Application/Capacity and Sizing of Pressure relief valves

[salar1363]

Hi everybody,

I'm a lttle bit confused with purpose of having Pressure Relief Valves on different points and i'm full of different questions in my mind. i appreciate recieving your technical helps.
On top of pressure vessels, we almost always consider PRV, which prevents from over pressurizing the vessel. In two phase seperators (liquid/liquid), the capacity of these PRV were set as big as the inlet flow rate to these vessels.
My questions at this stage :
1. What can cause over pressure ?? changing in the pressure of upstreams only, or other reasons it can have??
2. Shall we consider fire cases also ?? or the mentioned method has covered fire cases ??
3. what about cases with seperation of gas/liquid?? what shall be considered as the required capacity of PSV??

I'm looking forawrd to recieving your answers
Thank you so much

Salar

 

[MortenA]

you should read API 520 and API 521 and then come back with more questions.

Best regards

Morten

 

[salar1363]

Dear Mortena

Thanks a lot for the references, but before reading API standards, I would like to make sure the purpose of having PRV on processes. As far as I know, only one reason can be Fire, while on pressure vessels, instead of considering Fire cases, we usually consider PRV with the capacity of total inlet to the vessel. these questions arise after reading some similar questions, on thi site, that those guys have considered Fire case design for PRV, for same projects.
I really need first clarify this issue, and I'll go through API 520 and API 521 as you mentioned.
Thank you so much

Salar

 

[pleckner]

@salar1363:

No, before doing anything else you really need to do what Morten wrote and read API. To answer your question on what can cause overpressure, you want to start with API Standard 521. Then for sizing help you want to read API RP520.

To determine the size of the PSV you will ultimately choose you will need to calculate the size of a PSV for each credible scenario you can come up with. The chosen PSV will the the largest you calculate from all the analyses.

As far as determining what the credible scenarios are, only you can do that; we in the INTERNET world cannot for a couple of reasons. One is that we don't have a detailed P&ID of your system nor do we have your design conditions, equipment sizes or heat and material balance. Another is what I would call a major liability issue. We can give you suggestions but what if we miss something and you figure what we give you is complete?

Again, any reponses from us must be considered suggestions only. And here is a couple of suggestions. (1) If this is a flammable system that can sustain a fire then fire case would be a credible scenario. If neighboring equipment can catch on fire, then the PSV for this separator may be needed to be checked for fire. (2) If you can block in your separator and the source feed equipment (pump) can overpressure it, then this can be a credible scenairo.

Once you come up with all of your credible scenarios you need to determine the relieving rate for each. With this information you you must then determine the size of a PSV for each. The PSV chosen will be the largest once calculated.

Since you are a novice at doing this analysis, please find someone in your company to help you. Not only can your mistakes result in equipment damage and even the loss of a life, it will probablly be a violaiton of law (depening on your place of origin).

 

[salar1363]

Dear Pleckner

Thank you so much for your answer, I think I have some general Idea now, and I will read API 521 and API 520 , and then, ask my questions again,
Regards

Salar

 

[dcasto]

To get you started on scenerios,things like any thermal input, fires, heaters, solar radiation can cause a pressure increse. Mechanical equipment like pumps and compressors. These mechanical sourses of pressure increase are from failures of control valves or ESD systems as well as human error.

Failure of a high pressure tube in an exchanger can cause one side of the exchanger to leak the high pressure into the low pressure.

Chemical reactions in the process could cause the pressure to rise.

You will also need to establish how many fails must happen before a credible case of overpressure can occur. Cedrtain ly if 10 valves must fail from independant cause (not a loss of a common air supply for example), the probability is to small to be credible. Most companies allow only a single fault or in some processes two faults. If you want get even more detailed ISA 86 (don't have my book here) has a standard to evaluate probibities and levels of protection to document your companies risk assesments.