Fast leakage test Unitem
Basic concept for leak testing
There are no perfectly tight objects, but you can achieve a very high degree of tightness specified by safety standards (which is especially important in devices using explosive or toxic gases) or product standards specified by designers. Leak is a feature of every tank (chamber), regardless of whether it is a windshield washer fluid reservoir or high-vacuum apparatus used in the semiconductor industry. However, from the application point of view, it is important to answer the question what is the acceptable value of the leak at which the tank still performs its intended functions. Below this value we determine the rejection level of defective products. Leakage belongs to this category of concepts whose measure is always greater than zero.
There are three types of leaks: real, apparent and resulting from gas permeation through the walls of the tank (chamber). Real leaks are associated with holes, gaps or channels connecting areas at different pressures and allowing gas to flow between these areas. Such leaks occur most frequently in the vicinity of connections and are usually the result of structural, technological or assembly errors, and in rare cases – hidden material defects.
Apparent leaks are associated with the presence of gases adsorbed on the internal walls of the tank (chamber), gases trapped in micropores, residual lubricants, solvents and other organic impurities. Apparent leaks play an important role in the high vacuum technique (and higher – pressure below ~ 10-3 mbar), i.e. in technologies requiring high purity of the conducted process (e.g. deposition of metal vapors).
However, leaks due to gas permeating through the tank walls can generally be neglected.
In the art, the leakage value is often given in units of gas flow rate. The most commonly used units are: [W = Pa • m3 / s], [mbar • l / s], [mol / s], and [Tr • l / s].
Leak detection methods can be divided into vacuum and pressure methods depending on whether the gas pressure inside the tested object is lower or higher than atmospheric pressure.
Some of these methods allow the location of leaks, while others only allow measurement of the resulting leakage intensity.
The smallest leak rate detected by a given method determines the sensitivity of this method.
Unitem uses the following methods in its leak testing stations:
The most popular pressure leak test method used in the automotive industry is the method of slow pressure changes. A significant limitation of this method is the lack of indication of the leak location. Another disadvantage is the very long measurement times (at least tens of seconds) for details that deform significantly when compressed air is applied. Examples include such elements as oil sumps, tanks for various fluids made of plastic or rubber hoses. Motivated by aforementioned drawbacks , Unitem has developed thermal imaging based method. It enables the indication of the leak location and is insensitive to pressure changes inside the tested element due to its deformation (gas pressure increase), which enables a significant reduction in the measurement time. FLT thermal imaging leak testers provide the traditional information OK and NOK. Optionally, it is possible to view the thermograms in order to locate the leak. Examples of leak test results are shown below.
Fig. 1. Measurement of tightness of fittings in hoses installed in the cooling system of passenger cars a) view of the tested object, b) leakage in the AR01 area.
Fig. 2. Measurement of the tightness of the hose clamp a) view of the correctly made connection b) leak detected (black points on the infrared image)
Fig. 3. Plastic element analysis example: a) polymer tank mounted in cars, b) thermogram and c) example temperature characteristics
Fig. 4. a) Thermal image of a local leak (red point) in a stainless steel diaphragm and b) an exemplary temperature characteristic during pressure shock changes
Fig. 5. The body of the heater a) used in coffee machines and b) disassembled view, c) leakage (red and yellow area on the thermogram) and temperature characteristics
Fig. 6. Exemplary temperature characteristic and leak point thermogram (NOK)
Fig. 7. Schematic representation of the acoustic method.
Acoustic leak detection (Fig. 7) uses sonic or ultrasonic energy generated by an expanding gas leak. This method is very simple and fast, but has low sensitivity – leaks up to 10-2 mbar · l/s. Additional coverage of the leakage place with foam or water allows the sensitivity range to be increased to 10-4 mbar · l/s. This method is a vacuum and pressure method.
The air bubble method
The air bubble method (Fig. 8) is one of the most common pressure methods. There are two variants of this method. In the first one, the test object is pumped to a sufficiently high pressure and it is immersed in water and thanks to the formation of air bubbles it is possible to locate the place of leakage. In the second variant, the object is covered with a thin layer of an aqueous soap or detergent solution and after inflating it is observed in which places soap bubbles are formed. The sensitivity of this method for soap bubbles is about 10-4 mbar · l/s, while for the method with immersion of the tested object it is about 10-3 mbar · l/s. In the method with immersion of the examined object, credibility has a great impact on, among others lighting of the water reservoir, the degree of turbidity of the water, location of leaks and hence the sensitivity of this method is one order worse.
Both variants of the described method, although seemingly simple, often create many disadvantages. In the case of the immersion method, trouble increases with the size and weight of the objects tested, the more so that after the test they must be thoroughly drained. In the alternative, removing the previously applied solution is often quite laborious and thus expensive.
Fig. 8. Leak testing using the air bubble method (with a dedicated agent).
The method of slow pressure changes (pressure drop method, differential method)
Depending on the value of pressure generated inside the tested object, the method of slow pressure changes is classified as vacuum or pressure methods. In both variants it allows to assess the tightness of the object, however, without the location of the leak.
The sensitivity of this method in the vacuum variant depends on the measuring range and accuracy of the vacuum gauge used. For ionization vacuum gauge (pmin = 10-6 Pa), recorded leaks reach the level of 10-6 mbar · l/s. The principle of the pressure variant differs only in that the tested object is inflated to a pressure usually in the order of several atmospheres and after disconnecting the compressor the pressure drop rate is measured – Δp / Δt resulting from the occurrence of real leaks (pressure drop method). The sensitivity of the pressure gauges used is usually around 100 Pa (the lowest measurement resolutions are 0.01 Pa for the most expensive measuring devices) and hence the sensitivity of this variant is several orders of magnitude lower than in the case of the vacuum variant.
Several factors influence the difference in measured pressure drop (ΔP = P2 – P1), such as:
presence of leakage
volume difference ΔV
temperature difference ΔT
Therefore, when using the pressure drop method, the impact of temperature change ΔT and the impact of volume change ΔV on the measured pressure drop ΔP, should be taken into account. When using this method on the production line, the impact of the above physical quantities on the final result of the measurement should be taken into account. To eliminate the impact of these factors, a differential pressure drop method should be used. In other words, the differential method compensates for volume differences ΔV and temperature differences ΔT.
In the differential method, one of the basic laws used is Mariotte’s law (Boyle’s law), which for ideal gases takes the form:
P V = n R T
where: P [Pa] – pressure, V [m3] volume, n – amount of moles (amount of matter), R – constant for ideal gases (R = 8,31 J/mol·K), T [K] – temperature.
After taking into account the impact of temperature changes and volume changes, we get:
(P+ΔP)(V+ΔV) = n R (T+ΔT).
The volume V consists of: the volume of the test item, the volume of tubes used to connect the test tank, the volume of tubes and fittings inside the measuring device.
Fig. 10. Leak testing by pressure drop using a reference element
The figure shows the idea of differential measurement, in which we use a pattern (an element with an acceptable level of tightness) made of the same material with the same dimensions (volume, structure) as the tested element. Both elements are placed in the same climatic conditions (temperature, pressure). Due to this approach to measurement, the ΔT and ΔV values are the same for the tested and reference element and cancel each other out without participating in the measurement indicated by the leakage measurement device.
Methods using tracer gas
Methods involving the introduction of selected gases (markers) into the tested tanks and tracking their flow through leaks using detectors. Gas markers should meet a number of conditions: they should not be chemically active, toxic, they should not occur (in significant amounts) in the Earth’s atmosphere, they should not be explosive / flammable and should have the highest possible value of diffusion coefficient to other gases. The only gas meeting the above conditions is helium and it is often used. Sometimes, hydrogen (usually a mixture of up to 5% hydrogen with nitrogen, which is not explosive) or one of freons (now very rarely) is also used as gas markers. Hydrogen does not meet some of the above requirements, but its diffusion coefficient is of high value and is therefore used. The leak test method using gas markers can be used as vacuum or pressure method.
Vacuum methods using tracer gas
The vacuum method using tracer gas requires the mass spectrometer. The spectrometer’s design is selected so that it only detects gas tracer molecules. A stream of this gas blows through all chamber joints in turn. If the gas stream is directed to the leak, the gas enters the apparatus and then goes to the spectrometer, which registers its presence. In this method, the measurement of the leakage intensity takes some time (depending on the tank volume and the rate of helium pumping from this reservoir). The location of each subsequent leak requires pumping out the gas marker (e.g. helium) introduced during the previous test into the tested vacuum chamber. The time it takes to remove helium is called ventilation time. The response time and ventilation time characterize the inertia of the vacuum system when changes occur and must be taken into account in the leak detection procedures.
The professional leak detector is equipped with a quadrupole mass filter and pumping system with a turbomolecular pump, which makes this system relatively expensive.
The sensitivity of the described method is very high and is usually 10-11 mbar · l/s.
Fig. 11. The ASM 340 helium leak detector from Pfeiffer Vacuum
The test object is pressurized (most often with a mixture of helium or other tracer gas with air) to pressure exceeding atmospheric pressure, usually to about 2 – 5 bar. Then the detector is applied successively to the tested surfaces of the tank, in particular to all kinds of connections to detect leaks. When the tracer gas reaches the detector sensor through a leak, it signals its presence. This method requires a detector with a relatively high degree of complexity:
helium leak detector with mass spectrometer
a vacuum system to maintain a sufficiently low pressure in the spectrometer (often a turbomolecular pump + an initial vacuum pump)
valves controlling individual stages of the measuring cycle
electronic control and measurement system
The sensitivity of this method for the best solutions is estimated at 10-7 mbar · l/s. It should be remembered that the pressure method does not allow a precise estimation of the leakage intensity.
Such a method is also used in a significantly simplified variant, in which we get information about the location of the leak, while the size of the leak can be drawn based on the value of the gas tag given in ppm (parts per million). Detectors of this type are often not equipped with suction cups, and the cost of instrumentation is incomparably lower than in the mass spectrometer method. Unfortunately, in this variant we have no information about the gas flow rate through the leak.
Sensitivity range of the method[mbar · l/s]
Coated with water or foam
The air bubble method
Slow pressure changes