Pressure relief is the last line of defence when everything else on a plant has failed, and getting it wrong carries consequences that no operator wants to face. The design work is governed by well established guidance in API 520 and API 521 for sizing and disposal, supported by the pressure vessel codes that set the allowable accumulation. This checklist walks through the questions an engineer should be able to answer with confidence before a relief system is signed off. It is not a substitute for the standards, but it is a reminder of the traps that catch even experienced teams.
Have you identified every credible scenario?
Sizing starts with scenarios, not with equipment. The relieving case is the one that produces the largest required relief load, and finding it means working through every credible cause of overpressure. Common scenarios include the following.
- External fire raising the vapour generation rate in a vessel
- Blocked outlet while a pump or compressor keeps delivering
- Thermal expansion of trapped liquid between two closed valves
- Control valve failure in the open or closed position
- Loss of cooling, reflux or heat removal in a column or reactor
- Tube rupture in an exchanger passing high pressure to the low side
- Chemical reaction or runaway generating gas or heat
The governing case is not always the obvious one, and a fire case can be smaller than a blocked outlet on a system with a large pump. Documenting why each scenario was included or excluded protects the design against later doubt.
Is the relieving pressure and accumulation correct?
The set pressure, the allowable overpressure and the accumulation all follow from the vessel code and the service. A single device protecting against a non fire case is typically allowed a lower accumulation than the multiple device or fire case, and mixing these up leads to an undersized valve. Confirm the maximum allowable working pressure of the protected equipment, the set point, and the accumulation permitted for the governing scenario before any sizing calculation begins.
Have you chosen the right device?
A conventional spring loaded valve, a balanced bellows valve, a pilot operated valve and a rupture disc each suit different services. Back pressure is the usual deciding factor, since a conventional valve loses performance when the superimposed or built up back pressure grows, while a balanced or pilot operated design tolerates far more. For fouling, corrosive or two phase services the choice narrows further. Where a rupture disc sits upstream of a valve, the combination capacity and the risk of a pinhole leak both need thought.
Is the flow regime understood?
Sizing equations differ for vapour, liquid, steam and two phase flow, and the two phase case is where many errors hide. A relieving system that flashes as pressure drops, or a reactive system that produces both gas and vapour, may need the methods developed for two phase relief rather than a simple vapour calculation. Assuming single phase vapour when the real relief is two phase can undersize the device badly. When the phase is uncertain, the conservative path is to check the two phase case explicitly.
Does the inlet line respect the pressure drop limit?
The non recoverable pressure drop in the inlet line to a relief valve should stay within the recognised limit, commonly taken as three percent of set pressure, so the valve does not chatter. Chatter destroys valves and defeats the protection at the worst possible moment. Short, generously sized inlet piping with minimal fittings is the goal, and a line that fails the check needs to be reworked rather than accepted with a note.
Is the discharge system adequate?
The tail pipe, header and any flare or catch system all impose back pressure that feeds back into the valve selection and sizing. API 521 gives the framework for disposal, and the built up back pressure during relief must be consistent with the device chosen. A shared header can carry the discharge of one valve into the outlet of another, so simultaneous relief cases deserve a check. The final destination must handle the fluid safely, whether that is a flare, a scrubber, a knockout drum or a safe location for atmospheric venting.
Have you addressed reaction forces and thermal effects?
A discharging relief system generates significant reaction forces and can swing or vibrate if the piping is not supported for the transient. Cold discharge can embrittle carbon steel and thermal growth can stress connections. The mechanical design of the relief piping is part of the safety case, not an afterthought for the piping group, and the two disciplines need to agree on loads and supports.
Is the whole system documented and maintainable?
A relief device that cannot be isolated safely for testing will drift out of certification, so any isolation valves need administrative control such as a car seal or lock. The relief study, the scenario basis, the sizing calculation and the datasheet should form a traceable package that a future engineer can pick up. Inspection and testing intervals belong in the maintenance system so the protection stays proven over the life of the plant.
A pressure relief design is a chain, and it is only as strong as its weakest link from scenario to disposal. If you would like a second pair of eyes on a relief study or help revalidating an ageing system, a relief systems specialist can review the basis and the calculations with you.