Download

Sealed or Leaking? 

Sealing is a relativ term. There are no absolutely sealed parts. The requirement of sealing of a component always relates to the conditions of the specific application, therefore, the leak test method, the test pressure and the permissible leakage rate must be specified.

At this point the first doubts may already arise. Permissible leakage rate? – the workpieces must be sealed! Component parts of internal combustion engines for example must be water tight or oil tight, and others must have no fuel or gas leaks.

In this case the leak test must, at the earliest possible moment within a production process, give information on wheter a component is liable to leak or not under the subsequent conditions of use – the further machining or assembly of a leaking workpiece can the be avoided, this is called “Scrap Refining”.

In automatic leak testers either air or a special test gas is used as the test medium, even when the workpieces must be liquid-tight. The lower viscosity of the gaseous test medium compared with the liquid working medium allows substantially higher accuracy of testing. Through a porous area from which hardly any liquid escapes, the gaseous test medium can escape and be measured.

The existance and definition of a quantity of escaping air (leakage rate) ist a condition of automatic leak testing. A leakage rate of “Zero” cannot be detected by any test method not even with underwater visual testing. The requirement that no bubble must become visible is true of even the smallest bubble that might appear during a given period of observation.

An Example

In an underwater test, air escapes from the test piece, rising through the water on reaching a bubble size of 1 mm. If this takes place every 30 seconds, a leakage rate of 0.001 cm³/min can be calculated from this. Supposing this test piece is a car tyre (Interior volume about 40 l), on account  of this leak the tyre pressure would drop only 0.1 bar within 10 years. Is this tyre to be described as sealed or leaking? If the bubble with diameter of 1 mm rises every 3 seconds, this corresponds to a leakage rate of 0.01 cm³/min. The above-mentioned pressure drop in the tyre then occurs within a year. Is this car tyre now considered to be sealed or already leaking?

The permissible leaking rate of refrigerator compressors for example ist 1 g freon/year at 16 bars pressurisation. This corresponds to a leakage rate of 8.4 x 10^-7 standard cm³/min.

When asking the question “Sealed or Leaking” it is therefore a matter of defining the permissible leakage rate V_perm1. In this case the workpiece with a

                   Leakage rate <        V_perm1 = Sealed

                   Leakage rate >=      V_perm1 = Leaking

The underwater visual test is fairly accurate test method – with careful execution by the test and with plenty of opportunity to observe the workpiece. If the leakage rates attainable as a result are (as in the above mentioned example) to be detected by an automatic leak tester, the use of a mass spectrometer test method is necessary. The underwater visual test is however not only gladly used due to its accuracy, it is also comparatively cheap and – a very important advantage – the point of leakage is located at the same time. The disadvantages of this test method are – it depends on the operator (low test reliability), it takes a long time, and the workpieces get wet – the reasons for using automatic leak testers.

 
                    Common Leak Test methods in mass production

In the automatic leak testing of mass-produced parts, leak testing with air has become widespread. The reasons for using this test method are:

• High test reliability.
• Objective and quantified statement of quality.
• Short testing times.
• Low investment and running costs.
• Sufficient accuracy.
• Gentle treatment of workpieces.
• Easy integration in interlinked production processes.
• Easy calibration.

Automatic leak testing of mass-produced parts will in future further increase in significance in connection with the requirements from the fields of

• ISO 9000.
• Product liability.
• Obligation of documentation.
• Environmental conservation.
• Consumer protection.
• Workplace design.

The Test Specification

The leak test specification includes two data: test pressure and leakage rate.

The crucial question is: how high may the leakage rate be in order to ensure that no liquid escapes from a workpiece under operating conditions?

The permissible leakage rate is different for each application. It depends in many factors, such as e.g. the material of the workpiecem, the structural circumstances and – as already mentioned above – the subsequent stresses in the operative state.

The ratio between air and liquid leakage rates cannot be calculated in the lower range of leakage rates (liquid leakage rate almost zero). For higher leakage rates, the ratio of viscosity’s can be used as an approximation.

In this case, factors such as the wall thickness of the workpiece, the surface properties and the formation of porosity (blowhole, crack) also play a critical role.

A leakage rate value is always referred to a given test pressure. The pressurisation of the workpiece in leak testing is as a rule selected such that the subsequent operative state is simulated. Here it is a question of firstly the degree of test pressure and secondly the direction of testing.

In fixing the test pressure, the question arises of the correlation between test pressure and leakage rate. With high porosity the leakage rate behaves more or less in proportion to the test pressure, and with low porosity, the ration of leakage rate to test pressure becomes lower.

Small leaks are not constant. They vary with time, above all by underwater visual testing. If the result of an automatic dry test is to be verified by an underwater visual test, then this effect must be taken into account.

Moreover, economic aspects are critical in establishing the permissible leakage rate in industrial production. Testing is to be only as accurate as necessary - not as accurate as possible. If the set maximum leakage rate is lower than necessary, the proportion of rejects increases, and hence the costs rise too.

The final leakage rate permitted for production must be determined empirically. In some cases, empirical values are available or recommendations can be made.

In leak test methods with air as test medium (measurement of pressure or flow rate), the leakage rate is usually given in standard cm³/min, and in leak test using a special test gas (e.g. helium) indication in
mbar 1/sec is common.

1 standard cm³/min = 60 mbar 1/sec.

At this time it should be pointed out that 1 standard cm³/min is a gas quantity, which in the physical standard state (0° C and 1013 mbar) fills 1 cm³. 1 standard cm³ is not a spatial measure, but a gas quantity expressed in spatial dimensions. Indication in standard cm³/min is thus a mass flow rate and not a volumetric flow rate. For leak testing, this means that strictly speaking if the leakage rate is indicated in standard cm³/min either testing is carried out by the mass flow rate method or the pressure test signal must be corrected for the effect of barometric pressure and ambient temperature. If this is not done, the leakage rate should be given correctly in cm³/min or ml/min. The difference between standard state and test state is on average about 10 %.

Examples of permitted air leakage rates for car engine parts