PSV Setpoint for Fire Case 1

[akcullin]

It seems in the oil and gas industry that PSVs are always set at the vessel MAWP, even in fire case. However, it's occurred to me that if I have an inlet separator with an MAWP of 1440psig @-20/130F, then during a fire 130F will almost certainly be exceeded (as proven by many dynamic fire case simulations to calculate PSV size).

What do people on this forum think about this? Should the PSV be set to the maximum pressure of the vessel at the relieving temperature? (the calculation for that is another discussion) However, the temperature will continue to rise as the PSV relieves so then what temperature do I choose?

Does ASME, API, or any other code talk about this?

What are everyone's best practices?

 

[mk3223]

It correct that temperature is to be increased during the fire around the equipment.
My understanding is that the actual vapor discharge should be calculated per the fire temperature. Also, in addition to the PSV protection, the equipment may require to have an emergency depressing procedure in the fire event to protect the plant.

 

[Latexman]

I think you are worried about over-temperature protection, which is a separate issue from over-pressure protection. There is some pertinent discussion in the archives that may help you. Use Search, under the thread title, between Forum and FAQs. I got 20-50 hits depending on my search criteria. Here's one that may help you - thread1203-324332. If not, come back at us with more details, like dimensions of separator, is it piping or a vessel, the fluid involved, etc.

Good Luck,
Latexman
Pats' Pub's Proprietor

 

[georgeverghese]

Relieving temp is to be maintained within design conditions for non fire case relief scenarios only.

 

[MortenA]

When sizing you consider two cases:

a) A liquid filled
b) a dry

Most cases you would only consider the liquid filled - this is not completely liquid filled but filled to "normal level" - and for an inlet separator this would usually be around 50%

In this case the fire would cause the liquid to heat until its boiling point and then it would not become any hotter! The liquid would the cool the metal so that it dosnt heat much more. This may change as liquid boil off but is (usually) not considered. If you feel that its a case that your fire (w.o. external cooling from either deluge or fire fighting) can last for so long that the liquid composition changes so much (or that all of the liquid evaporates) then you must consider this. A solution could be a bursting disc that would depressurize your vessel. Would no tsolve all your problems but mitigate them to a large extent.

This is also why the point of deluge is not to quench the fire bur to cool the vessel!

For case b) - yes and very likely many empty vessels are undersized for this type of fire and will explode...

 

[don1980]

inlet separator with an MAWP of 1440psig @-20/130F

That's an unusually low temperature (130F). You could consider getting a vessel engineer to recalculate the MAWP at a higher temperature, but is that really worthwhile in this case? Very doubtful, because the fluid is almost certainly supercritical at the PSV set pressure. That means the PSV isn't going to provide any meaningful protection from fire exposure. The vessel wall temperature is going to rapidly rise to the yield point because there's no cooling action inside the vessel (no Hvap).

It's possible that the PSV set pressure (and relieving pressure) could be lowered below the critical point, but the fact that the vessel MAWP is 1440 psig indicates that it's very unlikely that your operating pressure is low enough to allow this option. So, this is a case in which you need to focus attention on other layers of protection. You still need a PSV for compliance reasons (and for other scenarios, if they exist), but there's no point in sizing it for fire exposure. Regardless of whether that PSV is a 1x2 or an 8x10, it's essentially useless in protecting from fire exposure.

Consider protective measures such as auto-depressurization, water-spray, and fire-resistant insulation.

 

[MortenA]

@Don,

I wouldnt call that unusual (around 50ºC) - makes some of your corrosion/material issues "go away". Also, while you may be above Pc for the gas phase you will still have a liquid phase at this pressure.