Rever flow of Double Check Valves on discharge side of Pump and SV-Capacity on suction side 1

[PDHPP]

Dear All,
The interpretation of the new API-521 regarding the failure rate of two check valves in series (as 4.4.9.3) is leading to missunderstanding and need your technical support
Foe example for the clean boiler feed water and high pressure BFW-Pump (50 Barg. discharge press.), asssuming both check vlaves are properly maintained and inspected (spare pump is running):
a) In the past: No leak for inspected check valves (of different type!)
or
b) New: 10% of the flow (as 4.4.9.3.3.)
Which scenario a) or b) shouild be used for capacity of the SV on suction side ?
Thanks in advance

 

[don1980]

There's been a lot of debate and wide disagreement within the industry on how to assess check valve risks. I think that the changes that you see in API 521 underscore this point. The key thing to recognize is that the guidance in API 521 is just that - it's guidance. There are no black & white prescriptive requirements that tells you how to determine whether the failure is credible, or how to calculate the required relief flow rate when it is determined to be credible. If you re-read 4.4.9.3.3 again, with an eye toward noticing this lack of specificity, you see what I'm talking about.

So, if your company is doing a good job of inspecting/testing these check valves, and the past records don't show problems with corrosion or wear, then it's perfectly acceptable for your company to conclude that reverse flow isn't a credible risk. It's also perfectly acceptable for your company to conclude that this particular application justifies more conservatism.

So, inform yourself about the details of the applications and your companies risk tolerance, and then proceed based on whatever makes sense for each specific application. There's no right or wrong answer.