Common Mistakes in PSV Sizing: Lessons Learnt and Case Studies From Years of Operation

Pressure safety valves are often the last barrier against loss of containment, yet design verification keeps finding valves that were sized wrongly. Drawing on years of revalidation work and incident reviews at upstream facilities, the authors catalogue the recurring errors, show what each one does to valve capacity, and give a checklist for catching them. Many of these flaws persist in older assets designed before current API guidance.

Key results

Assuming a fixed flare-tip pressure drop instead of modelling actual flow behaviour can cut valve performance by 50–60%
Ignoring sudden liquid swell in an upstream separator during a gas blowdown case left a valve 30% undersized, which had to be retrofitted after commissioning
A glycol dehydration PSV was sized with a rich-glycol latent heat of 681 Btu/lb, but under fire the water boils off first and pure glycol has only about 170–180 Btu/lb, so the valve was badly undersized
A lower-flow fire case with different fluid properties produced a worse inlet pressure drop than the governing blocked-outlet case and caused valve chatter. Every relief case, not just the governing one, must be checked for pressure drop

Approach

Reviewed past design verifications, field revalidations and incident investigations against API 520 Parts I and II, API 521 and ASME Section VIII
Re-modelled relief scenarios with process and relief-system simulators to check the original sizing assumptions
Grouped mistakes by category: using only the governing case for inlet and outlet pressure drop, over-claiming double jeopardy, latent check-valve failure, multiphase relief loads, unbalanced heat in columns, fire latent heat, missed global scenarios, and class-break placement
Distilled the findings into best practices and a field-validation checklist for PSV sizing

The problem

Recurring mistakes in PSV relief-scenario definition, relief-load calculation and piping design leave relief valves undersized or unstable.

Tools & methods

Aspen HYSYSAspen FlarenetHTRI Xchanger SuiteiPRSMAPI 520 / 521ASME Section VIII

Summary written for this site from the published paper. For figures, data and the full method, read the paper on OnePetro.