8 replies to this topic

[Kryz]

Dear All,

If I may can I ask your expert advise on a PSV sizing problem we have come across? We are trying to resolve sizing of a PSV for the MR Surge Vessel, with calculations completed with the PSV sizing program indicating that the valid FIRE GAS EXPANSION sizing case the vessel wall temperature will exceed the limit of the material before the design pressure is reached inside the vessel. The vessel contains mixed refrigerant with the only apparent valid relief case the fire gas expansion (as BLOCKED OUTLET is protected elsewhere). This conclusion has me concerned as according to this calculation there appears to be no valid case where the vessel can reach the design condition which I find hard to believe as this would indicate that no PSV is required! I am also concerned that it may be concluded from this that the materials of construction are not suitable if the vessel wall will fail before the vessel reaches the design pressure under a fire scenario but I guess the vessel material will fail in any case at this temperature regardless of whether a PSV is installed and is relieving. What is not taken into account by the sizing program I guess is also the impact of internal pressure within the pressure at higher wall temperatures and whether this leads to premature failure of the vessel?

WE ARE USING FARRIS SIZE MASTER SOFTWARE

THE MESSAGE:
Sizing Warnings:
SSMCalc.3014:Warning: The calculated relief temperature is greater than the wall temperature, therefore relief will not occur.
User may elect to disregard this overpressure scenario.

[pleckner]

The problem you are experiencing is not uncommon.

If this is an ASME code vessel, then you need a relieving device (PSV, rupture disk, etc.) or some other form of overpressure relief mitigation, e.g. instrumented safety system. This is true even if you can't come up with a credible relieving scenario.

We do not design equipment for fire. By this I mean we expect the vessel and surrounding equipment and piping to have to be replaced after a fire. The vessel and adjacent piping will fail. The object of the relieving device is to allow a controlled release of vessel contents to avoid a bomb. After the contents are boiled off, the vessel will fail. In the case of a vapor containing vessel, then the vessel may very well fail before it can overpressure.

When you lack a credible relieving scenario, I typically install a small 3/4" x 1" PSV and call the relieving scenario thermally blocked-in. You will not have to show any calculations for this type of relieving scenario. You might also consider installing a small rupture disk instead as this may be a cheaper alternative (or it may not; has to be checked).

[JoeWong]

Kryz,
I suggest you read this post again...

http://www.cheresour...?showtopic=3921

You may consider to reduce the PSV set pressure to lowest level as your system allow. Theoritically the relieving temperature will go down...that's it...

Your focus should be shifted to other protection system as highlighted by Phil e.g. RD, instrumented safety system, etc as well as depressuring system, fire proofing, increased fire water deluge density, F&G detection, etc


JoeWong

[Kryz]

Thanks to Phil and JoeWong for your reply.

One of other expert answer to us told about water spray. Is it something that makes sense to you???. And simply do not install any relief device.

OR

Perhaps small 3/4" x 1" PSV will be good solution as Phil suggest. However the problem is this is extremely large vessels approx. 13 m length and 3 m diameter (surge drum with big volume). The area exposed to the fire is very large 126 m2. It produces orifice size like M. (4 inch inlet by 6 inch outlet) much bigger than in Phil email. Sizing Farris software obviously gives us answer but attached this warning about wall temperature.

Perhaps installation big rupture disk would be better option as Phil said. Anyway see response about sprays below, and if you have any comments feel free to respond.

Regards,

Kryz.

His response:
"All the conclusions you have drawn are correct (I am assuming someone has audited the calculations). Under these circumstance you should fit water sprays to douse the vessel. These sprays should be fitted with temperature actuated valves. This arrangement is standard practise for LPG storage tanks, which I am guessing suffer the same fate - not that I have ever had the opportunity to calculate this - hence it remains a guess. You can not as a professional engineer permit a situation that the vessel becomes a potential bomb - not unless you want to enter the hall of shame."

[pleckner]

Your consultant gives good advice about the sprays (as did Joe when he wrote, "increased fire water deluge density".) but I don't see where it says don't install any relieving device. This is simply breaking the law. You must put a relieving device or install some other authorized method of eliminating an overpressure issue if this is an ASME coded vessel.

In your specific case as you presented it in the original post (OP) it seems you have no credible scenarios because even a fire won't overpressure your vessel without it failing first! The relieving device is there to prevent a bomb but if the vessel will not overpressure and just begin to fall apart, this is no bomb. It may be a non-controlled release but it isn't a bomb. So SIZEMASTER is only giving you a fictitious PSV size since the vessel can't really overpressurize if it fails first.

But with all this said I like the combination of Joe's idea to install a deluge system and a depressurization system and then install a minimum sized PSV to accomodate the law. In addition, I strongly suggest you look into dropping the relieving pressure as Joe also suggested.

One last thing and a lesson to all of us. This is why I strongly advocate only people experienced in relief system design perform these analyses. Too many novices just execute software like SIZEMASTER without really knowing how to interpret and use the answers they get.

[Kryz]

Thanks for your notes Phil.

I think the solution would be then install deluging system and fit also small size of PSV.
And ignoring provided by Farris PSV with M orifice.

That is the best what we can do.

Regards,

Kryz.

[JoeWong]

Thanks for your notes Phil.

I think the solution would be then install deluging system and fit also small size of PSV.
And ignoring provided by Farris PSV with M orifice.

That is the best what we can do.

Regards,

Kryz.

Kryz,
Just a quick note, not very sure where your plant is located. I am suggesting you to bring this issue up to your project and people dealing with authority. Get early consensus from authority so that you won't get into trouble in very late stage i.e. construction phase.

My experience informed me that sometime authority official may not think in the same way like we are...in some place...CONSERVATIVE approach is their only direction...


Good luck.

JoeWong

[Kryz]

Hi Joe,

Thanks for your notes. Also I refer to other suggestion given by you and Phil in previous post.

I lowered set pressure as you both suggested. We change from 270 psig to 130 psig as SET P. And for this pressure we do not have anymore problems with warning about relief temperature greater than wall temperature. The determined size of PSV is now with J oriffice. Inlet 2 and outlet 3 inch. Model Farris 26JA10-120.

Perhaps this can be an option.

Project poeple are already informed about the issue anyway. So I will see how we go.

Regards,

Kryz.

[pleckner]

Based on this new information, the PSV approach is now reasonable and I would go with it.

Water deluge is still something you should consider however. It would be nice to be able to not only prevent the PSV from relieving but also protecting the equipment and surrounding area with the water deluge as well.