24 replies to this topic

[hdsattar]

Hello all
can any body tell me how could i calculate "exposed surface area" for a heat exchenger (double pipe) in a fire case. both sides of the exchenger are feed gas, so one of the cases for relief load calculation is the case of fire. but I am not sure how to calculate the exposed surface area for the tube side.
I would appreciate if any one could help.

[Art Montemayor]

This thread belongs in the relief devices Forum.

A gas-to-gas, double-pipe heat exchanger should be protected during a fire case. The normal, usual way to ensure this is to vent all combustible gas from the system once a fire is detected. That means the exchangers should not be blocked-in, with pressurized gas. They should be empty.

I would still install a PSV on the exchanger – depending on whether there are other cases besides the fire case. But the fire case is protected by venting. The reason for this is that it is naïve to believe that if you have blocked-in gas inside the exchanger (you certainly are not going to have flowing gas inside the exchanger during a fire case) that you could, within a reasonable short period, develop a significant gas pressure rise inside the unit due to solely natural convection currents. Gas, of itself, is an excellent thermal insulator and is the main substance that allows all solid insulation to work as well as it does. Gas’ thermal conductivity is one of the worst in nature and is the reason for its insulating properties. Therefore, static gas will act as a natural insulating barrier to heat flow and the steel shell of the exchanger will reach very high values before any natural, internal convection currents start to stimulate heat transfer to the gas and make the internal pressure rise. By the time this probably takes place, your steel shell may be in the range of total collapse due to its weakened stress condition. Relying on a PSV to save you and the exchanger is folly in my opinion. Direct venting is much safer – and any connected gas pressure vessel(s) are also protected at the same time.

[ankur2061]

Art,

So well said as always. It is always a pleasure to read your lucid and clear explanation compared to the incoherent nonsense presented by some members writing all too frequently on the forum.

Regards,
Ankur.

[Lowflo]

If the vessel contaings only gas or supercritical fluid, it's usually a waste of time and money to size a safety valve for fire exposure. And worse, it gives the owner a false sense of security. The reality is that PSVs on such vessels don't provide any meaningful protection at all - regardless of the PSV size. The vessel wall temperature will continue to rise, unabated, and the vessel can fail in 10-15 minutes or less if it's enveloped by the fire.

Unfortunately, too many engineers proceed with sizing these relief valves for gas thermal expansion, without recognizing the dangeous futility. Instead, they should be pointing out the real hazard so the owner can do a risk assessment, and implement effective protective measure according to that risk.

As Art points out, de-pressurization is very effective protective measure. It reduces the stress on the vessel walls while also reducing the consequences if/when the vessel does eventually fail. Other measure to consider are water spray and fire-resistant insulation. Depending on the specific case (risk), one might need one, two, or all three of these layers of protection.

[fallah]

hdsattar,

You should provide, at least, following information to get a clear response regarding if fire case is applicable to your case and then if so, what would you do to protect the equipment:

- Elevation of the exchanger respect to local grade
- Type of fluid as feed gas
- If the exchanger could be isolated in fire case
- If schedule pipe has been used in exchanger or gauge tubes
- The trapped volume of the gas when exchanger is isolated

Depending on the system volume and configuration, it may be considered same as a piping system then fire case may not be applicable to it.
Obviously, if fire case would be applicable and the need to PSV/TSV installation would be justified, how to calculate the "exposed surface area", will be among last needed information for relief load calculation.

Fallah

[Propacket]

Well said lowflo. If the equipment is engulfed by fire, one should not wait for vapors to expand and then PSV to pop.The vessel may fail before the PSV lift.

[hdsattar]

Thak you all

[simadri]

Dear Art,
I have read your explanation regarding how PSV can not save the HE for fire case. It is very interesting. But I have few doubts.

1. I understood that HE will rupture due to temperature increase in fire case, not due to high internal pressure of gas resulting into gas expansion. I am not sure that what I have understood, that is correct.

2. If that is correct, then how providing a simple vent can resolve the issue?

3. A vent on heat exchanger will have a manual valve. But at the time of fire, who will open the valve if there will be no personnel working at the area?

4. The vent shall be provided on the heat exchanger or discharge line?

I hope, you will educate me in these regards.

Thanking You
Simadri Acharya

[Art Montemayor]

Simadri Acharya:

The following are my responses (in RED) to your further queries on the fire case imposed on a gas-filled pressure vessel:

1. I understood that HE will rupture due to temperature increase in fire case, not due to high internal pressure of gas resulting into gas expansion. I am not sure that what I have understood, that is correct.
The heat exchanger need not “rupture”. It may fail by collapsing, and fall – but it will not rupture because it would not have any internal pressure upon failure. The basis for this is that in the event of a fire case, the process is immediately shut down and all pressurized vessels (including the gas-filled heat exchanger) would be blocked-in (shut off by automatic valves) and the contained gas safely vented to a flare system or to atmosphere (if need be) by a remotely operated automatic valve. This is considered standard procedure in fire case procedures within process plants.

2. If that is correct, then how providing a simple vent can resolve the issue?
The venting is done by remotely actuating (normally at a central control room) an automatic vent valve connected to the heat exchanger.

3. A vent on heat exchanger will have a manual valve. But at the time of fire, who will open the valve if there will be no personnel working at the area?
As explained above, the operation is remote – not localized. This is a standard procedure for activating vent valves in the event of a fire case.

4. The vent shall be provided on the heat exchanger or discharge line?
You are trying to vent the heat exchanger’s internal pressure, so the obvious and common sense application is to vent the heat exchanger directly.

The purpose of venting is not to “save” the heat exchanger; rather, it is a step taken to prevent the catastrophic rupture of a gas-filled pressure vessel in the event fire fighters are in the immediate area. It is meant to save human lives.

A liquid-filled pressure vessel does not present the same potential hazard as a gas-filled vessel during a fire. As long as the vessel contains liquid, the vaporization (“boiling”) of the contained liquid inside actually works to keep the vessel walls relatively cool by the creation of internal vapor that is being continuously vented (relieved) by a properly designed PSV on the same vessel. A blocked-in, liquid-filled pressure vessel is actually a convenient heat sink during the fire case. BUT, it should be remembered that this is only during the time that it contains liquid. Once the liquid is depleted (due to a prolonged fire case), the vessel becomes a hazardous, blocked-in, gas-filled vessel subject to wall failure and rupture under pressure – a very dangerous scenario that requires complete evacuation of all human beings at this time.

I hope the above design basis of how I approach the design for a pool fire case has been explained to your satisfaction and understanding. As Lowflo has so well explained, the proper and further steps an owner can take to prepare for such a possible scenario can be applied by the engineer who should always apply a proper and well-designed safety system to avert or mitigate the catastrophic effects of a pool fire within a process.

[paulhorth]

Art,

I agree with all your comments in red in the last post.
With respect, I would like to add something to your remarks about a vessel containing liquid in a fire. The liquid does act to keep the wetted walls cooler by absorbing heat, but not the dry walls above the liquid level. If the internal pressure is not reduced by depressuring, and only a PSV is provided, the exposure to fire of the walls not wetted by the liquid can lead to what is termed a BLEVE (standing for Boiling Liquid Expanding Vapour Explosion) which can have severe consequences. The classic example was a propane sphere somewhere in Texas I believe, in the early 1970s, which led to the term BLEVE entering the literature. The sphere was being sprayed by firefighters and many of them were killed. I think a Google search might turn up the details.
For this reason it is important to provide both water deluge and depressuring for ALL hydrocarbon pressure vessels including the liquid-containing ones. As many others have already said, a PSV provides no protection duing a sustained fire.
Depressuring also is an important safeguard before a fire has started, if a gas leak has been detected.

Paul

[Art Montemayor]

Paul:

Excellent, accurate, and a very important point that you have covered correctly. I did not emphasize enough the importance that the size, location of the fire and the extent to how much the internal wall surface is wetted plays a very important part in analyzing the potential for a catastrophic event. I also failed to add that the liqud level is differentially disappearing in a liquid vessel during the course of the fire and thus, is increasingly more apt to rupture as time goes by - due to increasing dry wall exposure to the fire.

In my opinion, this topic cannot be emphasized enough and I am indebted to you for the valuable note(s) that you have contributed in this thread. It was my intent to have this subject aired in our Forum when I refused to respond with my comments to Simadri's personal message via my messenger service - simply because I believe this subject is too important to discuss only between two members instead of having the entire Forum membership available to read, discuss, contribute, and learn from these valuable experience lessons.

I hope every member reading this thread is mindful of the profitable key learnings that can be acquired through the valuable participation by active and experienced members like yourself. Thank you. You deserve the recognition and gratitude from all members reading this thread.

[simadri]

Dear Paul & Art,
Thank you very much for sharing the knowledge. I am sure, not only I, but also other members will be benefited from your posts.

Dear Art,
As I have already told to you, at that time I was unaware about the use of messenger and to ask a doubt in a blog created by someone, is right in the forum or not. But thank you very much for clearing me about the way of communication.

Still I have certain doubts. Please bear with me.

If we will provide a vent on heat exchanger, then also the gas inside HE (after pressure will be balanced with atmosphere) will form a natural insulating barrier to heat flow and the steel shell of the exchanger will reach very high values before any natural, internal convection currents start to stimulate heat transfer to the gas and make the pressure rise. Hence the steel shell may collapse.

For liquid handling equipment, can we provide PSV for fire case as the wetted part area will decrease with time as liquid will be vaporized? Shall we put a PSV and an emergency venting valve for depressurizing? How the system will work in case of fire?

Thanking You
Simadri Acharya

[fallah]

For liquid handling equipment, can we provide PSV for fire case as the wetted part area will decrease with time as liquid will be vaporized? Shall we put a PSV and an emergency venting valve for depressurizing? How the system will work in case of fire?

simadri,

To meet the code requirement you should provide PSV for all pressure vessels regardless of liquid or gas handling function.

For liquid HC, normally there is no need to consider BDV. Liquified HC's such as LPG is among exceptions and the pressure vessels cotaining them should be equipped with BDV along with the already provided PSV.

Fallah

Edited by fallah, 23 May 2012 - 04:20 AM.

[paulhorth]

Art,

Thank you for your most generous comments. My points would also be made by at least a dozen colleagues, who would have expressed the issue better, had they not been busy on real work and so not following this forum.

Fallah,

I am sorry if I have misunderstood you, but in your statement

For liquid HC, normally there is no need to consider BDV.

you seem to be proposing that vessels containing both liquid and vapour hydrocarbons, there is no need to provide a depressuring valve (BDV). Such vessels include oil-gas separators and compressor KO drums as used in all oil production facilities. These items should ALWAYS be provided with BDVs, if they operate at pressure greater than 6.9 barg, for the reason given in my earlier post.
LPG pressurised storage bullets are a special case as it it sometimes difficult to arrange a flare to accept depressurisation. Other means, such as earth covering, should then be used to protect these vessels against fire.

Paul

[fallah]

Fallah,

I am sorry if I have misunderstood you, but in your statement

For liquid HC, normally there is no need to consider BDV.

you seem to be proposing that vessels containing both liquid and vapour hydrocarbons, there is no need to provide a depressuring valve (BDV).

Paul,

Actually, when in response to simsdri's post about "Liquid handling equipment..." i mentioned "For liquid HC....", i meant the liquid filled equipment that can be isolated and are subject to fire.
In such equipment if would subject to a fire, the relevant PSV is rapidly reached to its set pressure due to liquid expansion following the sensible heat absorbtion. After initial relieving, because the vapor space of the vessel would be very small part of the vessel and gradually becomes larger due to simultaneous relieving and latent heat absorbtion, the system can keep the relieving pressure and the PSV can continue to be opened. Then the PSV itself functions as a depressurizing facility from the beginning of the fire occurence while adequate wetted area would protect the vessel from failure, and there is no need to BDV beside mentioned PSV.

Fallah

Edited by fallah, 24 May 2012 - 10:01 AM.

[paulhorth]

Fallah,

OK - for a liquid-filled vessel, I agree with you. No BDV is necessary.
I had indeed misunderstood what you said, and I apologise for that.

Paul

[simadri]

Thank you Fallah.

Then when can we use 'vapor expansion due to fire case' for PSV calculation as we shall go for BDV for gas service?

Dear Art,
If we will provide a vent on heat exchanger, then also the gas inside HE (after pressure will be balanced with atmosphere) will form a natural insulating barrier to heat flow and the steel shell of the exchanger will reach very high values before any natural, internal convection currents start to stimulate heat transfer to the gas and make the pressure rise. Hence the steel shell may collapse.

[fallah]

Thank you Fallah.

Then when can we use 'vapor expansion due to fire case' for PSV calculation as we shall go for BDV for gas service?

simadri,

As many times mentioned in this forum, the PSV itself, in most fire cases, may not be able to protect the vessel against failure especilly in gas services. But it should be considered to meet the code requirement.

Anyway, the BDV and PSV have different functions; the former is activated in the case of an emergency depressurization and the latter in any case the equipment is subject to overpressure.

Fallah

[Art Montemayor]

Simadri:

I thought I addressed this point, so I'll repeat it again:

The heat exchanger need not “rupture”. It may fail by collapsing, and fall – but it will not rupture because it would not have any internal pressure upon failure. The basis for this is that in the event of a fire case, the process is immediately shut down and all pressurized vessels (including the gas-filled heat exchanger) would be blocked-in (shut off by automatic valves) and the contained gas safely vented to a flare system or to atmosphere (if need be) by a remotely operated automatic valve.

What I mean by "collapse" is that the steel vessel may reach such high external metal temperatures that the metal actually starts to "soften" - begins melting. Under this type of exposure, the vessel collapses without any internal pressure because its own weakened state can support it no longer - it literally falls to the ground or grade in a deformed state. I have witnessed the collapse of a 110 ft distillation column after a catastrophic fire and it wasn't very pretty. Several persons resulted dead from this scenario. Such things have a way of instilling you permanently with the will to never allow such a thing to happen on your watch.

[simadri]

Dear Fallah & Art,

Thank you for your explanations. I am citing my understanding hereby after reading posts from both of you. Please bear me if I am wrong.

I understood that in gas service, PSV does not help because there is no internal pressure rise in the equipment due to external fire. Depressurizing is a better option through vent valve. But we shall provide PSV on such vessels to meet code requirements. API-521 also states equations for PSV relieving rate calculation for gas expansion case caused by external fire.

[fallah]

simadri,

My comments with red color as below:

Dear Fallah & Art,

Thank you for your explanations. I am citing my understanding hereby after reading posts from both of you. Please bear me if I am wrong.

I understood that in gas service, PSV does not help because there is no internal pressure rise in the equipment due to external fire (there is internal pressure rise due to external fire but the rate wouldn't be so high to activate the PSV before equipment body failure). Depressurizing is a better option through vent valve (In fact, it isn't an option in gas expansion case, because it shall be applied beside the PSV to prevent equipment failure due to delay in PSV activation). But we shall provide PSV on such vessels to meet code requirements. API-521 also states equations for PSV relieving rate calculation for gas expansion case caused by external fire.

Fallah

[Lowflo]

I hate to see threads drag out forever, but there still appears to be confusion and misunderstanding.

1. During fire exposure, a PSV is incapable of sufficiently protecting a vessel filled with vapor, or supercritical fluid, and that is true regardless of whether the PSV is activated or not.

Why?

• Unless boiling (Hvap) is occurring inside the vessel, there is no chance that heat can be removed (through the PSV) from the vessel at the same rate that it's being absorbed from the fire. With a vapor filled vessel, the fire heat is being transferred to the internal vapors. But, vapor has very little capacity for absorbing heat. That vapor increases in temperature and expands, but the amount of heat it absorbs is trivial.
• This means that the heat-in from fire far exceeds heat-out through the PSV. Therefore the wall temperature is rapidly increasing.
• Even if the internal pressure reaches the PSV set pressure, the PSV will open and then re-close, maintaining the pressure inside the vessel while the vessel temperature continues to rise.
• While the PSV opens and re-closes, the amount of heat removed is insignificant. The vessel temperature is still rising, and the PSV isn't doing anything to stop that.
• The hot metal walls will rapidly approach their yield stress, and then fail.
• This failure can be catastrophic because it occurs while the vessel is still pressurized.

2. Unlike PSVs, a de-pressurization valve provides real protection for vapor-filled vessels during fire exposure.

Why?

• A de-pressurization valve reduces the internal pressure, which in-turn reduces the stress on the vessel walls.
• The vessel wall temperature is still rising, just like the vessel protected by a PSV, but this reduction in stress buys you some time.
• If the fire continues, this vessel will also fail, but the failure occurs at a lower pressure which reduces the consequences of that failure.

[simadri]

Dear Lowflo,
Thank you very much for the detailed reply.
I understood well how deprsserizing is effective as compared to PSV for gas-filled pressurised vessels. But in fire case, how can we save the support of the vessel? Because if the support softens due to heat, absorbed from fire, then also the equipment may collapse.

[Art Montemayor]

Simadri:

At the cost of being redundant, I will repeat what has previously been noted by Lowflo and Paulhorth: You are free to apply mitigation design or procedures - such as:

• You can apply a concrete coat to the steel skirt or steel supports of the vessel; this forms an insulating coat for the support steel and is commonly used in many processing plants;
• You can apply a water deluge system, or water sprinklers, to the affected steel structure in order to absorb the extreme heat from the fire and allow the steel to remain relatively "warm" - but still strong enough to support an empty vessel.

I hope this explains what we consider to be normal, conventional techniques to mitigate the negative effects of a pool fire.

[simadri]

Art:
Thank you for answering me.