TUV Findings

Plant construction and the petrochemical industry have special requirements when it comes to safe, fast and simple leak testing. An important reason are the technical requirements in the field of clean air (Volatile Organic Compounds - VOC).

Depending on the type leak detection can better and cheaper solutions be achieved on the basis of the hydrogen leak detection method than with other tracer gases. Recently, a hydrogen-containing test gas was used for various applications for leak testing at a petrochemical company. A cheap standard mixture consisting of 95% nitrogen and 5% hydrogen (forming gas) was used. A gas mixture that is environmentally friendly, non-explosive, non-toxic, non-flammable and non-corrosive. The leak detection measurements were performed with selective leak detection equipment from Sensistor Technologies.

The leak detection took place with a hydrogen leak detector, which only reacts to hydrogen. The background concentration of hydrogen in the air is very low at 0,5 ppm, so that leak values ​​of up to 5×10-7 mbarl/s can be demonstrated. During localization, the measuring head is simply guided to the leak. There are no maintenance-intensive parts such as pumps or filters in the leak detection equipment. The leak detectors are compact, handy, do not require a power supply and are therefore also very suitable for mobile use. The leak detection results of components such as flange couplings, fittings and heat exchangers are displayed just as well as if the test were carried out with helium.

Hydrogen as test gas

Due to its physical properties, hydrogen is extremely suitable as a trace gas. The molecular speed is higher and the viscosity lower than any other gas. Hydrogen can be introduced into components and installations much faster than other test gases, it mixes with air faster than other gases, has a higher leak rate and is much easier to vent.

For leak detection and tightness testing, the hydrogen method uses a standard mixture of nitrogen and hydrogen (forming gas) in a ratio of 5% hydrogen (H2) and 95% nitrogen (N2). This mixture is classified as non-flammable in accordance with international standards (ISO 10156). As a result, there are no safety risks when using the gas mixture. The mixture is mainly used as an inert gas in welding and soldering work. It is available from stock from suppliers of technical gases (eg Lindegas, Air Liquide or Air Products) in gas bottles (10 liters or 50 litres) of 200 or 300 bar and costs only a fraction of helium.

Recently, the price of helium has risen significantly again due to increased demand. Since supplies of helium worldwide are limited, a further price increase can be expected. Hydrogen, on the other hand, is being offered at an increasingly affordable price due to its increasing use as an energy source. Unlike helium, hydrogen is a replaceable raw material, which is especially important for an environmental audit according to ISO 14001. In addition, it is non-toxic and non-corrosive. The differences in physical properties between helium and hydrogen also lead to differences and advantages in practice. Hydrogen spreads significantly faster. This effect contributes to achieving a uniform concentration in the interior of the part to be inspected after only a short time. In general, the part to be checked does not need to be emptied beforehand. Problems caused by an increased background concentration (“pollution”) are much less with hydrogen. In addition, hydrogen does not adhere to surfaces and can be deaerated much more quickly. In summary, hydrogen can be called the ideal leak detection tracer gas.

Hydrogen leak detection device

The hydrogen leak detection method requires a hydrogen leak detection device / leak detector with a corresponding measuring range and a high selectivity.
Sensistor/Adixen's H2000C is based on a semiconductor sensor, which is constructed in the same way as a field effect transistor. Hydrogen molecules adhere to the surface of the sensor element and decompose into hydrogen ions (protons). These can then pass through the grid structure of the sensor element and thus cause a change in the electric field. The generated electrical signal is processed and thus displayed on the leak detection device. The sensor element reacts selectively to hydrogen, the indication limit is determined by the natural background concentration of hydrogen in the air (0,5 ppm). Detection takes place directly in the measurement zone by means of diffusion. This eliminates the need for pumps to suck the gas into the device. The measuring zone is only connected to the leak detection device via a cable. Advantage: No filters and other maintenance-sensitive parts are required. Another positive effect of direct measurement is a very short response time that is not dependent on the cable length and an equally short recovery time. The leak detection devices are user-friendly and the current status of the sensor element is displayed during each calibration. Alternatively, calibration can also be carried out via a test leak. Due to the battery-powered variant of the device, it is possible to carry out leak tests independently of the mains power supply.

The leak detection method

Depending on the leak test, two main methods are followed in practice with the hydrogen method, via localization (sniffing) or a test chamber.
Localization consists of searching for leaks. Sometimes it is known that a leak is present before the measurements are started. It is then important to locate the leak as accurately as possible with as little effort as possible. Strictly speaking, it is not entirely true for localization or 'sniffing checks' (see DIN EN 1779, method B4) that test gas must be sucked up before the measurement. (This is not necessary with our unique leak detectors.) When using 95/5 forming gas, the indication limit for the leak detection device is 5×10-7 mbarl/s. In hydrogen technology – with sufficient safety measures – higher concentrations or pure hydrogen are also used in some cases. The indication limit also shifts to lower leak values ​​based on the concentration.

As with the localization, in the test chamber the part to be checked is filled with test gas, after which the increase in the hydrogen concentration in the test chamber is measured. The duration of the leak detection measurement depends on the allowable leak rate and the size of the test chamber. Depending on the application, a flexible liner or a fixed chamber can be used as a test chamber (method B3 according to DIN EN 1779).

The leak tests mentioned in the introduction were carried out during the periodic inspection of a hydrocracker. As part of planned maintenance work, the leak tightness of several hundred flange connections has been checked. For that purpose, test chambers (control parts) were placed around the flanges, taped with a slightly hydrogen-permeable adhesive tape. The measurement took place via a mobile, battery-powered hydrogen leak detection device from Sensistor (now Inficon)

Comparative measurements between helium and hydrogen leak detection

The TÜV Süddeutschland has carried out a number of tests to determine the leakage values ​​at seals. Two seals that are commonly used in practice have been chosen for this purpose, a ball valve and a flange coupling. Comparative measurements were performed with helium and a mass spectrometer as well as with hydrogen and a hydrogen leak detection device. The measured leakage values ​​are shown in the table. The leakage value is displayed according to VDI 2440 in a unit related to the seal length. It can be seen that the readings are independent of the trace gas used and the leak detector. The deviations between the measured values ​​are caused by the measurement accuracy of the devices used and can therefore be neglected. This result prompted TÜV Süddeutschland to certify the Sensistor H2000 hydrogen leak detection device for measurements based on the technical requirements in the field of clean air (TA-Luft).