Highly sensitive non-vacuum leak detection method for the automotive industry

Pressure and bubble testing are not accurate enough

Until now, most containers for liquids in cars and motorcycles have been tested for leaks in a water tank or by the pressure decay method. Although these two methods are inexpensive to test, they can only detect leaks up to 10-2 to 10-3 mbar l/s. In addition, due to the surface tension of water, small bubbles cannot form at all, so the bubble test in water cannot detect smaller leaks at all. At the same time, the test range for air conditioning components or fuel tanks with valves is 10-5 mbar l/s, but the test range for injection valves remains at 10-4 mbar l/s. Therefore, traditional leak testing methods are not suitable for the automotive industry at all. In addition, since some of the test parts are wet, this method cannot be used to test all parts, and as a pure visual task, the test results are completely dependent on the ability and concentration of the person performing the test.

When using the pressure decay method (or pressure drop test), air is introduced into the test part at a specific pressure. If there is a leak, the pressure will decrease and the difference can be detected. In theory it is very simple, but in practice you have to deal with all sorts of variables. This is because pressure changes are not necessarily caused by leaks. On the one hand, the pressure is also affected by changes in the elasticity of the test component itself. When air enters the test component quickly, the test component expands and then gradually contracts for testing. Since the measured difference depends on pressure and volume, you have to wait until the test component has regained its original volume (stabilization period) during the test. The greater the difference between the original volume and the pressure when the air enters, the longer it takes for the test component to reach the stabilization period. In production, the upper limit of the test component volume is about 5 liters.

If the stabilization period is too long, the test method is not cost-effective. In addition, when air enters the plastic test component, the creep behavior must be taken into account (the plastic part of the test component deforms due to the load). After the test, the volume of the test component will increase to varying degrees, depending on the molecular structure of the plastic, and the pressure will also decay. In this case, it is almost impossible to reliably measure low leak rates. In addition, the pressure will vary greatly with temperature. When the test pressure is 5 bar, the test volume is 1 liter, and the test time is 30 seconds, a virtual leak rate of 6*10-2 mbar l/s will be generated for every 0.1°C drop in temperature. However, if the temperature rises by 0.1°C, even if the leak rate is 6*10-2 mbar l/s, it will not be reflected on the test component. This is because, although the test pressure increases with the change in temperature, this increase in pressure is simultaneously offset by the pressure decay caused by the leak. Many of the parts needed to manufacture cars are directly from the production line. Due to the production process, the temperature of these parts is generally not low.

If these parts are cooled during the test process, the test results will be unrealistic, which is not conducive to effective control of the leak rate. We can of course offset this adverse effect by testing the test parts after they have reached a suitable stable period, but the large amount of time wasted is also a factor affecting the cost. Alternatively, we can set up a cooling zone, but this means a significant increase in investment. The temperature of the surrounding environment is also a major problem. Even changes in sunlight can cause temperature fluctuations, which can lead to changes in test results. Generally speaking, high temperatures are the most worthy of attention. Finally, air humidity can also change the test results, because the vapor pressure of water vapor in the air affects the measurement and produces differential detection. All of the above are the main characteristics of low-cost production equipment in Asia, so setting up a test area in Asia is extremely complex and challenging, and doing so will also eliminate our cost advantages. In this case, it is difficult to guarantee the same measurement results, and as the leak rate becomes smaller or the volume becomes larger, the probability of the same measurement data will continue to decrease. Therefore, for many requirements of mass production in the automotive industry, the pressure drop method can no longer guarantee the accuracy of its measurement data, or it can be said that this method has lost its effectiveness in some areas.

Vacuum leak detection is expensive

. We can also choose other detection methods: such as using helium as the detection gas to detect leaks using a mass spectrometer. Although the smallest leak rate (10-11 mbar l/s) can be detected with a minimum of helium, this must be carried out under high vacuum conditions. In this regard, this method is too expensive. The vacuum box must be highly sealed and equipped with various high-performance pumps to produce the vacuum effect, which makes the production and operation costs too high. This method is very effective, even the smallest leaks can be detected, the test takes only a few seconds, and the system is so advanced that it can maintain the best test state after several rounds of testing, even when the helium concentration continues to increase.

However, this system is not the best solution to the above problems, because the test is extremely sensitive, up to thousands of factors can cause the test results to change, and its acquisition and operation costs are too high, which greatly offsets its advantages. In addition, evacuating the test chamber to test the component means that the pressure difference is usually as high as 1 bar. Since such high pressures are not usually considered when designing the parts, many plastic parts may be damaged. In this case, vacuuming the test chamber and the test part at the same time is equivalent to filling the test part with helium at low pressure. In addition, for large test parts such as car radiators and other parts, the size of the vacuum chamber must also be changed accordingly. Therefore, on the one hand, the pumping time of the vacuum treatment of the test chamber and the entire test time will be extended; on the other hand, the installation project is larger, resulting in higher investment costs. In today's automotive industry, which is transforming to low cost and high efficiency, neither the extension of time nor the increase in cost is applicable.

Alternative leak detection without vacuum / mixed helium leak detection

In the automotive industry, testing with air has reached its limits in many respects, but testing under vacuum is too expensive. There is indeed a range between 10-2 and 10-5 mbar l/s where high-performance, fully automatic testing methods can be used. Today, leak detection with helium or hydrogen in an accumulation tank at atmospheric pressure (accumulation method) has become the most economical solution to fill this range (see Figure 1). Hydrogen, more precisely a synthetic gas with a hydrogen content of 5%, can also be used for leak detection like helium. As a test gas, it is more reasonably priced (in the United States, it is only 1/3 of the price of helium, in China this figure is 1/10). However, testing with this gas does not maintain the same sensitivity, and for smaller leak rates such as 10-3 mbar l/s, the test time is about 5 minutes (11 seconds for helium leak detection) with a net volume of 10 liters of the test component.

Figure 1: At atmospheric pressure, helium leak detection can fill the area between 10-2 and 10-5 mbar l/s. (Ideal: 10-6 mbar l/s)

The hydrogen leak detection method is closer to the pressure drop method in terms of detection, but it does not have the same weaknesses in terms of elasticity and temperature. Helium is also suitable for leak detection with a leak rate of 10-4 to 10-5 mbar l/s. Since the detection is at normal pressure, a mass spectrometer cannot be used for this detection method, so a highly sensitive sensor is required to measure the increasing helium concentration. These sensors are patented by Wise Technology and are only used in the Infracore system. In the detection chamber, helium is introduced into the detection component through the test gas connection. If there is a leak, the test gas will escape into the accumulation box through the leak.

A fan ensures that the helium is evenly distributed in the chamber. In this way, accurate measurements can be made without knowing the leak location. The sensor detects the content of the test gas in the atmosphere. In the Infracore T-Guard system, the sensor consists of an evacuated glass tube with a helium-permeable quartz membrane on the top. This quartz membrane is like a sponge and only molecular helium can pass through it. The change in helium concentration in the glass tube can be measured by the change in current in the auxiliary pressure measuring device. (See Figure 2)

Figure 2: The sensor is evacuated glass tube with a helium-permeable quartz membrane on top.

Using this method, the sensor solution can accurately distinguish whether the concentration has increased by 25 ppb or 0.025 ppm, and the detection can reliably detect leaks within 10-6 mbar l/s. In production, for a 5-liter free volume test chamber (free volume is the test chamber volume minus the test component volume), this method can detect a leak rate of 1*10-4 mbar l/s in about 30 seconds. For a test chamber with a volume of 1 liter, it only takes 11 seconds to detect such a leak rate. Including the startup set-up time, the detection time for each test component is only about 16 seconds, and 225 test components can be detected per hour. (See Figure 3) [page]

Figure 3: Relationship between the leak rate and net volume of the test component during helium leak testing at atmospheric pressure

Other special applications of this test method include testing the tightness of exhaust gas recirculation systems inside and outside, with a leak rate of 2 sccm, which is about 3.3 * 10-2 mbar l/s. In a test chamber with a volume of 40 liters and a mixed gas (containing 10% helium) as the test gas, the test time is 11 seconds, and the complete cycle time for each test part is 45 seconds. In addition to the high speed, this test method has other advantages, such as the fact that it can test plastic test parts. The test part volume increases after the gas is introduced, but this does not affect the measuring system. In addition, temperature, humidity and elasticity have no influence. In order to check whether the system can detect the leak rate limit of the test part, the test part can be placed in different positions in the test chamber to further check whether the helium concentration in the test chamber is consistent.

Sensors and systems

are components of leak detection systems. Sensitivity is not the only requirement for the measuring system. Easy integration, individual settings and low-maintenance operation are also required. In addition, to meet the needs of factory engineering, the inspection system must also be designed to be simple and compact, so that the system can be well connected with hydraulic parts and electronic components, and can be flexibly used in multiple inspection modes. (See Figure 4)

Figure 4: T-Guard can work in a simple test chamber at atmospheric pressure, without the need for a complex high vacuum chamber and vacuum pump.

The advantages of this accumulation method are obvious. First of all, there is no need to vacuum the standard atmosphere, so the use of turbomolecular pumps, vacuum boxes with extremely high airtightness requirements, and high-sensitivity mass spectrometers can be avoided. This greatly simplifies the design of the measurement system and can better control or reduce the procurement and maintenance costs of the entire system. In addition, this method can detect reliable and reproducible test results, even if household plastic boxes are used. In addition, even if the test parts are very large, high temperature or humid, the measured values ​​of this mixed helium leak detection are still highly reproducible, and the test parts do not need to be cooled or dried for a long time before testing. From an economic point of view, fast and fully automatic testing is the most critical part of the production line, so this helium leak detection method that does not require vacuum conditions is an ideal means for the automotive parts manufacturing industry, which can meet the dual needs of the automotive manufacturing industry for quality and cost.

Figure 5: Mixed helium leak testing with an accumulation tank has been used to test torque converters in the automotive industry.

Figure 6: The inspection box needs to be as close to the inspection part as possible. The limited free volume makes the inspection time shorter and more productive.

INFICON is the world's leading developer, manufacturer and supplier of leak detection instruments. Its leak detectors are widely used in industrial processes with high difficulty in production and quality control. INFICON's main customers include manufacturers and service providers of refrigeration and air-conditioning equipment, automobile manufacturers and automotive parts suppliers, the semiconductor industry and leak detection system integrators. Almost all major automobile manufacturers and parts suppliers in the world are INFICON's customers, including airbags, air conditioners and components, fuel tanks, injector systems, various fluid container manufacturers, etc. (end)