Depressuring Valves to avoid BLEVE during a Fire

[Leclerc]

What should be the air fail action for depressuring valves installed, or nominated, to avoid a BLEVE when fire surrounds a pressure vessel?

If air fail closed, and if fire also causes damage to air system then depressuring will not happen and BLEVE possible.

If air fail open, then vessel depressurizes when there is no fire and at a far higher frequency than fire, leading to undesired/ unacceptable release of organics.

As a rider to the general problem above, we have, in the past, nominated certain PCVs to also serve as depressuring valves. These PCVs tend to have air fail closed actions. Are there any comments on this state of affairs?

 

[sshep]

What about a fusible link design that opens with spring actuator when link melts? This would only open in the event of a fire and are independent of the air system.

just a thought, sshep

 

[Homayun]

Leclerc,

This is a very common problme/ question. I used to work on the design of a NGL plant and there we had the same discussions again and both arguments are true. But I guess that one would certainly want to have the Blow down valves be opened during a fire event. So I tend to make them Air FAil open. But I will then make sure that I have a relibale source of Air that doesn't fail every now and then.

Other thing is to look for a similar design that your company has with other systems.

 

[1969grad]

I prefer to have any ESD system go to a de-energized state or position during emergencies. That means the blowdown valve should go to the open position on loss of instrument air and/or signal to the solenoid controlling the valve.

I know that some companies go to an energized state during emergencies but I fail to see the logic and advantages.

In my opion a BLEVE is a catastrophic condition to the point that inadvertant release of organics is something you may have to put up with.

 

[StoneCold]

Leclerc
We have an interesting transfer system for a pyrophoric liquid that may be usefull to you. How is works is there is plastic tubing running along the transfer line. This plastic tubing suppies air to the pneumatic valve on the large tank, allowing it to open. In the event of a fire (leak) the plastic tubing melts and the valve closes. You could have the same type of system. Loss of air to the tank dump valves due to melted hoses causes the valves to open. You could run the valves off of the air system with an intermediate storage tank for the air. Have a check valve on the storage tank so that if you loose your compressor you still have a tank of air to supply the valves until you restore air to the plant.
You have to keep an eye on the tubing and change it out on a regular basis because it gets brittle.

Goodluck
StoneCold

 

[EGT01]

Leclerc,

I don't think that there is any one answer that is correct for all situations. As the others have indicated I think the typical action would be depressuring valves fail open and shutdown valves fail closed but that does not have to be so in every case.

I believe what you are doing is the correct way to approach the problem and you need to evaluate the advantages and disadvantages of your options. Pick the fail position that makes the most sense for your system but expand your evaluation to more than just the fail action of the depressuring valve.

I can't offer any thorough reference for depressuring systems but API RP-521 does have a little information that you may want to review. Basically, the key points they make include...
"The valves should remain operable for the duration of the emergency or should fail in a full open position."
I would put the emphasis on the "or" and don't necessarily interpret that to mean that all depressuring systems should fail open.

API 521 goes on to say...
"Fireproofing of the power supply and valve actuator may be required in a fire zone."
So there are other factors that should be part of your design that should affect your decision.

For example, I have seen emergency air supplies (small pressurized containers) sufficient to actuate an air operated valve located close connected with valve. Maybe you could locate your depressuring valve outside the fire zone or maybe install redundant valves. A fusible link sounds interesting, some relief systems on compressed gas cylinders include a combination rupture disc and fusible plug to ensure release only during a fire event. What about depressuring your system to some type of disposal system (flare or scrubber).

I happened to find this article on the Instrument Society's website that talks about using a Safety Integrity Level (SIL) selection method relating to depressuring of a hydrocracker reactor. You might want to consider using an approach like this to help decide which fail action is best for your system.

http://www.isa.org/InTechTemplate.c...Management/ContentDisplay.cfm&ContentID=30532

 

[Leclerc]

My thanks to all for their comments so far.

I did know that, in the absence of any particularly difficult problems arising, it is better to have a safety system fail safe (air fail open), as 'Homayun' and '1969grad' reminded me.I just wanted to air the general problem again and see what comments came back.

At the moment, in view of the fact that the material released is directed to a scrubber,that there is some scope for locating the valve outside of the fire zone and that the air supply to the valve can be judiciously routed, I am minded to go for "air fail open".

 

[jsummerfield]

I worked on an offshore project that split the failure - differently for loss of the signal from the SIS than loss of air pressure. The SIS energized a solenoid to initiate a blowdown; however, pneumatically, the system was fail-safe - loss of air pressure opened the blow down valves.

John