Expansion of CO2 3

[EngFire]

I'm not sure if this is thr correct place to post this question but i'll give it a shot.

I'm trying to determine what the pressure build up will be in a room where CO2 is to be released. I'm trying to plug some values into a spreadsheet to determin if I need to provide some pressure reliefe venting.

This space being protected is a mechanical space on a Offshore production facility where the walls are of steel construction.

I know the amount of Liquid CO2 that will discharge. Should I be using "Ideal Gas Law" or "Real Gas Law" (van der Waal) considering that CO2 is in liquid state while in storage?

 

[ColourfulFigsnDiags]

As a first approximation it is reasonable to use the Ideal gas Law PV=nRT, however, you will need to consider what temperature the discharged CO2 will be at, as this will change the pressure, possibly significantly. Adiabatic expansion of the CO2 from liquid state will result in a considerable cooling of the gas.

Read the Eng-Tips Site Policies at FAQ731-376

 

[Montemayor]

EngFire:

If you are offshore and trying to protect a structural housing, you will require over-pressure protection in the case of injecting CO2 for fire extinguishing purposes. This is not a case of whether you need it or not; depending on the amount of CO2 employed, for total safety, you should protect the enclosure from rupture or collapse. There is no economy or practical sense in doing a spreadsheet to determine IF you need it or not. The probability and hazard of overpressure is present the moment you inject CO2 for the sole purpose of putting out a fire. The subsequent resulting pressure build-up is something you inherit with the system employed and you must provide for a safe, mitigated pressure situation. And this is assuming you can safely rely on a positive liquid CO2 shutoff valve. This is sometimes difficult to accomplish due to improper, experienced mechanical design and I wouldn't put too much trust in it.

You are certainly doing the right thing by showing concern for a safe, correct design and I’d like to help.

First of all, you cannot reasonably and accurately determine the pressure build-up resulting from an injection of liquid CO2. And this has nothing whatsoever to do with the Ideal Gas Law or Johannes Diderik Van der Waals’ pioneering equation of state (which, although the first, is very inaccurate and cumbersome). The reason for this is that when you inject liquid CO2, you have no means of determining how much CO2 has been administered nor what temperature (& this will be increasing) the total gas mass is at. But even if you knew the total amount of CO2 injected, it would do you no good since the application of liquid CO2 onto a fire consists of expanding liquid CO2 into a resultant mixture of SOLID CO2 (dry ice “snow”) and GASEOUS CO2. The initial make-up of this 2-phase mixture is 50%-50% and it initially exists at -109 oF while continuously being warmed up by the fire it is suffocating and the surrounding enclosure mass. The solid CO2 sublimates (reverts to the gaseous state directly from the solid state) differentially in accordance with the surroundings and the equilibrium it is trying to establish in the enclosure. This is a very complex and dynamic heat and mass transfer model that I defy any industrious engineer to try to simulate with accuracy. It simply is never even attempted – much less done.

The practical way your problem is resolved is that a “fuse” type of apparatus is employed to rupture at a pre-determined pressure (less than the allowable for the enclosure) and release the CO2 to the atmosphere should the rupture occur. This “fuse” is usually installed in the form of a thin membrane of aluminum or other weak material that will ensure that it will fail before the structure starts to weaken. Its size (or capacity) should be based on liberal allowance for pressure build-up. Depending on the size of the protected enclosure, I don’t think a 3 ft -4 ft diameter is out of the picture. You should guide yourself by conservative design, taking into consideration the worse case scenario.

My basic recommendation is to not worry about calculating the rate of pressure buildup. Base yourself on the conservative stance that a pressure buildup will occur and that you must provide a relief device to safely protect the enclosure and personnel affected by its failure. By trying to predict the theoretical pressure (which any experienced engineer would not accept on face value) you will be putting your efforts on the effects and not the cause. Attack the problem of providing pressure relief directly and conservatively and you can’t go wrong.

 

[JimCasey]

You can protect the structure with blowout panels available from your favorite manufacturer of rupture discs. These are used on structure subject to possible deflagration/detonation/other overpressure and are designed to rupture before any structural damage occurs. THey will burst before a duct-taped piece of reinforced polyethylene sheet will burst, and are much more consistent as well. The poly sheet trick does not withastand wind or weathering well, either.

I agree with everything Montemayor said except I would expect the "fuse" device, such as the woods-metal element in a sprinkler head, to fail at a predetermined temperature, not pressure.

CO2 fire suppression systems are sold by a number of suppliers, and professional application assistance is available from these suppliers to ensure the most effective mitigation of an incident.

Halon fire suppression systems are not as popular as they once were, as Halon is an extremely effective fire suppressant but a halogenated hydrocarbon and might be bad for the ozone, as opposed to the tons of God-Knows-What that might be released if the fire gets away from you and burns down the entire plant.

 

[Montemayor]

Jim:

I apologize for not communicating as well as I should have. What I meant by a “fuse” device was exactly what you have so accurately defined: blowout panels. I want to make sure no one interprets my post to suggest or recommend that a fusible device – such as that used in a sprinkler head – be used to initiate the spraying of liquid CO2 within a confined area where it is feasible to have humans. Designing for the automatic injection of an inert gas – such as CO2 or Nitrogen – into a space where human beings could be located runs the danger of asphyxiating the personnel or causing a panic situation. This was not my intent nor do I consider this a safe design. EngFire failed to state that the enclosure in question was totally devoid of human occupation and that the operation of liquid CO2 injection would be a manual one. I failed to specifically note that I assumed that the enclosure would be not occupied by humans – such as an electrical substation – and that administrative control would ensure that the manual injection of liquid CO2 would not occur until the area was secured and positive knowledge indicated that no humans would be caught in the enclosed CO2 atmosphere. This is a very important premise and I apologize again for not being detailed and specific in pointing it out. Your post has alerted me to this omission and I am indebted to your contribution.

As engineers we are all cognizant of the fact that if we are not detailed and careful, we are capable of creating a worse scenario than the original if we fail to apply safe and thorough principles to protect personnel at all times. The relieving of CO2 pressure within an enclosure is a positive step to protect the platform personnel from potential structural failure; however, in doing so, we should also be constantly on the alert to avoid a potential asphyxiation scenario or a situation where a “cloud” of cold CO2 vapor could prevent safe visual personnel evacuation.

Your note recommending professional assistance be used in applying a CO2 fire suppression system should, in my opinion, always be heeded and applied.

 

[abeltio]

if the installation in question is subject to the authority of the US Fire Marshall, or if the unit is built under american codes... the standard to apply is NFPA 12.

One of the items to look at is the hinges and latches of the doors where the CO2 release takes place.

From your description, this installation is similar to the fire protection system on gas turbine compartments.
The initial CO2 release required by NFPA 12 is so violent that i've seen quite a few times doors being blown open and off the hinges, flying a couple of feet creating 2 problems:
1. they are projectiles
2. the CO2 concentration per NFPA 12 cannot be maintained.

in this case the doors act like the blowout panel... rendering the CO2 system ineffective.

check NFPA 12 std... the use of blowout panels may be contraindicated in order to maintain the required concentration of CO2 to ensure the fire is extinguished... holes in the walls mean that AIR may RUSH IN after the initial burst creating a potentially explosive environment (remember the movie BACKDRAFT?)

specially being an offshore installation... i'd much rather have a bulged bulkhead than a system that was released, has no more CO2 available and the fire is still live.

again... make sure that the door hinges and latches are strong enough to sustain the initial burst of CO2... AND, to ensure that the doors will sustain the initial CO2 discharge: BY NO MEANS MAKE THE DOORS TO OPEN INWARDS IN THE PROTECTED AREA! THAT IS A DEATH TRAP!!

also, make sure that the fire DETECTION system WIRING is rated same or HIGHER than the fire (heat) detectors themselves... e.g. if the fire detectors are rated at 325degF the wiring must be rated at higher than that (usually 400degF or 600degF) otherwise... the wiring may fail BEFORE the detectors detect the fire!!!

HTH



saludos.
a.

 

[abeltio]

forgot to mention:

the first rule of firefighting is to CONTAIN the hazard... i.e. avoid the spreading of the fire... the blowout panels will allow the fire to spread to other locations.

also: all fans, and air-conditioning and ducting in/out should be shut-off, the ducting usually by means of CO2 actuated dampers, to maintain the required CO2 concentration during the prescribed time.

HTH

saludos.
a.