Propane fire case PSV relief

[larryli2004]

I have a few 2000 bbl propane bullet with a design pressure 250 psig. We have sized the PSV for fire case. My question is, can we relief all PSV to atmopshere? If we bring the PSV relief to flare system, the header will be very large. If we vent the relief to atmosphere, is there any code requirement as far as stand point of spacing. For example, how far we should keep the relief away from ignition sources? Do we need to build any berm or dyke for bullet? Thanks.

 

[Chance17]

1. In general it is allowed for emergency venting to go to atmosphere.
But double check with your local air emission regulatory agency. Your facility permit may have some special clauses.

2. I am not aware of any specific horizontal space seperation for vents other than
a) the recommended API tank spacing and
b)spacing requirement to maintain the area electrical classification.
If the vent is near a working platform API calls for the outlet to be 10-ft above and 50-ft horizontal from platform.
For dispersion of flammables, the tail pipe velocity should be greater than 100-ft/sec for the smallest contingency.

Hopefully steam is available at your facility to be piped at end of tail pipe.
If you had a release (or small seepage) which ignited, the steam would be available to snuff the flame.

 

[larryli2004]

Thanks Chance17. Yes, we will contact local regulatory body and check any concerns they have. We have a flare stack on site for some thermal relief PSVs and manual blow down. Currently it is 50 m away from propane bullet. We don't have steam available at site. Our site is relatively small and if the PSV is vented to atmosphere, we are not planning to direct the PSV outlet to anywhere further. Probably it will be only 2 m and then go to atmosphere. I got a little bit worried about the distance to flare stack. I don't want the PSV venting cause another fire or explosion at flare stack.

 

[MJCronin]

I think this is a case where the rules contained in NFPA-58 must be followed, rather than the adoped generic API rules for storage of flammables.

NFPA-30 may be useful also.....

 

[larryli2004]

Yes, I checked NFPA-30 but I could find the direct answer. It is allowed per code. I am just not sure any additional requirements to be followed. When fire happens, vessel will collapse eventually but proper design may buy some time. If PSV is vented atmosphere, it may cause the second fire.

 

[dcasto]

If there is a fire at the tank, and the PSV is vent right at the tank, how can there be a second fire? If the C3 from the PSV catches fire, its a contained fire over the tanks.

 

[larryli2004]

I mean the PSV pop to atmosphere may make the things worse.

 

[MortenA]

Worse for whom?

You must remeber that the purpose of a PSV in the fire case is to prevent escalation of the accident by having the vessel rupturing.

So venting to atmosphere near the vessel may or may not make thing worse. I gues everybody knows the texas city incident. This promted BP to state that they would use cold vent. This is rather impractical in some applications. However - a LPG bullet may be one case where a cold vent could be dangeours e.g. in case of overfilling.

If this is a Shell job the theres a DEP about the subject - but i guess you would already know that.

Best regards

Morten

 

[larryli2004]

Thanks Morten. I see the prons and cons. I have seen a few plants with PSV vented to atmoshphere, I just personally haven't done any NGL project previously. What would be your best suggestion? We potentially have a few more bullets after expansion. Sizing for worse case and have all PSV connected to flare would make the header so big.

 

[MJCronin]

All over the US, many, many propane relief valves discharge directly into the atmosphere.

Propane bullets are protected from truck impact and are at set distances from structures, buildings etc.

Your propane storage facility should be located in conformance with NFPA 58 and the requirements of the local authorities...then

Draw a line around the propane tanks on the PID and...

Do not discharge into a flare header ..... IMHO this is a case where you have a utility (propane) that is "outside" of the rules of the process space.

Some "fun reading" about propane is contained in:

http://www.aristatek.com/newsletter/jan09/TechSpeak.pdf

http://ncsp.tamu.edu/reports/CSB/csb_4.pdf