Thermal Relief Valve Sizing

[EngMattP]

Hi All, I am new here. But have a question.

I am looking for a way to size pressure relief valves for thermal relief or fire relief. The application is a suction scrubber on a gas compressor. Normally, I size all my relief valves to relief full flow of the compressor. This is leading to us having some really big valves on the compressor skid, and that leads to more money and bigger pipe, and clearance issues, ETC ETC.

So I am looking for some credible source of reference to size a relief valve for thermal expansion/fire. I am seeing sizing in some places, sizing for this is based on how much gas you have contained in the area you are protecting. That said, you would need to be able to relieve that pressure, i.e. protecting your suction scrubber and piping. Well, what rate does that pressure/volume need to be released at? What criterion exists?
Just need some threads to grab at to see where I can go with this.

 

[zdas04]

Do you have a credible scenario that puts full compressor discharge pressure/flow onto the suction scrubber? That would be one funny looking P&ID.

If you look at API 520, it talks at length about credible scenarios, what you need to do is look at the categories that are possible (e.g., I've never seen the "runaway reaction" scenario to be credible in a compressor suction scrubber) and determine which are credible. It is usually 3-4 of the long list, most often a suction scrubber will need to evaluate "failure of controls", "blocked outlet", "Thermal", and "Fire".

If you look at the section on fire case you'll see that a jet fire (most common on a compressor) is not a credible overpressure scenario because the metal will fail in the jet long before the vessel overpressures from the added heat. A pooling liquid fire is sometimes credible on a diesel fueled compressor and you have to look at that. For the thermal relief case to be credible there has to be a credible way for the vessel to become liquid full and blocked in. I've never found a compressor where the suction scrubber had a credible way of becoming liquid full and then blocked in (I've often seen thermal relief required on a suction blowcase, but not the whole vessel).

Once you determine that a scenario is credible, then you have to look at the flow rate in that case. For thermal relief you are starting out liquid-full and blocked in so the flow rate is ml/hr and a very small PSV is called for. For a liquid-full vessel in a pooling fire the credible flow rate can be huge. For a wellsite compressor, the most common source of overpressure is the reservoir, which you should be able to determine what it will flow at the PSV set point.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.

 

[BigInch]

Relieve the gas volume that is retained in the vessel and surrounding piping after the compressor unit shuts down. You do have an ESD on the suction side of the plant, before the scrubber and shut down block valves on both the suction and discharge sides of the compressor unit, correct?

"People will work for you with blood and sweat and tears if they work for what they believe in......" - Simon Sinek