CV Failure Philosophy 4

[Jefka]

Let's say there is a control valve with a system upstream designed for 200 psig (i.e. the upstream relief valve set pressure). Operations plans to actually run the upstream equipment at no more than 150 psig.

There's also a a relief valve downstream of the control valve (system downstream design = 50 psig). When sizing this relief valve for a CV fail open case, what does one assume for the CV upstream pressure in these calcs? Would you use:

a.) 150 psig
b.) 200 psig

Assume the client is indecisive and leaving it up to the designer.

 

[SNORGY]

Personally, I would size the downstream PSV for whichever upstream pressure gives rise to the higher mass flow through (and thus downstream of) the failed control valve.

Regards,

SNORGY.

 

[Chance17]

For the information given, I would use 90% of the upstream safety valve setting i.e. 180-psig.
It is not likely for the control valve to fail at the same time the safety valve is 200-psig.
But the upstream is capable to continuously operate at 180-psig.
Planning to operate at no more than 150-psig is only good intensions.

 

[Jefka]

Thanks, Chance- my argument, exactly. There is quite a bit of debate between myself and several another engineers on this.

The designer has no idea how operations will run the plant after startup and my experience with operations companies is that they're always pushing the limits of the equipment in terms of flow and pressure.

If anyone else has an opinion on this (latexman?), I'd like to hear it.

 

[MortenA]

Will it matter?

Since PSV's only comes in fixed sizes - being really cleaver chosing your relief rates dosnt always make a difference?

Why dont you check for both pressures - and if they both ead to the same size go for the larger?

Best regards

Morten

 

[Jefka]

Doing the math, it can matter. Quite a bit. In some situations this assumption will cause the valve to go up 1-2 sizes.

 

[pleckner]

Who is to say that if the upstream side went into relief an operator wouldn't try to depressure that side by manually opening the control valve all the way?

There is no engineering merit in just choosing an arbitrary upstream pressure like 180 psig. If anything, I can justify saying the upstream operating pressure would be 175 psig because it is pretty common to set the design pressure the greater of 10% or 25 psig higher than the maximum operating pressure.

You must size the downstream PSV using a delta P of 150 psi (200 - 50) and the full open Cv of the control valve.

 

[MortenA]

Jefka - it cant take that long to size in both cases - but then i will assume that you have done just that and found that its larger in the HP case - so i will rest my.

Best regards

Morten

 

[BenThayer]

i can't really argue with Chance's suggestion since above 90% of set, you are likely approaching "weeping" for that PSV and one should not choose to operate there.

however, another consideration would be to size for the MAWP of the upstream equipment which might be more than the 200 psig setting of the PSV.

Operations is notorius for changing PSV settings to match the MAWP and not look at all the downstream ramifications.

also consider a slightly smaller control valve or reduced trim if practical. this could limit the flow and help out.

 

[CJKruger]

Jefka, I agree with Chance17.

API 521 is very clear that we only need to look at one unrelated event at a time.

For the "Control valve fail open case" the upstream pressure can be taken as the normal operating pressure, except if you can see a reason why it would increase when the control valve fails open.

That being said, the normal operating pressure is typically 10% or 25 psi below the set pressure.

Also remember that for a control valve that fails open, the governing case may be after you lose liquid level in the upstream vessel.

Bottom line - in your case I would use 175 psig inlet pressure.

If they want to you to use 150 psig, then they need to change the relief valve set pressure to 175 psig.

Cilliers

 

[pleckner]

You keep on talking about a failed-open control valve and this is indeed a credible scenario but I would like someone to justify that an operator cannot rationalize to manually open that control valve to depressure the upstream system going into relief?

 

[sheiko]

All
Just follow Pleckner piece of advice! It's worth it.
Some people would even use the upstream relieving pressure, ie 220 psig...

"We don't believe things because they are true, things are true because we believe them."

 

[maddocks]

Fully agree with Pleckner - use the 200 Psig case.

 

[pleckner]

Yes, Sheiko is even more correct. Upstream pressure should be 220 psig (unless relieving scenario is fire of course)... I wasn't thinking it all the way through; need to include the allowable overpressure during relief.

 

[sheiko]

Not mentionning that, if there is a manual bypass valve (for maintenance, for startup,...) in parallel to the main control valve we're talking about, the rated Cv of the bypass valve should be added to the rated Cv of the main control valve for determining the relieving flowrate...

"We don't believe things because they are true, things are true because we believe them."

 

[CJKruger]

pleckner,

- So your scenario is: Vessel inlet control valve fails open. Overpressure the vessel. Operator or instrumentation opens the control valve in question.

If this is a valid scenario, then I agree that the control valve inlet pressure should be based on accummulated pres.

- For the control valve fail open scenario however, the inlet pressure should be based on the normal operating pressure.

In both cases you can take credit for the flow that continues to flow downstream.

Cilliers

 

[CMA010]

CJKruger: What do you define as normal operating pressure?

 

[Armen75]

As stated by the first post, I would take Normal Operating to be 150 psi

Note also that differential pressure of the control valve in this case is the normal operating pressure at the inlet and the relief pressure plus over pressure of lower system at the outlet.
so, assuming 10% overpressure, set is at 50 psi... gives us 55 psi. Total relieving flow for valve fail open scenario to consider 150-55=95 psi delta across the valve.

so, one needs to calculate the flow through the valve with normal inlet conditions with a pressure drop of 95 psi

 

[Latexman]

I'd like to know the function of the CV? Control pressure? Upstream or downstream pressure? Control flow? Liquid or gas? Something else? The answers to these will help determine what is going to happen when in "automatic". As others have stated, what the operations department allows during emergency conditions will determine what sizing secenarios to consider credible. If the plant does not have an operations department yet (design phase), then the assumptions made will determine and/or limit them on what they can do safely.

Good luck,
Latexman

 

[Jefka]

@Latexman: I was speaking for a completely hypothetical but let's say the control valve is an FCV taking plant air from an accumulator to a downstream user. That's a case where operations may be tempted to increase the pressure to increase the holding capacity of the vessel.

Manually relieving the upstream equipment is a very good point. However, I suppose am primarily concerned with a more general question:

Is assuming that operations may not always run the equipment as in the operations manual considered a "double jeopardy" case? While most operators would know not to mess with the PSV's or CV's, I've known some who would change an operating pressure or flowrate without hesitation.