33 replies to this topic

[S.AHMAD]

Afshin

1. If I were you, I will use risk analysis to determine the need of PSV for fire case.
2. Based on the info from this forum, I believe the probability of occurrence for fire case is remote. The consequence is also remote since it is unmanned and water service (consequence of death/serious injury is remote). Thus the overall risk is small.
3. In analyzing the consequence please take note that the pipeline will also be destroyed. Therefore, from business point of view, make little difference between installing and not.

[ZBR]

Afshin,
I agree with your description in Post no. 21, except for Step 3. The pressure will not decrease after the PSV opens, and the PSV will continue to discharge a small volume of water due to thermal expansion as the temperature rises.
I do not agree with ZBR's point no. 2. There is no need for two PSVs, just one sized for fire relief with steam flow will do the job. Furthermore, water will not continue to flow during a fire, as it is certain that the source of water, which is the water injection pipeline, will be closed by an ESDV on detection of fire.
You say the platform is normally unmanned. If a fire happens when unmanned, there is no fire protection and the filter may rupture, but this is acceptable as no-one is exposed to the risk. If a fire happens when a crew is present, the fire water system will be started, but you have no deluge. The fire crew using the monitors will be exposed to an explosion risk from this filter, so it would be better to provide a small deluge on the filter, or relocate it, to avoid this hazard. Please consult a professional fire protection engineer on this point.
I agree with Fallah's post no. 22.

Paul

Thank you for your comment paul,

However Paul you miss understood my point number 2 . It is important to understand the original question which asks either to consider a psv of smaller size that was sized for solar thermal expansion or use bigger PSV for fire case scenario with user asking if fire scenario should be considered at all because fire scenario psv relieving temmperature is above filter design pressure and line would automatically rupture befoe fire scenario PSV releases.

Now coming to my point number 2 i told him to use fire case sized psv because that relieving temperatue wont be achieved instantaneously with fire. till all the water is evaporated you wont get that kind of relieving temperatures that would fail the pipe................................................................ p
-
-
-
but now i have a question with that ESDV valve shutting water supply.... what are the consequences ....if we dont shut that water down ...... because we may be potentially putting our system to superheating and rupturing pipe line

[paulhorth]

With respect, I must disagree with Ahmad's point in post no. 25 about the problem of chattering if the PSV is sized for fire. Chattering occurs with Gas PSVs because the valve is driven rapidly wide open, and then closes again because the capacity is much larger than the required flow. With a liquid-filled system, before the vaporisation point is reached, the valve will not do this, but will just leak a few drops sufficent to relieve the thermal expansion. So in my view it is OK to fit the larger valve sized for steam vapour relief during fire. Only one PSV is needed. If you have two, the smaller thermal expansion one will have to be set at a lower pressure than the larger one, or they will both open on thermal expansion, which eliminates the hoped-for benefit. However, that could mean that it will leak in normal operation.

So I also agree with ZBR that the fire case is the correct sizing case. But, ZBR also says that the pipeline will rupture if the flow is shut off (as it definitely will be on fire detection). This is not so. The facility that Afshin is discussing is a platform. The water pipeline is below the sea and is not subject to overpressure from fire. The facilities on the platform are a pig receiver, a filter and a water injection manifold. It is the blocked-in filter which is at risk of rupture. The manifold is also at risk, but the volume contained in this will be relatively small. Afshin has said that there is currently no deluge, and I am recommending that the filter has a small deluge installed.


Paul

[Afshin445]

With respect, I must disagree with Ahmad's point in post no. 25 about the problem of chattering if the PSV is sized for fire. Chattering occurs with Gas PSVs because the valve is driven rapidly wide open, and then closes again because the capacity is much larger than the required flow. With a liquid-filled system, before the vaporisation point is reached, the valve will not do this, but will just leak a few drops sufficent to relieve the thermal expansion. So in my view it is OK to fit the larger valve sized for steam vapour relief during fire. Only one PSV is needed. If you have two, the smaller thermal expansion one will have to be set at a lower pressure than the larger one, or they will both open on thermal expansion, which eliminates the hoped-for benefit. However, that could mean that it will leak in normal operation.

So I also agree with ZBR that the fire case is the correct sizing case. But, ZBR also says that the pipeline will rupture if the flow is shut off (as it definitely will be on fire detection). This is not so. The facility that Afshin is discussing is a platform. The water pipeline is below the sea and is not subject to overpressure from fire. The facilities on the platform are a pig receiver, a filter and a water injection manifold. It is the blocked-in filter which is at risk of rupture. The manifold is also at risk, but the volume contained in this will be relatively small. Afshin has said that there is currently no deluge, and I am recommending that the filter has a small deluge installed.


Paul

Paul,

Thnk you for your complete answer.I am also agree with you for using one PSV sized for Fire case.
My concern is about deluge.It can be use for filter body cooling during fire or easily can be use as a fire extinguisher?
In this regards, it can only works when fire water ring is connected to supply boat, otherwaise fire water is dry and deluge can't work.

Afshin

Edited by Afshin, 11 January 2012 - 08:03 AM.

[martin_h]

Interesting that we should all be thinking that the filter will fail before it reaches 357 C. Looking at the allowable stresses for the metal in question (I looked in API B31.3 for the data), I see the allowable stress at 100F is 23.3 KSI and at 700 F it is 18.3 KPI. These temperatures are well below the creep region and with a small reduction in allowable stress like that indicated the probability of failure is vanishingly small. I suggest you discuss this with a mechanical engineer next to the vessel drawings and get his/her opinion.

[Afshin445]

Interesting that we should all be thinking that the filter will fail before it reaches 357 C. Looking at the allowable stresses for the metal in question (I looked in API B31.3 for the data), I see the allowable stress at 100F is 23.3 KSI and at 700 F it is 18.3 KPI. These temperatures are well below the creep region and with a small reduction in allowable stress like that indicated the probability of failure is vanishingly small. I suggest you discuss this with a mechanical engineer next to the vessel drawings and get his/her opinion.

But as fallah says in Post # 20 till filter have liquid water inside it can't fail due to water vaprization.Am I right Fallah?

Afshin

Edited by Afshin, 17 January 2012 - 01:07 AM.

[fallah]

But as fallah says in Post # 20 till filter have liquid water inside it can't fail due to water vaprization.Am I right Fallah?

Afshin

Afshin,

As long as there is liquid water in filter the relieving temperature would be constant at 357C. Hence, if the filter can withstand (based on material and thickness) this temperature at relieving pressure (148.3 barg) you are completely right.

Fallah

[Afshin445]

In this case, it's not better the filter thickness(Mechanical design) calculated based on filter relief condition? Then we are ensure filter can sustain in relief condition and then releif valve can consider as a protection.

Afshin

[fallah]

In this case, it's not better the filter thickness(Mechanical design) calculated based on filter relief condition? Then we are ensure filter can sustain in relief condition and then releif valve can consider as a protection.

Afshin

Afshin,

It is not absolutely better!

From economical/practical standpoint, the relieving temperature of the fire case wouldn't be the base of vessel's mechanical design at all.

Indeed, suppose you take first relieving temperature (357C) as design temperature of the vessel to protect it by fire case PSV , then what do you do for vessel protection at higher and higher relief temperatures would be occured after vaporization of the last liquid droplets inside the vessel?

Fallah

Edited by fallah, 19 January 2012 - 12:07 AM.