the test time. The tracer gas partial pressure
increment ∆p, the tracer gas flow Q, residual volume V and the test time ∆t can be expressed with a
sample equation:

∆p = (Q · ∆t) / V

This method is used in very special applications, e.g. very small components to be tested for tiny leaks.

VTECH_leak_detection -22- Rev. 1 – 29 Mar. 06

7.6.3 Vacuum chamber inside-out leak testing
Vacuum chamber inside-out leak testing is the most complex system for leak detection, but it is
theoretically suitable to find very small leaks, using a proper tracer gas. The equipment is composed of
one or more vacuum chambers, large enough to house the unit to be tested. The chamber is connected to
a vacuum pumping group equipped with the tracer gas detector, for chamber evacuation and gas
detection. A second vacuum group is required to evacuate the unit under test before filling it with gas. A
tracer gas-filling device completes the testing apparatus. The unit to be tested is put into the vacuum
chamber and connected to service hoses. Then the vacuum chamber and the unit are evacuated. During
chamber evacuation, the part is pressurized with the tracer gas and, after a stabilization time, the detector
is linked to the vacuum line, so to detect the tracer gas flown through a leak and sucked by the pumping
group. In this way, the leakage is detected. It is a “go-no go” test and to locate the leak requires other
techniques.

This method has some advantages. This technique is fully automatic, so it is not very operator dependent.
Its sensitivity, depending to tracer gas and test time, can reach 10-10 mbar · l/s, even if in a practical
application in the refrigeration industry the limit is 10-6 mbar · l/s.

There are also some drawbacks. Depending on the vacuum chamber dimensions, the evacuation group
can have a high pumping speed. Some gas detectors require a periodic maintenance to ensure proper
functioning, since they are complex system composed of vacuum pumps, mass spectrometer and vacuum
fittings. Besides, the cost of the tracer gas may be significant and, in case of a particularly expensive gas,
the used of a suitable gas recovery and reclaim system should be considered, further increasing the
overall costs and the system complexity. Concerning the tracer gas used, in case of big leak in the part
under test, a lot of tracer gas escapes from the leak. A long pumping time could be required to lower the
tracer gas in the detector to an acceptable level compatible with system function. The system is unusable
during this time. It is good practice to use a preliminary leak testing system to reject parts with gross
leakages. It is possible to integrate this preliminary test, i.e. a pressure decay test with the vacuum
chamber test, so to simplify the system.

Another disadvantage is that this method does not identify where the leak exists, only if a leak is or is not
present. It detects the total system leak and more than one leak can exist. Leak location requires other
techniques.

7.6.4 Outside-in leak testing

In this testing technique, the unit to be tested is put into a containment hood containing the tracer gas. The
part is connected to a vacuum group and evacuated. A tracer gas detector is placed in the vacuum line to
detect the tracer gas flow in a leak and sucked in by the pumping group.

This method has some advantages. This technique is fully automatic, so it is not very operator dependent.
The sensitivity, depending on the tracer gas and test time, can reach 10-6 mbar · l/s. The containment
hood can be designed to prevent dispersion in order to reduce working area pollution and tracer gas
consumption, saving money and avoiding the need for a recovery system.

There are also some drawbacks. The difference from inner and outer pressure is limited to values slightly
above the atmospheric pressure. Concerning to the tracer gas used, in case of big leak in the part under
test, a lot of tracer gas escapes from the leak. A long pumping time could be required to lower the tracer
gas in the detector to an acceptable level compatible with system function. The system is unusable during
this time. It is good practice to use a preliminary leak testing system to reject parts with gross leakages. It
is possible to integrate this preliminary test, i.e. a pressure decay test, in the tracer gas-detecting machine,
so to simplify the system.

VTECH_leak_detection -23- Rev. 1 – 29 Mar. 06

Another disadvantage is that this method does not identify where the leak is, only if a leak is or is not
present. It detects the total system leak and more than one leak can exist. Leak location requires other
techniques.

In the following sections, the main applications of the previously reported methods are described.

7.6.5 Halogen leak detectors

The working principle of the halogen detector is based on the measure of a positive ions emission due to
the halide presence inside an electronic cell. This ion current is related to the halogenated gas
concentration and, therefore, to the leak size. Less sensitive detectors are based on infrared light
absorption by halogenated gas. The main application for halogen detectors is for inside-out systems,
while their use for outside-in methods is very limited.

In the inside-out method, halogen leak detectors are used in the detector-probe mode, requiring that the
system be pressurized with a tracer gas containing an organic halide, such as CFC, HCFC and HFC. The
exterior of the system is then scanned with a sniffer probe sensitive to traces of the halogen-bearing gas.
The achievable sensitivity can be 10-5 mbar · l/s.

In outside-in mode, an evacuated vessel is connected to a halogen detecting instrument and is sprayed by
halogenated gas. In this manner, its performance is up to 5.10-7 mbar · l/s and is used in rough, medium
and high vacuum. This method is quite complex and has high environmental impact, so it is rarely used.
Halogen leak detectors have a wide spread usage in refrigeration and air conditioning maintenance to
locate leaks in refrigerant charged system due to their high sensitivity. In manufacturing industry,
however, their use is limited because of several disadvantages and drawbacks. Refrigerant has a higher
specific volume than air; therefore refrigerants will fall when exposed to atmospheric pressures. This
means leak detecting on the bottom sides of the piping or components will be more effective in detecting
a leak.

Since electronic halogen detectors are sensitive to many gases, included non-halogenated ones, such as
carbon monoxide and alcohol, their sensitivity is strongly determined by the tracer gas type and
environmental conditions. The best performance is achievable if used in a controlled atmosphere room.

If a unit previously charged with refrigerant has to be evacuated and re-charged, the pre-evacuation phase
is difficult and very time consuming. Halogenated gases are costly, more expensive than other tracer
gases, like helium and nitrogen-hydrogen mixture. Halogen gases have a high environmental impact and
their dispersion in the atmosphere is severely regulated, if not forbidden, in many countries.

7.6.6 Inside-out helium sniffer detectors

A Helium sniffer is a detector probe and an ancillary accessory of leak detectors. A vacuum pump inside
the leak detector maintains the helium spectrometer in high vacuum (up to 1·10-2 Pa or 7.5·10-2 µm Hg).
One side of the sniffer is connected to this vacuum group, while its detection probe, provided with a
calibrated orifice, is opened to the atmosphere. Air, with helium, flows through this hole into the mass
spectrometer, where the helium concentration is measured and the leak rate is computed.

In inside-out techniques, the unit to be tested is evacuated and then pressurized with helium. An operator
moves the sniffer over the part and tests with the probe around suspected leak sites. The orifice
dimension establishes the probe flow and then the detection performance. Flow, QHe, pumping speed,
SHe, and partial pressure, PHe, are related as:

QHe = PHe · SHe

VTECH_leak_detection -24- Rev. 1 – 29 Mar. 06

The minimum leak rate measurable by a helium sniffer is the concentration of helium in the working
area. In the ideal case, the atmospheric concentration of helium is 5 ppm, so its partial pressure is 0.5 Pa
(3.7 µm Hg). Then, considering a standard pumping speed of 1 cm3/s, the sensitivity is:

QHe = PHe · SHe = 0.5 · 1 · 10-6 Pa · m3/s = 5·10-6 mbar · l/s

The slower the pumping speed, the higher the sensitivity will be, but the slower the helium will arrive at
the spectrometer and longer the delay time to get the measurement. The most commonly used pumping
speed, a trade-off between sensitivity and response time, is 1 cm3/s that, with a pipe 5 m (16.4 ft) length,
provides a delay time of about one second. As a consequence of this delay time, the sniffer’s moving
speed and its distance from the part to be tested are critical points. The nearer the probe is, the higher the
helium concentration entering the mass spectrometer and better the test quality. On the other hand, the
faster the detector movement, the smaller the pumped helium quantity is. Furthermore, the sensitivity of
this method is strongly limited by the background helium level and is not as good as that achievable with
others techniques based on mass spectrometers.

The helium sniffer leak testing method has the big advantage of determining leak location but, as shown
before, has some drawbacks. Helium sniffer leak detection implies manual operation, strongly operator
dependant, and the operator’s experience is a determining factor in the outcome of the testing procedure.
Helium disperses slowly into the atmosphere, so in the case of leaks, the background level increases
limiting the successive tests. In case of a big leak, the working area may become unusable for long time.
It is good practice to use a preliminary leak test, such as a pressure decay test, to detect gross leakages.
This test may be integrated in the same helium inside-out machine.

Leak detection sensitivity is related to the internal pressure, as shown in the following diagram. However,
the maximum usable pressure is limited by operator safety concerns. Unless using a robotized arm,
sniffing inside a protective hood is not feasible.

Fig. 15 – Inside – out helium sniffer dependency on internal pressure

VTECH_leak_detection -25- Rev. 1 – 29 Mar. 06

Helium is quite an expensive gas, and, as stated before, it is not convenient to spread it into the working
area, so is good practice to use helium recovery/reclaim stations to empty the unit at the end of leak
testing. The recovered helium can be reutilized for successive tests. Nevertheless, even using a recovery
system, some helium will be lost. A small quantity will be, in fact, dispersed in the test area and some
more will remain inside the circuit. Furthermore, during the recovery phase, helium can be contaminated
by air infiltration, moisture and oil. The storage tank must be checked to ensure that helium inside has
enough concentration and is pollution free so that leak testing is affordable. Helium concentration
restoration and tank cleaning procedures must be periodically executed.

7.6.7 Outside-in helium spraying

In the outside-in helium spraying technique, the sample to be tested is connected to a vacuum group and
to a mass spectrometer. The unit is evacuated and its surface is "probed" with a pointed jet of helium by a
suitable diffuser. Coming over a leak, the detector senses the helium entering from the leak allowing the
location and data about the leak size. This method is classified under vacuum test methods, even if it is a
special case. This technique can be used where it is needed to locate small leaks since its sensitivity is not
limited by the background helium level.

However, it is not free of limitations. This method, like sniffing, is operator dependant. Moreover, helium
is lighter than air; it tends upwards, so helium sprayed in the bottom of the unit may pass through a leak
in the upper part. Helium has a tendency to accumulate and then saturate the working area if it is not used
with good ventilation. In this condition, leak location becomes difficult. In case of a big leak, a lot of
helium reaches the mass spectrometer and a long time may be needed to reduce the helium level to a
value compatible with testing, but, during this time, operator can perform other tasks.

Since the unit to be tested is evacuated, its internal pressure is lower than atmospheric pressure.
Therefore, if the working condition requires a pressure inside, this leak testing method will stress the part
in the opposite way. This testing technique is, however, is widely used in research and in all applications
involving big plants which cannot be leak tested using other methods. It is not the most suitable solution

for testing in high productivity lines or series production manufacturing industries. It is important to note
that parts to be checked must be kept in an area free from helium contamination before leak testing. If
high pressure air is used for a preliminary pressure decay test, the high pressure air compressor inlet must
also be in a “fresh air” atmosphere.

7.6.8 Outside-in helium leak testing

In the outside-in helium leak testing technique, the part to be tested is covered with a suitable hood to
contain helium and connected to a vacuum group and mass spectrometer. The test consists of evacuating
the unit and flooding the hood with helium. Helium, in fact, due to its atomic characteristics, has a high
penetration capability. So a mass spectrometer can detect the helium leaked into the component through
cracks and porosities not detectable using other systems. The part is evacuated to less than 15 Pa (100 µm
Hg). The unit may be also subjected to a vacuum rise test to ensure that it is “clean” from water vapor
contamination, other non-condensable gases, and free of very large system leaks. After evacuation, the
internal circuit is connected to the mass spectrometer. If any helium has leaked into the circuit, the mass
spectrometer detects it. In this way, it is possible to quickly establish whether a sample leaks and to
establish the total leak rate. This method is able to detect leak rates up to 1.8·10-5 mbar · l/s, that is 2.5 g
(0.1 oz) of R134a refrigerant per year. The sensitivity of this method is related to the helium
concentration. The following picture shows the relationship between helium leak rate and refrigerant leak
considering 10% helium 90% air.

VTECH_leak_detection -26- Rev. 1 – 29 Mar. 06

 Fig. 16 – Outside – in helium leak rate

There are several practical realizations of this testing method. The test system can be designed with one
or multiple stations, usually two, for production rate requirements since multi-station machines allow
testing several parts, one for each station, simultaneously. The containment hood, appropriated to the
product sizes so as to completely cover the test unit, may be created in such as way to reduce helium
dispersion during loading and unloading operations. Advances in vacuum and helium technologies
provide improved sensitivity and faster test cycles even with low helium concentration, even values
typically used of 10% helium 90% air. In this case, the containment hood uses a helium ratio transducer
to monitor the mixture inside it. Recirculation fans and mixing ducts maintain the preset concentration of
helium to air uniformly distributed in the internal volume. The system will only replenish the helium
when needed. Adoption of this state of the art solution saves helium thereby reducing operation costs.

This leak testing method has some advantages. It is a final global leak test and this technique is very
useful on production lines where a test piece must be accepted or rejected. The test is fully automatic and
hands free. The total process time to leak test a part, regardless of its volume, is low, even less than 80
seconds for a medium size unit.

However, there are also some shortcomings; some precautions must be taken for the proper use of this
system. This method is a “go – no go” test and is for leak detection only. For the leak location an
additional system is required. If the test unit has a large leak, i.e. an over looked brazed joint, the escaping
helium, at this much volume and concentration, would render the system inoperable for long time. It is
convenient to use a preliminary leak test for identification of gross leakages. A pressure decay test can be
easily combined in the outside-in leak testing machine. This initial nitrogen filling, besides the
mechanical stress and leakage test, allows purging the circuit to be tested, lowering its humidity. On a
successful completion of the pressure decay test the component is ready for final test. At the end of cycle,
the part may be filled again with nitrogen, at a low pressure, to avoid helium to enter into its internals
when disconnected.

Since the unit to be tested is evacuated, its internal pressure is lower than atmospheric pressure.
Therefore, if the working condition requires a pressure inside, this leak testing method will stress the part
in the opposite way. This problem can be mitigated using a proper preliminary pressure decay test.
It is important to note that parts to be checked must be kept in an area free from helium contamination
before outside-in leak testing. If high-pressure air is used for a preliminary pressure decay test, the high-
pressure air compressor inlet must also be in a “fresh air” atmosphere. At the end of cycle, the unit may

VTECH_leak_detection -27- Rev. 1 – 29 Mar. 06

be filled with low-pressure nitrogen to avoid helium infiltration into the part when disconnected.

7.6.9 Inside-out helium vacuum chamber leak testing

In inside-out techniques, also known as “global hard-vacuum test”, the test configuration is reversed. The
component is placed inside an air tight chamber equipped with service hoses and a vacuum pumping
group with a mass spectrometer. The unit is connected to the service hoses, the chamber is closed, and
the part is evacuated and then pressurized with helium. The chamber is subsequently evacuated and, once
a suitable vacuum level is reached, the inlet valve of the leak detector is opened. The leak detector begins
to analyze the residual gas molecules present in the chamber. Helium molecules escaping from the
component are conveyed and measured in the leak detector. At the end of cycle, the vacuum chamber is
vented. The leak detector finds the leak and it gives the total measure. Referring to the complexity of the
test and the desired degree of automation, different test systems may be realized from the simplest to
more elaborate. As an example, a two chamber machine can be designed so that while in one chamber
the test cycle is in progress while in the other the part is being unloaded and loaded. In many industrial
applications, to reach the vacuum level inside the test chamber in acceptable times, an auxiliary pumping
group is needed. The dimensioning of the pumping group depends on different parameters, such as the
size of the expected leaks, the dimensions of the parts to test and the cycle time.

The vacuum chamber helium inside-out leak test method has some advantages. A sniffer is not used;
testing is automatic and not very operator dependant. The sensitivity achievable with this test is very
high. In laboratory applications it is not impossible to get 10-10 mbar · l/s. However, in practical
applications in the refrigeration industry, due to unit uncleanness, dirt accumulation inside the chamber
and oil and moisture contamination, the limit is considered 10-6 mbar · l/s. The following picture shows
the theoretical relationship among refrigerant leak rate, internal helium pressure and helium leak rate.

Fig. 17 – Vacuum chamber inside – out helium leak detection

Despite these advantages, vacuum chamber inside-out leak test method has several drawbacks. It is an
expensive test system. While the sniffing test requires only the leak detector and the ability of the
operator, the hard-vacuum test requires more complex and expensive equipment, but it enormously limits
the reliance on the human factor. This fact makes this test attractive in many industrial contexts, while the
sniffing test has its application in the maintenance and the analysis of defective parts.

VTECH_leak_detection -28- Rev. 1 – 29 Mar. 06

In design and realization of the vacuum chamber, particular care must be taken to avoid helium
contamination. In fact, if some helium remains on the chamber inner surface or in some internal
components, the background helium level will be high and the sensitivity will consequently be reduced.
Consequently, test equipment performance, especially its sensitivity, is strongly affected by vacuum
chamber clearness and contamination. To achieve the best performance from this machine, periodic
maintenance, cleaning and purging procedures are required. If the test unit has a big leak, large amounts
of helium would leak out of the test unit and into the vacuum chamber, the mass spectrometer and the
vacuum pumping station. The helium, at this much volume and concentration, would render the system
inoperable for long time.

To avoid this phenomenon, is good practice to use a preliminary leak test to detect big leaks. A pressure
decay test can be combined with the vacuum chamber inside-out leak test. This initial nitrogen filling,
besides the mechanical stress and leakage test, allows purging of the circuit to be tested, lowering its
humidity. On a successful completion of the pressure decay test the component is ready for final helium
test. At the end of cycle, the part may be filled again with nitrogen, even at a low pressure, in order to
purge its internal circuit from the helium residuals before vacuum chamber venting.

Unlike the sniffing test, the hard-vacuum test is a global method, that quantitatively defines the total leak
of the piece, but that does not determine its position. Other techniques are required for determining leak
location. Because helium is a quite expensive gas and to reduce the helium concentration it is not
acceptable to spread it in the working area, so is good practice to use helium recovery/reclaim stations to
empty the unit at the end of leak testing. The recovered helium can be reutilized for successive tests.
Helium recovery systems are costly equipments and, as already discussed above, they are not trouble
free.

7.6.10 Inside-out hydrogen sniffer detectors

As in the other inside-out techniques, the unit to be tested is evacuated and then pressurized with the
hydrogen/nitrogen mixture. An operator moves the sniffer over the part and tests with the probe around
suspected leak sites.

Comparing with helium sniffer leak testing, this method has the same difficulty of operator dependency,
but has many advantages. Hydrogen is ideal for leak testing. It is the lightest element, with higher
molecular speed and lower viscosity than any other gas. Therefore, it easily fills the unit to be tested,
mixes quickly with any gas, and is easily evacuated and dissipated. Since hydrogen mixes very quickly
with other gases, this method requires a low vacuum level compared with helium filling. As a result, the
vacuum group is very simple – a Venturi vacuum generator is enough. Hydrogen easily disperses into air,
so its concentration in the working area falls quickly to the background level even in case of big leaks.
In addition, hydrogen has the highest leakage rate of any other gas. Hydrogen detectors are based on
semiconductor sensors and, do not contain any moving parts, are robust for industrial use and completely
maintenance free. They are unaffected by the presence of other gases. Moreover, these sensors do not
require suction for functioning, so the probes can be used without worrying about dust. In addition, the
probe can be equipped with a protective cover that allows it to be used on wet objects. Hydrogen has a
very low background concentration (0.5 ppm) and the detectors are sensitive enough to detect up to 5·10-
7
mbar · l/s (or 0.5 g/y (0.018 oz/y) R134a refrigerant) using a mixture of 5% hydrogen 95% nitrogen, ten
times more sensitive respect to helium sniffing. This blend is standard industrial grade mix, inexpensive,
non–flammable (as for ISO10156 specification) easily available from industrial gases suppliers. The
following figure gives an idea about the relationship among tracer gas leak rates, internal test pressure
and refrigerant leak rates. Obviously the maximum internal pressure is limited by operator safety
concerns.

VTECH_leak_detection -29- Rev. 1 – 29 Mar. 06

 Fig. 18 – Hydrogen diffusion probe leak detection sensitivity at several internal pressure

There are some shortcomings or rather some suspicions in using hydrogen as a tracer gas. Hydrogen
escapes rapidly from a leak almost in vertical direction, so a leak can be precisely detected and located
only if the detector is directly above it. This characteristic causes someone to think that leak detection

with hydrogen is more difficult than with helium. However, even if a skilled operator has no difficulty in
managing this problem, detectors can be easily integrated in special designs, e.g. a funnel-shaped device,
to facilitate leak detection. Another common doubt of the hydrogen/nitrogen mixture is component
separation and hazards related to hydrogen. Even though the hydrogen/nitrogen mix is widely available,
big industries with high production lines think that in big cylinders hydrogen tends to accumulate in the
higher section of the bottle. This would cause working with different hydrogen concentrations during the
bottle life. On other hand, in-house mixing of hydrogen and nitrogen would assure the proper mixture,
but it is very hard to achieve due to the risks related to hydrogen handling.

Hydrogen leak testing is a manual operation but, with a proper sampling probe, it can also be automated.
Approaches similar to the accumulation method can be favorably applied. Although hydrogen quickly
disperses in air and, even in case of big leak, the working area is not saturated by hydrogen, it is
convenient to save tracer gas using a preliminary leak test for gross leakage identification. A pressure
decay test can be easily integrated in the hydrogen testing machine.

8. HOW TO CHOOSE THE TEST METHOD

The specification of the acceptable leak rate and the test method are the first parameters to take into
consideration in the plan of a product.

The questions to be answered are:

 Which type of total leak can damage the product?
 How many points (welding points, joints etc…) must be tested?
 How long must the product last?

The answer to these questions is usually established in terms of a volume or gas mass that flows in a
definite time (e.g. 10 g (0.35 oz) of R134a refrigerant in seven years). Sometimes answering this question
VTECH_leak_detection -30- Rev. 1 – 29 Mar. 06
is not exactly straightforward. For example, to define the total leak specifications of a refrigeration
system is simple, but to define the valid specifications for a single component and a single welding point
is much more complex.

Every leak testing methodology has advantages and disadvantages. For leaks comprised in the range 10-1
and 10-3 mbar · l/s all the test methods are able to recognize a leak. Besides sensitivity, in selecting the
proper testing method, many other parameters have to be considered. Among them are repeatability,
accuracy, report capability, operator dependency, cost of the equipment and the necessary work force
required to run it. In addition, the cost of the tracer gas has an important impact especially in case of
series controls. In some cases a gas recovery system, cutting down its consumption, should be
considered.

For clarification, in the following a sample application is treated.

Let 84 grams (3 oz) of R134a in 5 years, at internal pressure of 1800 kPa (261 psia) and room
temperature, be an acceptable leakage. This leak corresponds to 16.8 g/y (0.59 oz/y) of refrigerant

F = 84/5 = 16.8 g/y = 16.8 / (365 · 24 · 3600) = 5.33·10-7 g/s

The equivalent leak rate can be computed by the kinetic theory of perfect gas. Since the molar weight of
R134a is 102.03, the equivalent molar flow is:

Fm = F / M = 5.33·10-7 / 102 = 5.22·10-9 mol/s

It is known that a mole of a gas occupies a volume of 22.4·10-3 m3 (0.79 ft3) at atmospheric pressure and
0 ºC (32 ºF). Then the molar volume at room temperature and atmospheric pressure is:

Vm = 22.4·10-3 · 300 / 273.16 = 24.6·10-3 m3/mol

Let p the atmospheric pressure, the volumetric leak flow will be:

Q = Fm · Vm · p = 5.22·10-9 · 24.6·10-3 · 105 = 1.28·10-5 Pa · m3/s = 1.28·10-4 mbar · l/s

The internal pressure has effects on the leak rate. The flow through an orifice depends not only on the
pressure difference at the two sides of the orifice, but also on their absolute values. Let P1 and P2 the
absolute pressures, to both edges of an orifice, that cause the rate of Q1 leak, and P3 and P4 the absolute
pressures that cause, in the same orifice, the Q2 leak rate. Then the following relationship yields:

Q1 P1 2 - P2 2

Q2 P3 2 - P4 2

This equation allows computing the equivalent leakage when different pressure values are used. For
example, resuming the previous case, the leak of Q1 = 1.28·10-4 mbar · l/s from P1 = 1800 kPa (261 psia)
to P2 = 100 kPa (14.5 psia), corresponds to a flow rate Q2 from the atmospheric pressure to vacuum:

4 12 - 0
Q2  1.28 10  2 2  3.9 10 7 mbar · l/s
18 - 1

If the leak testing is executed in a vacuum chamber, pressurizing the unit to 100 kPa (14.5 psia), the limit
leak rate should be 3.93.9·10-7 mbar · l/s.

VTECH_leak_detection -31- Rev. 1 – 29 Mar. 06

The flow through a leak also depends on the fluid viscosity; the less the viscosity, the larger the flow rate.
Therefore, the flow Q1 of a medium of viscosity 1 is related to the flow Q2 of a medium of viscosity 2
by the relationship:

Q1  2

Q2  1

Returning to the previous example, considering that the tracer gas is helium, the R134a leak can be
converted in the equivalent of leak for helium, based on various the viscosity () of two gases

Q R134a  He

Q He  R134a

Since the viscosity of the R134a is 0.012 cP and that one of helium 0.0178 cP, it follows that the
permissible leak for helium is:

QHe = 3.9·10-7 · 0.012 / 0.0178 = 2.6·10-7 mbar · l/s

Even though these considerations are useful to identify the best suited testing techniques, they are not
precise. Several approximation and simplification errors are introduced in the various steps. The perfect
gas assumption is a useful approximation but, especially for refrigerants that should be more properly
considered vapors, is not exactly respected. In addition, there is no confidence that if a leak occurs at

1800 kPa (261 psia), this leak happens also to 100 kPa (14.5 psia), especially if the pressure acts in
opposite sense. Moreover, the flow computations are applicable accurately in known flow regimen, but,
obviously, this is not the case. In order to obviate these error sources, it is better to carry out the tests with
sensitivity ten times higher than that obtained with theory, to provide a wide safety margin.

9. CONCLUSION
The growing demand for components and systems with less and less acceptable losses seems the industry
trend due to several compelling market demands, such as economic requirements, environmental
protection specifications, safety constraints and quality production requirements. The result is stricter
quality controls for leak testing. Researchers, technicians, scientists, producers etc… working with
hermetically closed elements and vessels, vacuum or only tight seals have to become familiar with
measurements and location of leaks. However, this technical field is nearly unknown even in engineering
and in important project organizations. A brief analysis of some of the most commonly used leak
detection techniques, with particular attention to the refrigeration industry were presented. Every
methodology has advantages and disadvantages; the right choice is a trade-off between them and the
production requirements. For determining the best choice of test method to be used, it is necessary to
accurately consider all admitted leak limits and all the other factors, not only the technological
requirements, but also the corporate image, the regulation developments and the new requirements of the
market.

VTECH_leak_detection -32- Rev. 1 – 29 Mar. 06

 ANNEX A
Vacuum unit conversion chart

1 atm = 760 Torr = 1013 mbar = 101325 Pa = 14.7 psia = 760 mm Hg

Leak rate unit conversion table

At constant P of 100 kPa

std cc/s mbar l/s Torr l/s Pa m3 /s

1 std cc/s 1 0.75 0.1
1 mbar l/s 1 0.75 0.1
1 Torr l/s 1,3 1,3 0.13

1 Pa m3 /s 10 10 7.5

Approximate Leak Rate Equivalents for Refrigerant Leakage

Gas leak rate Conversion Helium leak rate

(mbar l/s) factor
1 g/y R22 = 9.00 x 10-6 x 0.66 5.9 X 10-6 mbar l/s
1 g/y R12 = 6.43 x 10-6 x 0.64 4.1 X 10-6 mbar l/s
1 g/y R134a = 7.63 x 10-6 x 0.71 5.4 X 10-6 mbar l/s
1 g/y R600a = 1.34 x 10-5 x 0.40 5.3 X 10-6 mbar l/s
1 g/y SF6 = 5.33 x 10-6 x 0.77 4.1 X 10-6 mbar l/s

Approximate Leak Rate Equivalents for R134a Refrigerant Leakage

Gas leak rate Helium leak rate Bubble immersion

(R134a) (mbar · l/s) (time to form one
bubble)
300 g/y (9.6 oz/y) 1.8 · 10-3 13 seconds
100 g/y (3.2 oz/y) 1.2 · 10-3 45 seconds
30 g/y (1 oz/y) 1.5 · 10-4 130 seconds
10 g/y (0.3 oz/y) 8 · 10-5 270 seconds
3 g/y (0.1 oz/y) 1.2 · 10-5 23 minutes
1 g/y (0.03 oz/y) 5.4 · 10-6 180 minutes
0.5 g/y (0.02 oz/y) 1.5 · 10-6 210 minutes

VTECH_leak_detection -33- Rev. 1 – 29 Mar. 06

 ANNEX B

Gas Properties

Viscosity and Molar Weight of Some Gases

Gas Formula Viscosity (Pa · s) at 15 ºC Molecular weight (g/mol)

Air 18.3 · 10-6 28.8
Nitrogen N2 17.6 · 10 -6
28.0
Oxygen O2 20.0 · 10 -6
32
Argon Ar 21.9 · 10 -6
40
Hydrogen H2 8.7 · 10-6
2.0
5% H2 95% N2 17.2 · 10 -6
26.7
10% H2 90% N2 16.7 · 10 -6
25.4
Helium He 19.4 · 10 -6
4.0
10% He 90% N2 18.4 · 10 -6

Carbon dioxide CO2 14.5 · 10-6 44

R134a CH2FCF3 11.61 · 10-6 102.03
R290 C3H8 8.114 · 10-6 44.1
R404A 11.87· 10-6 97.6
R407C 12.31 · 10-6 86.2
R410A 13.19 · 10-6 72.59
R600a CH(CH3)2CH3 7.397 · 10-6 58.12

Conversion of Leak Rate in Laminar Flow

Relative leak rate

Gas
(helium) multiply by
Air 1.077
Nitrogen 1.12
Argon 0.883
Hydrogen 2.23
Neon 0.626
Helium 1
Water vapor 2.09

VTECH_leak_detection -34- Rev. 1 – 29 Mar. 06

 Conversion of Leak Rate in Molecular Flow

Relative leak rate

Gas
(helium) multiply by
Air 0.374
Nitrogen 0.374
Argon 0.316
Hydrogen 1.41
Neon 0.447
Helium 1
Water vapor 0.469
Carbon dioxide 0.301
R134a 0.198
SF6 0.167

Diffusion Coefficient

Gas Diffusion coefficient (mm2/s)

Nitrogen 17.5
Oxygen 17.5
Argon 14.7
Hydrogen 67.1
Helium 69.7
Water vapor 21.9
Carbon dioxide 13.4

VTECH_leak_detection -35- Rev. 1 – 29 Mar. 06

 ANNEX C

Equation of State

An equation of state is a constitutive equation describing the state of a system subject to a given set of
conditions. An equation of state is useful to depict the properties and the evolution of such a system.
Referring to physical systems, and specifically a thermodynamic system, it provides a mathematical
relationship among two or more state functions associated with the matter, such as temperature (T),
pressure (P), volume (V), etc…. The most common use of an equation of state is to predict the state of
gases and liquids or solids, and their mixtures.

Ideal Gas Law

One of the simplest and best known state equations is the ideal gas law:

p V  n  R  T

where p is the gas pressure (Pa or N/m2), V is the volume occupied by the gas (m3), T is the gas absolute
temperature (K), n is the gas quantity in number of moles, and R is the universal gas constant:

R  8.314472 J/ (mol · K)
This equation is roughly accurate for gases at low pressures and high temperatures, and becomes
increasingly inaccurate at higher pressures and lower temperatures. Furthermore, it fails to predict phase
transition, like condensation from gas to liquid and evaporation from liquid to gas. Therefore, many other
equations of state, more accurate, have been developed. At present there is no single state equation that
accurately predicts the properties of all gases and vapors under all conditions.

Van der Waals Equation of State

The Van der Waals state equation (1873) was one of the first gas law working better then the ideal gas
law:

a V
(p 2
)  (  b)  R  T
(V n) n

where p is the gas pressure (Pa or N/m2), V is the volume occupied by the gas (m3), T is the gas absolute
temperature (K), n is the gas quantity in number of moles, R is the universal gas constant, a and b are
constants depending on the specific material. They can be computed from the critical properties as:

VC
a  3  PC VC b
2

3
where Pc, Vc, Tc are the pressure, volume and temperature at the critical point.
Now this equation is considered obsolete, its agreement with experimental data is limited. Modern
equations are only slightly complex but much more accurate.

Redlich - Kwong Equation of State

The Redlich-Kwong state equation (1949) it is still used due to its simple form:

VTECH_leak_detection -36- Rev. 1 – 29 Mar. 06

 R T a
p 
(V n)  b T  (V n)  (V n  b)

where p is the gas pressure (Pa or N/m2), V is the volume occupied by the gas (m3), T is the gas absolute
temperature (K), n is the gas quantity in number of moles, R is the universal gas constant, a and b are
constants depending on the specific material. They can be computed from the critical properties as:

0.42748  R 2  TC 0.08664  R  TC
2.5
a b
PC PC
where Pc, Vc, Tc are the pressure, volume and temperature at the critical point.

Peng-Robinson Equation of State

The Peng-Robinson state equation (1976) provides reasonable accuracy near critical points, particularly
for the compressibility factor computation. Moreover, it allow one to predict liquid density.

R T a 
p 
(V n)  b (V n)  2  b  (V n)  b 2
2

where p is the gas pressure (Pa or N/m2), V is the volume occupied by the gas (m3), T is the gas absolute
temperature (K), n is the gas quantity in number of moles, R is the universal gas constant, , a and b are
constants depending on the specific material. They can be computed from the critical properties as:

0.42748  R 2  TC 0.08664  R  TC
2.5
a b
PC PC
T 0.5 2
  (1  (0.37464  1.54226    0.26992   )  (1  ( ) ))
TC
where Pc, Vc, Tc are the pressure, volume and temperature at the critical point and  is the acentric factor
for the specific fluid. The term  is a temperature dependent function which takes into account the
attractive forces between molecules.

VTECH_leak_detection -37- Rev. 1 – 29 Mar. 06

 ANNEX D

Mass Spectrometer

This instrument allows revealing a wide range of leaks with a high facility of uses. It is possible to carry
out measurements either quantitatively either qualitatively. Qualitative measures are easier and suitable
for quality control applications.

The detection technique is based on the separation of a gas, from the other gases, in vacuum due to an
ionization process. In very low pressure (or vacuum) the molecules of rest gasses are transformed in ions
by electron impact. The ionized particles of different mass to charge ratios (q/m) are separated by a
magnetic field and their ions are collected in different position. These ionic current are related to each gas
concentration, therefore it is possible to state the partial pressures of present gases. The low pressure (less
than 2·10-2 Pa or 1.5·10-1 µm Hg) required for operation of the mass spectrometers is produced by an
integrated high vacuum pump system. The auxiliary vacuum pump required for rough pumping the
tested equipment can either be incorporated or be attached via suitable connection.

Two types of mass spectrometers exist, the simplest has a narrow passing band with single gas (normally
helium), the second has a wide passing band and programming function to select among a gases ranges
(helium, R12, R22, R134a, R600a, etc.) in order to recognize and to measure the leak for every type of
gas.

Mass spectrometers used as leak detectors are the most sensitive instruments for leak testing.

VTECH_leak_detection -38- Rev. 1 – 29 Mar. 06

 ANNEX E

References

Leak detection and location are widespread treated in literature. Here is reported a short list of internet
sites or papers where to go deep into this issue:

[1] Introduction to Helium Mass Spectrometer Leak Detection – Lexington, USA.

[2] Leak Detection Methods and Defining the Sizes of Leaks - 4th International Conference of Slovenian
Society for Nondestructive Testing - Ljubljana, Slovenia

[3] Advanced Leak Test Methods, Austin Weber, Assembly Magazine, Issue Date: 01/01/2001

[4] Leak Testing with Hydrogen, Claes Nylander, Assembly Magazine, Issue Date: 02/01/2005

[5] ANSI 7.60 "Leakage rate testing of contaminant structures”, American Nuclear Society, La Grange
Park, Illinois

[6] Boiler & Pressure Vessel Code Section V, Leak Testing, American Society of Mechanic Engineers,
New York, N.Y.

[7] Annual ASTM Standards, Part II, American Society for Testing & Materials, Philadelphia, PA.

[8] Leak Testing Standards, American Vacuum Society, New York, N.Y.

[9] Bulletin #1005 "Leak Testing Large Pressure Vessels". "Bubble Testing Process Specification", King,
Cecil Dr. - American Gas & Chemical Co., Ltd.

[10] Leakage Testing Handbook; NASA Contractor Report; NASA CR952, Marr, J. William, - NASA,
Washington, D. C. 1968

[11] Leak Testing Volume of ASNT Handbook, McMaster, American Society for Nondestructive
Testing, Columbus, OH, 1980.

VTECH_leak_detection -39- Rev. 1 – 29 Mar. 06

VTech provides complete turn-key systems or individual equipment for your specific production line
requirements. Pre-evacuation Stations, High Sensitivity Automatic Leak Detection Systems, Leak
Location and Repair Devices, Refrigerant Supply equipment, Refrigerant Charging Machines,
Refrigerant Recovery Systems, Run-Test equipment, Process Software, as well as repair stations and
personnel training. We can provide conveyors, fixtures, leak test fittings, etc.. Our extensive expertise can
assist you in finding the answers for your manufacturing needs. We have the solution for you. Contact us.
We are ready to help you.

VTECH_leak_detection -40- Rev. 1 – 29 Mar. 06

 This document is property of:

Without the written authorization, it may not be used for manufacturing the object
represented, transmitted to third parties, or reproduced in any way.

The proprietary company will protect its rights with the full rigor the law.

Information in this document is subject to change without notice and does not
represent a commitment.

VTech reserves the right to make modifications to the equipment herein described
for the purpose of improving the machine.

The software and/or the database described in this document are furnished under a
license agreement or non disclosure agreement.

VTECH_leak_detection -41- Rev. 1 – 29 Mar. 06