Leakage rate and pressure loss

Last edit: 03/08/2023


Technical standards often require external leakage test on components to be used in a gas plant, or even on the plant itself; however, the measurement of gas leaks is not always easily feasible. So how can we run this test?


Leakage test is essential for each of the components that make up a gas plant: each gas leak increases the risk of fire and explosion. To help manufacturers and users in risk management, product standards state the maximum permitted leakage rates for each component. We can find an example in the EN 13611, which concerns safety and control devices for burners and equipment with gas and/or liquid fuels. In this case, the recommendation is:

[7.2.1] “gas controls shall be leak tight, in accordance with the leakage rates given in Table 3”

Not only the individual components that have leakage rates limits: entire gas plants also have recommendations regarding tightness. For example, ISO 13577-3 (Industrial furnaces and associated processing equipment – safety – Generation and use of protective and reactive atmosphere gases) states:

[4.2.7] […] The external leakage rate shall not give rise to dangerous condition, combustible and or toxic, in the foreseen circumstances of the equipment or installation. […] It is generally agreed that an external leak rate of 1≈dm3(n)/h for gas or 1≈cm3(n)/h for methanol will not create a dangerous condition in typical ventilated industrial installation.”

From a practical point of view, measuring leak rates in a large plant can be extremely complex; even considering a single component, such as a valve, being able to quantify exactly the leakage rate value is not easy.
There is, however, an easier method for quantifying losses, i.e. measuring pressure loss. Once every opening of the gas plant have been closed and the internal pressure is raised to the test value, the gas leaks generate a loss in internal pressure, a loss easily measurable through a pressure gauge. Once the pressure loss has been measured, it can be related to the leakage rate via the appropriate formula.

EN 13611 [Annex C.2] itself illustrates how to apply this method, giving both the test conditions and the formula:



  • qL is the leakage rate [cm3/h]
  • Vg is the total volume of the control under test and the test apparatus [cm3]
  • p’abs is the absolute pressure at the beginning of the test [kPa]
  • p’’abs is the absolute pressure at the end of the test [kPa].

The formula is derived from the general gas equation, refers to a test lasting 5 minutes under standard conditions. 

However, a 5 minute test may be not long enough: EN 13577, for example, put the minimum duration of the test at 30 minutes for gas plant, and even longer test can be necessary to generate a pressure variation high enough to be measurable. The formula can then be extended to the following form:



  • Δt duration of test [s].

Using this formula, it is also possible to derive the minimum duration time of the test, knowing the sensitivity of the pressure loss measuring instrument. In this way, the tester doesn’t have to use high sensitivity tools, but can simply extend the duration of the test.



  • Δtmin minimum duration of the test to detect leakage rates above the highest value permitted [s]
  • Vg is the total volume of the control under test and the test apparatus [cm3]
  • Δpmin lowest pressure variation detectable by the instrument [kPa]
  • qL-max highest leak rate permitted by the standard [cm3/h].

If a pressure loss is not measured after the minimum time Δtmin, the plant tightness complies with the standard.


The leakage test is one of the most common tests in the world of gas plants, both for individual components and for the entire plant. One of the easiest ways to get the leakage rates is to derive it from the pressure loss they generate. Through the appropriate formula and a measurement by pressure gauge it is therefore possible to obtain the gas leaks and compare them with the limit value given by the standards. 


Safety in Collaborative Robotics
There is no “Collaborative Robot”. That is one of the first statements you hear from people working in Collaborative Robotics. The reason is because...