Total Backpressure at Actual Capacity of PSV 1

[chemter]

Hello.
I understand that the nameplate capacity of the relief valve (corrected for the fluid at flowing conditions at 10% overpressure) is used to determine the system's inlet and discharge pressure drop to verify the inlet losses do not exceed 3% and the backpressure (superimposed and built-up) does not exceed 10%, for a conventional relief valve. However, for a case where a PSV's set pressure is less than the MAWP of the equipment being protected (for example SP = 80 psig but MAWP = 100 psig) and you are allowed to use a relieving pressure of 100 psig + 10% Accumualation, you are therefore using an overpressure of 37.5% (80 psig + 37.5% Overpressure = 110 psig). Assuming this is acceptable for the system, my question is that since you've already proven the system's piping meets the guidlines for the 3% inlet loss and the 10% backpressure at the rated capacity, do you need to perform the pressure drops again at the higher actual relieving rate to show that the system is still adequate and that the system's backpressure does not exceed the overpressure?
Thank you.

 

[greg87]

The primary issue is that at the point that the discharge line's pressure drop exceeds about 10% of set point (8 psi) a conventional valve will close. It will then reopen as soon as the backpressure descreases, and so on. This rapid and possibly violent opening and closing can damage and potentially destroy the valve. A valve for gas service will basically initially open all the way, so if that flow rate causes too much backpressure you could have the problem.

If the valve is an ASME, which it would need to be if it is protecting an amse vessel, the valve manufacture will not state a capacity at a relieving pressure that is over 21% (fire case) of the setpoint [that may be true of any relief valve],so although you could calculate an approximate figure, your documentation would not be appropriate for ASME requirments.

 

[pleckner]

I will second greg87. The vendor will only guarantee the PSV stamped capacity at 10% overpressure using the set pressure as the basis, not the design pressure or MAWP (if different).

And I will argue why even go in this direction (I hope this is just for knowledge)? If you are so inclined to allow the relieving pressure to go to 110 psig (in your example)in the first place, why not just specify the PSV set pressure at 100 psig and don't worry about trying to become creative with the calculations?

The reason we sometimes set the PSV for a lower pressure than design (or MAWP) is because perhaps the contents can exhibit a run-away if the relieving pressure (and coincident temperature) was higher. Another reason is that for a multi-product vessel, perhaps too high of a pressure (and coincident temperature) will have a degradation effect on the contents. Otherwise, set the PSV at design pressure, you may even get away with a smaller PSV as a bonus.

 

[chemter]

greg87,
Thank you for your response. But for my question, I am assuming that the backpressure at the nameplate capacity (at 10% overpressure) is not greater than 10%. Therefore, once this is established, if I allow the overpressure to increase beyond 10% to 37.5%, the backpressure may increase above 10%, but so will the overpressure. Since I am now at these new conditions,do I need to re-calculate the pressure drops to make sure my backpressure does not exceed the overpressure? Thanks.

 

[pleckner]

Aa conventional style spring loaded PSV can be allowed to exceed the 10% back pressure limitation as long as the larger back pressure is less than the larger overpressure; and this allowance is clearly stated in API RP52o and 521 (well, almost clearly). Check Section 5 - Disposal Systems and API RP520 (7th Ed.), Section 3.3.3.1.3.

Do you need to calculate this? Absolutely and keep it with the PSV records. You will need to prove the system was designed safely in the event something happens and you get a visit from any number of authorities.

P.S. they don't like to have to interpret anything and want to see everything spelled out for them.

 

[chemter]

Pleckner,
Thank you for your response. I understand your statement "The vendor will only guarantee the PSV stamped capacity at 10% overpressure using the set pressure as the basis, not the design pressure or MAWP (if different)." However, I would tend to think that the PSV capacity would still be the same through a relief valve with a set pressure of 80 psig + 37.5% overpressure as it would through a PSV with a set pressure of 100 psig + 10% overpressure. Using the same valve and the same orifice size(say a 1" x 2" with an E orifice), in either case your pressure entering the PSV nozzle would still be the same, wouldn't it. Pleckner, I know that you are stating what the code says and that my example above for the lower set pressure with 37.5% overpressure would not be considered to follow the code or verifed by a PSV vendor, so again I respectfully do not argue your statement, I simply want to understand what is actually happening during these pressure events. But back to my original question, what I am trying to determine is for cases where you have an overpressure greater than 10%, do you have to (or should you) recalculate your pressure drops to ensure the backpressure does not exceed your overpressure. Maybe for a fire case, where you have are allowed 21% accumulation over your MAWP. If you verify the pressure drops are in accordance with the code at the nameplate capacity, do you need to also verify them at 21% accumulation (assuming SP=MAWP for the fire case)? Thank you very much.

 

[chemter]

pleckner,
Sorry, I believe I responded at the same time you were submitting your last post.

 

[pleckner]

"chemter":

First, one clarification in what is Code and what is not. API RP520 and 521 are recommended practices, not Code. The term Code is usually reserved to indicate "law" and the RPs are just guides; albiet considered "good engineering practice" by most authorities.

To respond to your first comment, of course in your example the pressures are the same but this is only because you designed the system incorrectly in the first place as I pointed out in my post. You determine the orifice size needed by determining a maximum relieving rate at some defined relieving pressure and temperature and physical properties. In your case, your system should have been calcuated for a relieving pressure of 88 psig. If your system was calcualted at a relieving pressure of 110 psig, perhaps you might acutally have gotten away with a smaller PSV and a smaller stamped capacity.

At the risk of repeating my last post, if you truly want a relieving pressure of 110 psig, then why not just specify a PSV with a set pressure of 100 psig and stop with the creative calcuations?

To respond to your second coment, what is happening with these pressure events? On this conventional PSV, as the back pressure increases, it puts pressure on the spring to try to force it to close. The only thing keeping the PSV opened is the opposing inlet force or set pressure + overpressure. That's why you can sometimes let the back pressure exceed this 10% criteria.

Do you need to recalculate the back pressure? Well, what is the capacity of the PSV at your new conditions? Do you know it; it is not the same as the stamped value? As I pointed out in my last post, the vendor determines the PSV capacity at 10% overpressure and this is what is put onto the stamp. So you would have to find some way of adjusting the stamped capacity if you are indeed letting the system exceed the testing parameters.

You are in luck, somewhat. ASME gives you a simple equation to convert the stamped capacity to a pseudo capacity at a different condition. I don't like it becuase it is nothing more than the use of the ideal gas law. So you can use this new calculated flow, and using the new properties at the higher pressure and temperature than what the capacity test was based on, determine your new pressure drops in the discharge line. So yes, you need to do some more calculations.

 

[chemter]

pleckner,
Thank you very much for your response. I agree that the discharge line should be checked to make sure the backpressure does not exceed the overpressure. One more question(s) though, what about the inlet pressure drop. Does this also need to be reconfirmed at the higher capacity? Also, does ASME or API suggest using the Required Relieving Capacity (for example, from the fire case which may exceed the nameplate capacity at 10% overpressure) or the Actual Capacity at the MAWP + 21% (for the case of fire with SP = MAWP)? Thanks again for your guidance.

 

[pleckner]

Well, in this case it's ASME and only a recommendation at this time and they require that the stamped capacity be used in the calculation at 10% over pressure. This now brings up a problem when you are doing what you are doing because there is no real way to adjust this criteria. Saying it another way, there is no guidline on what to do if the relieving pressure is over 10% of set pressure! Some countries other than the US have other guides.

The whole reason for inlet line check is somewhat similar to the back pressure check. A PSV will close when a certain pressure at the valve is reached; this pressure is called the blowdown pressure and the pressure drop is based on a percentage of the PSV set pressure. The 3% criteria (3% Rule) was set to maintain a margin between the blowdown pressure and the PSV set pressure so the valve will stay opened at all times until the relief is over. The blowdown pressure is a function of the valve and not the over pressure. Thus if you have a higher overpressure, then in reality there is a greater differential between blowdown and the real pressure at the inlet to the valve and you can tolerate a greater pressure drop. But the guides at this time don't really let you take any of this into account.

You can talk to the valve's manufacturer about this and get their take but don't hold your breath for much. They won't tell you to deviate from the guides as it then becomes a liability issue for them.

And by the way, one question I have for you. Why? Why go through all of this? Why can't you just set the PSV at MAWP?

 

[chemter]

pleckner,
Thanks again for your comments. To answer your question, "Why can't you just set the PSV at MAWP?", it seems that would be the easiest path to take, however, it also seems to me that there are numerous cases where the PSV's set pressure is below the MAWP for existing systems. Why? I do not know, but it often seems that if there is a small gap between the set pressure and the MAWP, and a slight increase in overpressure will get you what you need, you may be leaned on to just show it this way. The reasoning being what I stated earlier, that the same orifice is present and all you are doing is letting the pressure increase the same way it would with a higher set pressure. However, I understand your reasoning and wish that the Code would go into more detail on this issue since it seems to come up all the time. Maybe it would put an end to these type of situations. I look for literature on PSV sizing and calculations in general, but it is difficult to find specifics. Can you recommend any books that discuss PSVs and their design in the same way in which they are actually sized by engineers? API 520, 521 give a great foundation as well as ASME, but I would love to go into the sizing even more. I realize there is always some differences in methodology based on clients' practices, but some of the fundamentals are still there. I guess the rest is just experience and discussions like these with individuals that volunteer their time to offer help like you. Thank you again pleckner for your time.

 

[pleckner]

You are very welcome and I glad to be of help.

"Sizing" a relief valve is fairly easy and straight forward. The equations given APR RP520 are the current way to go. It's coming up with the relieving scenarios and calculating relieving rates is the real engineering. Each is very specific to the application so it would be very hard to generalize. There are similarities in many systems but then they too can be turned into specifics that would throw the similarity out the door.

A good book to have is:

Center for Chemical Process Safety, "Guidelines for Pressure Relief and Effluent Handling Systems," American Institute of Chemical Engineers, New York, N.Y., 1998.

There is a move to update this book with the latest information on 2-phase relief. When the new edition will ever come out is totally unknown at this time.