Leak Testing in production lines is done with various sensors. One of the most important things and reason for misunderstandings is the difference between the limits of these sensors and the limits of the resulting leak test machines. The detectable leak rates with leak testing machines are very often 1.000 to 100.000 times less compared to the sensor in the laboratory. The detectable leak rate of a leak test machine is more difficult to calculate because of different influence in different applications. In the following we will look to a few methods concerning their use in production lines.
Commonly used sensors for leak testing
| Description | Operator water bath testing (bubble test) | Pressure difference | Hydrogen (H2LT) | Helium-Atmosphere (HELT-A) | Ultrasonic- Leak Tester |
Helium- Vakuum (HELT-V) |
| Sensor | Eye | Pressure sensor | Semi- conductor sensor |
e.g. Wise Technology™ | Ultrasound transducer | Mass- spectrometer |
 |
 |
 |
 |
 |
 |
 |
| What is detected / measured? | Gas/Air bubble approx. 1mm diameter | Pressure 1 – 0,1Pa |
Current | Current 2·10-10A at 5ppm He |
Gas/Air bubble < 0,1mm diameter |
Current 10-15A |
| Detection limit of sensor | 2,6·10-5 mbar·l/s at 20s |
10-4 mbar·l/s at 1Pa, 0,1l volume and 10s |
10-6 mbar·l/s |
10-6 mbar·l/s |
10-8 mbar·l/s in 60s |
10-11 mbar·l/s |
We find tables as the following not very helpful. Quite the contrary: if you don’t look carefully, such tables are misleading. In many cases the detection limit of the sensor is mixed up with the detectable leak rate of a leak test machine.
| What can be reached by the leak test machine under production conditions | ||||||||||||
| Leak rate in mbar·l/s | ||||||||||||
| Test method |
1 |
10-1 |
10-2 |
10-3 |
10-4 |
10-5 |
10-6 |
10-7 |
10-8 |
10-9 |
10-10 |
10-11 |
| Operator water bath | ||||||||||||
| Pressure difference | ||||||||||||
| Helium Atmosphere | ||||||||||||
| Hydrogen | ||||||||||||
| Helium Vacuum | ||||||||||||
| Ultrasound gas bubble detection | ||||||||||||
The important question is not: What can the sensor detect?
but: What can the corresponding leak tester find?
In the following we only talk about automatic leak testing, as mentioned above, with just one exception: operator water bath. This is because this method is still the most common all over the world.
| Sensor: |
What is detected / measured? | Detection limit of sensor | |
| Eye | |
Gas/Air bubble approx. 1mm diameter | 2,6·10-5 mbar·l/s at 20s |
Method
Part to be tested is pressurized and immersed into the water. Bubbles produced by a leak are detected by an operator looking to the water bath.
Test gas
Mainly Air, sometimes Nitrogen
Detection limit of sensor
Up to 10-5 mbar·l/s s (test in experiments but under optimised conditions, very difficult do detect this all thru a 8h shift).
Detectable leak rate in production
In dispute, but around 10-2-10-3 mbar·l/s.
Characterisation
| Operator independent |
– – –
|
Test result completely depend on operators decision. Strong influence from illumination, contrast and dirt in the water |
| Invest costs |
+ +
|
Mainly simple water bath |
| Part dry after testing |
– –
|
Part to be tested must be immersed into a fluid |
| Operating and maintenance costs |
+ + +
|
Depending on labour rates |
| Measuring time |
+
|
Mainly short because the operator can watch over a long time with high concentration |
| Volume dependence |
+ +
|
In principle no, but for big parts it is difficult for the operator to watch the whole area permanently |
| Temperature dependence |
+ + +
|
No temperature influence |
| Influence of a change in volume |
+ + +
|
No influence |
| Localisation |
+ + +
|
Direct, fast localisation of leak by the operator |
| Test gas |
+ + +
|
Air (sometimes Nitrogen) |
| Coarse leak problem |
+ + +
|
No |
| Sensor: |
What is detected / measured? | Detection limit of sensor | |
| Pressure sensor | |
Pressure 1 – 0,1Pa | 10-4 mbar·l/s at 1Pa, 0,1l volume and 10s |
Method
Part to be tested and reference volume is pressurised. The leakage is detected in an indirect way by measuring the pressure difference between the part to be tested and the reference volume in case of a leak.
Test gas
Air, sometimes Nitrogen.
Detection limit of sensor
Pressure differences in the range of 0,1 – 1Pa
Detectable leak rates in production
Up to 10-2 mbar·l/s (very strongly depending on the test volume)
Characterisation
| Operator independent |
+ + +
|
Yes |
| Invest costs |
+ + +
|
Mostly the cheapest test method (if it does not work reliable, a waste of money!) |
| Part dry |
+ + +
|
Yes |
| Operating- and maintenance costs |
+ +
|
Periodic calibration required |
| Measuring time |
+ +
|
Mainly short, because long measuring times rise the probability of temperature influence |
| Volume dependence |
– – –
|
The same leak cause completely different pressure differences. For a big volume there is just a small change. For a very small volume there is a much higher change in pressure |
| Temperature dependence |
– – –
|
Typically the temperature influence is much bigger then the detection limit of the pressure sensor. In addition the temperature influence rise with the test pressure |
| Influence of a change in volume |
– – –
|
If the test volume is changing during the measurement this causes also a change of pressure |
| Localisation |
– – –
|
No localisation |
| Test gas |
+ + +
|
Air, sometimes Nitrogen |
| Coarse leak problem |
+ + +
|
Yes, when used to completely closed parts (pressure increase in chamber) |
| Sensor: |
What is detected / measured? | Detection limit of sensor | |
| Semiconductor sensor | |
Current | 10-6 mbar·l/s |
Method
Part to be tested must be put in a chamber, evacuated (no high vacuum required) and pressurised with test gas mixture, typically 5% Hydrogen and 95% Nitrogen. If there is a leakage the concentration of the test gas in the measuring chamber will rise. The rise of concentration is measured by the semiconductor sensor.
Test gas
Forming gas (Hydrogen-Nitrogen-mixture, 5% Nitrogen )
Detection limit of sensor
10-6 mbar·l/s
Detectable leak rate in production
More or less in the range around 10-2mbar·l/s (depending on the volume in the test chamber)
Characterisation
| Operator independent |
+ + +
|
Yes |
| Invest costs |
+
|
No vacuum chamber required (normal chamber can be used) but part has to be evacuated before filling with the test gas to get homogeneous distribution of test gas |
| Part dry |
+ + +
|
Yes |
| Operating- and maintenance costs |
+
|
Periodic calibration required, maintenance of vacuum pump(s), costs for test gas low |
| Measuring time |
–
|
Strongly depending on the chamber volume. For big volumes the increase of test gas concentration is very low. |
| Volume dependence |
– – –
|
See measuring time. |
| Temperature dependence |
+ + +
|
No influence under normal environment conditions |
| Influence of a change in volume |
+ + +
|
No influence |
| Localisation |
– – –
|
No localisation |
| Test gas |
–
|
Forming gas (Hydrogen-Nitrogen mixture) |
| Coarse leak problem |
– – –
|
Contamination of chamber and environment due to coarse leaks |
| Sensor: |
What is detected / measured? | Detection limit of sensor | |
| e.g. Wise Technology™ | |
Current 2·10-10A at 5ppm He | 10-6 mbar·l/s |
Method
Part to be tested must be put in a chamber, evacuated (no high vacuum required) and pressurised with test gas (mixture). If there is a leakage the concentration of the test gas in the measuring chamber will rise. The rise of concentration is measured by the Helium sensor.
Test gas
Helium, Helium-Air mixture, He-Nitrogen-mixture
Detection limit of sensor
10-6 mbar·l/s
Detectable leak rate in production
More ore less in the range around of 10-2mbar·l/s (strongly depending on chamber volume)
Characterisation
| Operator independent |
+ + +
|
Yes |
| Invest costs |
+
|
No vacuum chamber required (normal chamber can be used) but part has to be evacuated before filling with the test gas to get homogeneous distribution of test gas |
| Part dry |
+ + +
|
Yes |
| Operating- and maintenance costs |
+
|
Periodic calibration required, maintenance of vacuum pump(s), test gas |
| Measuring time |
–
|
Strongly depending on the chamber volume. For big volumes the increase of test gas concentration is very low |
| Volume dependence |
– – –
|
See measuring time |
| Temperature dependence |
+ + +
|
No influence under normal environment conditions |
| Influence of a change in volume |
+ + +
|
No influence |
| Localisation |
– – –
|
No localisation |
| Test gas |
– –
|
Helium, Helium mixture |
| Coarse leak problem |
– – –
|
Contamination of chamber and environment due to coarse leaks |
| Sensor: |
What is detected / measured? | Detection limit of sensor | |
| Ultrasound transducer | |
Gas/Air bubble< 0,1mm diameter | 10-8 mbar·l/s in 60s |
Method
Part to be tested is pressurised and immersed into the water. Raising bubbles are detected and localised by an ultrasonic system.
Test gas
Air, sometimes Nitrogen
Detection limit of sensor
Up to 10-8 mbar·l/s
Detectable leak rates in production
Up to 10-5 mbar·l/s
Characterisation
| Operator independent |
+ + +
|
Yes |
| Invest costs |
– –
|
Ultrasound technology requires more sophisticated water baths and some design requirements for machine building |
| Part dry |
– –
|
Part must be immersed into a liquid |
| Operating- and maintenance costs |
+ + +
|
Very low |
| Measuring time |
+
|
Only according to leak rate |
| Volume dependence |
+ + +
|
No influence of volume |
| Temperature dependence |
+ + +
|
No influence of temperature |
| Influence of a change in volume |
+ + +
|
No influence |
| Localisation |
+ + +
|
Direct, fast localisation of leak bubbles by ultrasonic system (additional immediately analyses by operator possible) |
| Test gas |
+ + +
|
Air, sometimes Nitrogen |
| Coarse leak problem |
+ + +
|
No problem |
| Sensor: |
What is detected / measured? | Detection limit of sensor | |
| Mass-spectrometer | |
Current 10-15 A |
<1·10-11 mbar·l/s |
Method
Part to be tested and chamber must be evacuated. Part is filled with test gas. If due to leakage the test gas expands into the chamber, the mass spectrometer detects the Helium.
Test gas
Helium, Helium-Air mixture, He-Nitrogen mixture.
Detection limit of sensor
10-11 mbar·l/s.
Detectable leak rates in production
Up to 10-7 mbar·l/s.
Characterisation
| Operator independent |
+ + +
|
Yes |
| Invest costs |
– – –
|
Vacuum chamber required, dimensions of pumps must be good enough to compensate environmental variations (e.g.. high humidity) |
| Part dry |
+ + +
|
Yes |
| Operating- and maintenance costs |
– – –
|
Periodic calibration required, maintenance for vacuum technology, test gas costs. |
| Measuring time |
+
|
Measuring time itself mainly short. For big chamber volume the time to reach the required vacuum can be high |
| Volume dependence |
+
|
See measuring time |
| Temperature dependence |
+ + +
|
No influence |
| Influence of a change in volume |
+ + +
|
No influence |
| Localisation |
– – –
|
No localisation |
| Test gas |
– –
|
Helium, Helium mixture |
| Coarse leak problem |
– – –
|
Contamination of chamber and environment due to coarse leaks |
Our leak testing methods
> Ultrasound Bubble Detection
> Helium-Leak-Tester
> Hydrogen-Leak-Tester
> Pressure Difference
> Contact form
> Service & support
> Technology Centre
> Workshops
Leak rate
> Definition
> Calculation
> Conversion
Machine Building Applications in Ultrasound Computer Aided Measuring Systems Electric and Acoustic Sensor Technology
