Pressure Testing of Pipe - Leave the Control Valves and Instrumentation in place? 1

[ChEMatt]

This is a cross post from another forum as I'm not sure I posted it in the right place.


Question has come up now and again and I don't have a really good answer for it. If this is the wrong forum, please direct me to the correct place to post this. Thanks.

Let's say I have a system that is designed for 300 psig maximum operating pressure. I ensure that all my instrumentation is designed to handle up to MAOP (or MAWP, if you like) pressure. The pipe was assembled with the control valve, pressure, temperature, and flow instrumentation.

The pressure test calls for 1.5x maximum allowable pressure. The control valve has 300# ANSI flanges on it, but as designed it says it's only good to 300 psig. Do you leave it in the line? Do you leave all the other instruments in the line and expose them to 450 psig? Should they be capable of handling that pressure if the system is designed only for 300 psig?

Sorry if the question isn't clear or lacks details. Thank you for your help.

-m

 

[sshep]

Hey Matt,
It sounds like the control valve should be replaced with a temporary spool. The instruments should be disconnected. One project i worked on crushed many pressure capsules by not isolating instruments during hydrotest- unless you are sure, it is better not to risk it.
Best wishes,
Sshep

 

[moltenmetal]

The "leak test" required by B31.3- which is actually a pressure integrity test, NOT a leak test- requires you to test hydrostatically to 1.5x the MAWP (times the ratio of stresses), or to pneumatically test to 1.1x the MAWP. What is the MAWP? It is the RELIEF PRESSURE, or more properly, the maximum pressure to which relief may be safely set. Yes, defining things that way seems a little backwards- but in my opinion, it helps people understand the terms better so they use them more precisely. The maximum operating pressure and the nature of the relief device being used, set the minimum MAWP you may use in design. Relief valves "simmer" (i.e. leak, a lot) at 90% of set pressure, and the best rupture disks don't like to be operated at 90% of their set pressure either. A prudence margin of about 30% between max operating and relief setpoint (MAWP) is the minimum we'd allow, unless there are very special circumstances. So no, I wouldn't equate maximum "allowable" operating pressure with MAWP- that just causes confusion.

All devices which retain pressure in the line must ultimately be tested, or (like pumps) be covered by another standard that requires them to be tested.

If the connection between devices and piping is made by means of flanges, and a particular item has been bench tested, it need not be tested again in the line. It can be replaced with a spool, and each spool may be tested on the bench from flange to flange. But if a component has a MAWP lower than that of the rest of the line, it determines the real MAWP for the line and hence the maximum relief pressure setpoint.

If a line contains nothing but pipe and flanges, and the pipe and flanges are designed to a spec which sets the MAWP equal to the flange rating (which is often the case), then the line may for convenience be tested to the MAWP of the flanges. However, that convenience falls in the toilet when you're testing a line consisting of more than just pipe/fittings and flanges- the MAWP may be limited by any component- in this case it sounds like it's the body of your control valve. There is no CODE requirement to test to a pressure higher than the multiple of the MAWP set out by the code, and per code ANY component may limit the MAWP. Flange-limited design is a convenience, and some consider it to also be a prudent choice, but it is not a requirement of the code.

The only device you should need to remove from the line for a pressure integrity test is the relief device itself, for obvious reasons! The test is at a multiple of its set pressure. That is, unless your process isn't tolerant of water or your other test media.

A post-assembly leakage test at max operating pressure is still required prior to start-up with anything hazardous in my opinion.