9 replies to this topic

[Lowflo]

A colleague of mine suggested that fire exposure should generally be considered as an credible relief scenario for the water side of exchangers. In my opinion, the only cases where this is justified are those where: (1) the water flow is very low , or (2) the water flow can be stopped. Otherwise, the risk of overpressure due to fire is trivial....certainly not worth using as a basis for a relief valve size.

Is there an industry-consensus answer to this question? I didn't see any discussion on this in API-521.

This has probably been discussed before in this group. If so, please excuse me. I just joined this forum.

Thanks.

[CMA010]

For evaluating a fire case it is generally considered that the equipment is blocked in (if this is possible obviously), and therefore all flows to and from the equipment are stopped.

However, rightly of wrongly, external fire is usually not considered for heat exchangers because it is deemed that heat exchangers are difficult to block-in. External fire is furthermore usually not considered for the tube side because it is for the most part "shielded" from the heat of the fire by the shell side.

Whether these assumptions apply depends on the specific situation and views, opinions, insights from the responsible engineers.

[fallah]

However, rightly of wrongly, external fire is usually not considered for heat exchangers because it is deemed that heat exchangers are difficult to block-in.

If the design pressure of the exchanger shell would be higher than 15 psig and to be designed as per ASME,and also located in fire zone external fire should be condidered.

[JoeWong]

For evaluating a fire case it is generally considered that the equipment is blocked in (if this is possible obviously), and therefore all flows to and from the equipment are stopped.

However, rightly of wrongly, external fire is usually not considered for heat exchangers because it is deemed that heat exchangers are difficult to block-in. External fire is furthermore usually not considered for the tube side because it is for the most part "shielded" from the heat of the fire by the shell side.

Whether these assumptions apply depends on the specific situation and views, opinions, insights from the responsible engineers.

The underlined sentence and bold word would be a result/conclusion of analysis. However, it is not the first consideration nor assumption.

I believe fire contingency shall be analyzed if there is still possibility. However, it may be eliminated by Quantitative Risk Analysis.

[astro]

Is there an industry-consensus answer to this question? I didn't see any discussion on this in API-521.

Depends where you look. Here's a direct quote from API STD 521 5th Ed. Hydraulic Expansion - Causes, s5.14.1 with emphasis in the relevant spots thrown in by me:

Hydraulic expansion is the increase in liquid volume caused by an increase in temperature (see Table 3). It can result from several causes, the most common of which are the following.
a) Piping or vessels are blocked in while they are filled with cold liquid and are subsequently heated by heat tracing, coils, ambient heat gain or fire.

Lower down in the same section there is the following:

In certain installations, such as cooling circuits, the processing scheme, equipment arrangements and methods, and operation procedures make feasible the elimination of the hydraulic-expansion relieving device, which is normally required on the cooler, fluid side of a shell-and-tube exchanger.

There's a further caution regarding taking credit for locked open block valves and the need for reliance on administrative controls.

The other point about tube or shell side consideration. The key criteria for external pool fires is exposed area. The tube side has a smaller exposed area that is a function of the head(s) of the vessel. However what do you do if the shell side is vapour and the tube side is liquid? Liquid relieving rates will generally be higher than vapour on a per unit area basis, so the vapour relief rate could end up being higher than the liquid. In any case, this consideration is irrelevant because the relief contingencies affect different process systems.

I don't see how you can disregard the analysis of one side or the other, unless you can take credit for a mitigating effect such as an open circuit situation like a cooling water system.

Could I be cheeky enough to say "case closed"?

[CMA010]

However, rightly of wrongly, external fire is usually not considered for heat exchangers because it is deemed that heat exchangers are difficult to block-in.

If the design pressure of the exchanger shell would be higher than 15 psig and to be designed as per ASME,and also located in fire zone external fire should be condidered.

See UG-133, ©.

Rightly of wrongly the thought behind is that heat exchangers are usually not equipped with remote operated block valves and will have to be blocked-in manually by an operator. Expecting an operator doing so in the case of a fire is hardly realistic. Without the exchanger being blocked-in there remains an open path.

To be clear: I'm merely stating the standard practices of some big, well respected operating firms in regard to overpressure protection of heat exchangers. I'm not saying that it is the standard to follow, the responsible personnel will have to make that judgment.

Personally i feel that Joe (relation of Wing Y Wong?) is correct, QRA.

[fallah]

[/quote]

See UG-133, ©.

Rightly of wrongly the thought behind is that heat exchangers are usually not equipped with remote operated block valves and will have to be blocked-in manually by an operator. Expecting an operator doing so in the case of a fire is hardly realistic. Without the exchanger being blocked-in there remains an open path.

To be clear: I'm merely stating the standard practices of some big, well respected operating firms in regard to overpressure protection of heat exchangers. I'm not saying that it is the standard to follow, the responsible personnel will have to make that judgment.

Personally i feel that Joe (relation of Wing Y Wong?) is correct, QRA.[/quote]
If the fire scenario would be applicable for an equipment such as HEX,then in fire case it should be capable to be isolated by SDVs from its adjacent equipment.

[CMA010]

[/quote]
If the fire scenario would be applicable for an equipment such as HEX,then in fire case it should be capable to be isolated by SDVs from its adjacent equipment.
[/quote]

Why would (should???) there be ESD valves be located between an exchanger and adjacent equipment?

[fallah]

Why would (should???) there be ESD valves be located between an exchanger and adjacent equipment?

Part of hydrocarbon plant to be subjected to fire,including one or several equipment,should be isolated from adjacent parts by ESDVs and SDVs in order to being depressurized through relevant BDV to flare or safe location.
Performing above operation would prevent fire propagation to other parts.

[JoeWong]

CMA010 / Fallah
Fallah has tried to point out that the heat exchanger is within a system. The Shutdown valves (SDV) isolates this system from other system. In case of fire adjacent to this system, Shutdown system is activated and isolated this system. This system is completed isolated. Anything equipment and piping within this system will potentially expose to fire. The SDV is preventing hydrocarbon liquid from another system entering this system.

Now come to fire on an equipment. PSV is just to buy time for evacuation, initiation of all protection system, etc. However, it don't really protect the equipment. The equipment will fail at the end if continuous fire impingement.

Now the next question is how much time this PSV can buy ? This really subject to heat flux, how the flame impinging the equipment, how great is the flame momentum impinging the equipment, etc. If conservative approach is taken, then the time is a rather short. If the time is so short, there is minimum value in providing this PSV.

However, if we look at it from different perspective. What if the heat flux is not too great, not at the correct angle, etc. It take some time before the vessel failed. This period of time would probably sufficient for operator to take quick action, inventory evacuation, etc. In this perspective, the PSV is valuable.

I have many chances to discuss this issue with those engineer in the "...big, well respected operating firms". I just sorry to inform that the conclusion are not consistent throughout...This made me believe "this is another art of the sciences".