How should API520 PartI, section 3.3.2.2 be interpreted: "Balanced spring-loaded or pilot-operated PRVs should be considered if the superimposed back pressure is variable. However, if the amount of variable superimposed back pressure is small, a conventional valve could be used provided......."?

For our application:
1) A conventional PSV protects column C209 that has a design pressure of 7 barg.
2) The Net Spring Setting (NSS) of this PSV is 2.3 bar.
3)The PSV discharges into vessel V208 that has a design pressure of 3.5 barg.
4)V208 has a normal operating pressure of 0.2 barg; hence the operating pressure range is from 0.2 barg to 3.5 barg
5)We hence have "Variable Superimposed Backpressure". BP % goes from 8% during relief at 2.5 barg (0.2/2.5) to 60% during relief at 5.8 barg (3.5/5.8)
6)Built-up back pressure during relief is less than 0.1 bar (hence < 10% NSS)

From a built-up back pressure point of view (<10% NSS), a conventional PSV is suitable (column C209 will never exceed its design pressure). However, we do have variable superimposed back pressure, which varies from 0.2 to 3.5 barg and hence cannot be considered small. What is the reasoning behind API's requirement for not applying conventional PSVs in services with variable superimposed pressure, if in our case, the vessel being protected will not exceed its design pressure? Is it because of the fact that the PSV will open under different setpoints (i.e. from 2.5 to 5.8 barg, depending on pressure in discharge vessel V208) and that this is not desirable?

Any clarification that you can provide in this regard is much appreciated.

Many thanks
Rihard Cooper

Rihard,
Welcome.

Good question.

There are at least three factors. i) maximum allowable overpressure(%) ii) instability iii) capacity.

I don't have the P&ID and don't know the exact routing...i guess your receiver is having another PSV and protecting it from overpressure and set at 3.5 barg. If so, the maximum backpressure to column could be 110% of 3.5 barg.

i) maximum allowable overpressure(%)
Built-up back pressure (dP) is ~0.1 bar < 0.23 bar & Pb + 0.1 < P design ==> OK

ii) instability
Built back pressure (dP) < 10% of Set pressure ==> stabil

*Previously was written as "Built back pressure (dP) < 10% of NSS ==> stabil"


iii) Capacity
If Pb = 0.2 + 0.1 barg, the Pb/Ps ratio will be (0.2+0.1+1.01325)/(0.2+0.1+2.3+1.01325) < 55% ==> Critical flow
If Pb = 110% x 3.5 + 0.1 barg, the Pb/Ps ratio will be (1.1x3.5+0.1+1.01325)/(1.1x3.5+0.1+2.3+1.01325) > 55% ==> sub-critical flow

*Previously pressure ratio was gauge-to-gauge

There are transition between critical and subcritical flow ==> Take extra attention on the PSV capacity. It could be peak relieving load and maximum relieving area (maxi relieving load may NOT be maximum relieving area) in some point in the middle...

Sorry for quick and untidy response.

If Pb = 0.2 + 0.1 barg, the Pb/Ps ratio will be (0.2+0.1)/(0.2+0.1+2.3) < 55% ==> Critical flow
If Pb = 110% x 3.5 + 0.1 barg, the Pb/Ps ratio will be (1.1x3.5+0.1)/(1.1x3.5+0.1+2.3) > 55% ==> sub-critical flow



Why did'nt you use absolute pressure ratio?

Regards

Oo...lah..lah.... ai...ya...ya......I should not made such mistake...

OK according to API 521 Part 1, section 3.6.1.4, it shall be absolute pressure ratio.

Hi Joe Wong

Thanks for your reply.
Sorry, I forgot to mention that this is a liquid relief at non-critical flow. Nonetheless, your explanation is clear.

Much appreciated.
Richard

Could we use NSS as Design Pressure for evaluating the built-up back pressure?

In API 520, everywhere mentioned "set pressure" it is equal to NSS? Have the API 520 assumed value of "set pressure" after reducing spring load of PSV for compensating superimposed backpressure?
Regards

NSS is set pressure minus constant back pressure (superimposed back pressure). This is only applicable to conventional valves. In the case of rdcooper the set pressure is 5.8 barg and the NSS 2.3 bar, if V208 is equipped with a PSV set at 3.5 barg.

Where does the API say you can't use conventional PSV's for applications with a "variable" (?) superimposed back pressure?

You are requested precisely pay attention to my questions along with a glance to previous discussions in this post.Rdcooper has evaluated built-up backpressure based on NSS not set pressure.Simply i want to know for conventional PSV, built-up backpressure in the case of e.g. 10% allowable overpressure should be 10% NSS or 10% Set Pressure?Is the Set Pressure mentioned in API 520, equal to NSS?

Ok, i guess my answer was unclear unless you didn't pay much attention to the answer. But you gave the answer yourself basically, when NSS is set pressure minus superimposed back pressure than NSS is not equal to set pressure.

Dear Joe Wong,
Would you explain why did you use NSS instead Set Pressure for evaluating built-up backpressure in your previous post in current topic?
Regards

Set pressure to be used.

Thanks for your reply,but as last point:
Don't you think NSS instead Set Pressure (for evaluating built-up back pressure) might be considered only in conservation point of view?
Regards

I think fallah bring us interesting points:

1/ Shall we use NSS or set pressure for evaluating built-up back-pressure?

2/ Moreover, when specifying a conventional releif valve, shall we consider the "set pressure" as the MAWP of the equipment protected or the differential set pressure, that is: MAWP minus superimposed back-pressure?
Indeed, API RP 520 states: "" The set pressure is the inlet gauge pressure at which the pressure relief device is set to open under service conditions."
Then, For a conventional valve, Pupstream - Pdownstream = Pset

API stated Pset is inlet gauge pressure. Why the definition change to differential pressure ??

Why ? Please elaborate your points instead of just asking questions ?

OK but what about the "under service condition" part of the statement? how to interpret it?

The PSV in "service condition" can be interpreted as PSV from it put in service (before it open), open when set pressure is reached, continue to lift due to overpressure, reseat during blowdown and finally back to normal close position as the internal pressure is dropped below set pressure.

Some may take opportunity to say that "service condition" is from PSV starts open to it reseat position. If this is interpretation, are we saying that we the PSV is close, it is not in "service condition" ?


Having struggle with this kind of defintion which is rather difficult and debatable. I would leave it to API to make necessary call.

In the doubt, maybe we should be conservative and stick with the higher end. that is, consider the differential set pressure as the set pressure in our data sheets and also to evaluate the built-up back-pressure...

Sheiko,

The point that i would like to bring up is seek for API clarification. Take the correct definition and apply instead taking unnecessary and probably costly approach.

Dear all,

I have contacted a vendor (Farris) that told me that the set pressure Set pressure is the gauge pressure at the valve inlet, for which the safety relief valve has been adjusted to open under service conditions. In liquid service, set pressure is determined by the inlet pressure at which the valve starts to discharge. In gas or vapor service, the set pressure is determined by the inlet pressure at which the valve pops. this is in agreement with what Joe said.

He also pointed out, that for conventionnal relief valve, total variable back-pressure cannot exceed allowable accumulation, otherwise balanced bellows or pilot valves are required. Note that the total variable back-pressure is the sum of the variable portion of super-imposed back-pressure and built-up backpressure.

Then if we consider the example of this post:
variable portion of super-imposed back-pressure = 3.5 - 0.2 = 3.3 barg
built-up back-pressure = 0.1 bar
total variable back-pressure = 3.3 + 0.1 = 3.4 barg > 0.7 barg
So, conventionnal type is not to be used.