Pressure Relief Valves are the last line of defense in any process system. When they are designed for liquid service, the performance of most Pressure Relief Valves depends heavily on the behavior of an incompressible fluid, showing; rapid pressure transmission, modulating lift, and steady discharge.

However, some repair and service providers and even end-users still test these valves using Gas.., compressed air or nitrogen gas, assuming it’s a harmless simplification. And, while we understand the practical advantages of testing with gas, in reality, gas testing of liquid-service valves may produce false results, mechanical damage, and serious safety hazards.

Same design, different performance

Generally all pressure relief valves look the same from the outside. They have the same building style but in case of liquid service the difference lies inside. The trim, the internal assembly, existing of the nozzle, the disc, guides and the spring distinguishes a valve for liquid service from one for gas or steam service. Basically the difference in the design of the trim parts is ‘compressibility’ of the process fluid, or in this case you can say..; the lack of it.

This can be further explained as follows;

• Liquids are incompressible, so when a valve opens suddenly, the escaping liquid can create an extreme hydraulic shock or even chatter. For the people that don’t know the phenomenon; chatter is a rapid, repeated opening and closing (vibration) around the set pressure point. In liquid service, chatter is particularly problematic because, as said, liquids are nearly incompressible, so any small change in flow or pressure is transmitted instantaneously and forcefully. The typical liquid trim avoids this effect by modulating the lift – the valve opens just enough to relieve pressure smoothly.
• Gases are compressible, so they require the valve to snap open to full lift immediately to handle large expanding volumes safely. Hence, a pop-action trim! So, even if the valve body looks identical for gas and liquid service, the trim is engineered differently to match the physical behavior of the process fluid. This clearly explains why it is important, and sometimes even critical, that Pressure Relief Valves intended for liquid service are also tested with liquid.

Set pressure definition

So, how is this done..? Just as with testing on gas, the answer lies in the definition of the set pressure – the key criterion of testing pressure relief valves.

The most commonly used definition of the set pressure for liquid service valves is the ‘inlet pressure at which there is a first steady stream of liquid flow from the valve’. Under the increasing pressure the valve begins to open gradually. Before real lift, there may be a few intermittent drops or small dribbles.  The “set pressure” is not when the first drop appears, but when a continuous, steady discharge of liquid can be observed. It’s the point where flow is stable and sustained – the valve is functionally open. With this clear definition in mind, we can turn to the technology of the test bench. Achieving a stable and precise control of the test pressure and maintaining a constant flow without pressure spikes and pulses ‘cannot’ be accomplished with a high pressure pump. No, instead, for this application, we use the gas-over-liquid principle, which is actually quite straightforward – particularly when the test bench is already equipped with a gas testing system with test vessel for conventional valves used in gas service.

After the valve has been properly clamped on the clamping system, and prior to the start of the test procedure, both the valve and a vertically mounted, cigar-shaped test vessel are filled with water. Once the entire system is filled, the test begins by introducing gas – typically air or nitrogen gas – via the top of the test vessel. The gas pressure is then increased gradually and steadily, pressing on the water column. The pressure rises until the valve reaches its set pressure and begins to discharge water. When this discharge transitions into a steady, continuous stream, the corresponding pressure is recorded as the valve’s set pressure. The process does not necessarily need to be time-consuming; however, maintaining a stable flow inherently requires a constant supply. This can only be effectively ensured through the correctly configured gas-over-liquid test system, particularly when the system is intended for operation across a broad pressure range.

Conclusion

Testing a liquid-service Pressure Relief Valve with gas is potentially unsafe, non-representative, and non-compliant with international standards.
Gas behaves entirely differently from liquids, and the results of such a test cannot be used to certify the valve or verify its correct operation in service.

Therefore, for any Pressure Relief Valve intended for liquid service;

• Always test with a non-compressible fluid (preferably water with an anti-corrosion inhibitor).   
• Use a gas-over-liquid systems
• And, above all, always carefully follow OEM instructions and settings

In case you have any doubts or questions about testing liquid service Pressure Relief Valves, please feel free to contact us.