8 replies to this topic

[kpavan]

Hi all,
Here is a topic which was continuously on discussion forums for a long time, but once again I would like to put this on this forum hoping some more output will come.
I have exchangers which process vapor on both sides of the exchanger. Normally I consider Thermal expansion for tube side (when tube rupture case is not a valid one) and if the shell side consists of liquid then I would never have considered fire case for tube side, but as this particular exchanger consists of vapor on both sides I have evaluated fire case also by taking the total surface area of tubes along with front and rear head of the exchanger. And I am getting fire case as the governing case.
Expecting your valuable comments/suggestions on this topic.

Thanks in advance.

Regards,
Pavan

[fallah]

I have evaluated fire case also by taking the total surface area of tubes along with front and rear head of the exchanger. And I am getting fire case as the governing case.

Because in fire case the tube side isn't subject to fire,no need to consider total surface area of tubes in your evaluation.

[latexman]

Direct fire exposure to the two heads on the ends will definately occur.

Heat transfer from the fire through the shell and into the tubes can be significant. I have also seem where the heat from the fire to the shell limits the heat transfer, i.e. the shell fluid was capable of transferring more, but could not due to the heat from the fire being limited by the shell area.

Sum these two quantities of heat to get a more realistic answer.

Edited by latexman, 02 September 2009 - 09:22 AM.

[JoeWong]

Direct fire exposure to the two heads on the ends will definately occur.

Heat transfer from the fire through the shell and into the tubes can be significant. I have also seem where the heat from the fire to the shell limits the heat transfer, i.e. the shell fluid was capable of transferring more, but could not due to the heat from the fire being limited by the shell area.

Sum these two quantities of heat to get a more realistic answer.

I agree with Latexman opinions. Direct fire exposure to tube heads and heat transfer from fire to fluid in shell and further transfer to tube side is credible when shell side operating pressure at temperature same as relieving temperature in tube side not exceeding shell side PSV setting. Otherwise, shell side PSV may open prior to opening of tube side PSV...

I have exchangers which process vapor on both sides of the exchanger. Normally I consider Thermal expansion for tube side (when tube rupture case is not a valid one) and if the shell side consists of liquid then I would never have considered fire case for tube side, but as this particular exchanger consists of vapor on both sides I have evaluated fire case also by taking the total surface area of tubes along with front and rear head of the exchanger. And I am getting fire case as the governing case.
Expecting your valuable comments/suggestions on this topic.

I have different opinion...

Liquid normally having higher heat transfer capacity compare to vapor or gas. Higher heat transfer rate from shell side to tube side if liquid is present in shell side (compare to vapor or gas). In the event of liquid in shell side, there is higher potential of heat transfer and higher tendencies to consider total surface area of tubes...

[kpavan]

Thanks fallah, Latexman & Joe for your reply.

As suggested by Latexman & Joe wong
I have calculated the fire case for tube side by taking surface area of shell and two heads. And still it is the governing case but requries a smaller orifice area compared to the one i have calculated earlier (By taking total surface area of tubes instead of shell). Ofcourse this is because shell surface area is less than the total tube surface area.

I have exchangers which process vapor on both sides of the exchanger. Normally I consider Thermal expansion for tube side (when tube rupture case is not a valid one) and if the shell side consists of liquid then I would never have considered fire case for tube side, but as this particular exchanger consists of vapor on both sides I have evaluated fire case also by taking the total surface area of tubes along with front and rear head of the exchanger. And I am getting fire case as the governing case.

I have different opinion...

Liquid normally having higher heat transfer capacity compare to vapor or gas. Higher heat transfer rate from shell side to tube side if liquid is present in shell side (compare to vapor or gas). In the event of liquid in shell side, there is higher potential of heat transfer and higher tendencies to consider total surface area of tubes...

Joe,
I agree with you that normally liquids have high heat transfer capacity, and this is the same reason which made to think not to have a fire case in case if liquid is in shell side. As liquid is capable of taking more heat and even requires more heat for converting from liquid to vapor, so shell side material is saved for more time with PSV of it when compared to gas in shell side.
Adding to this, one of the post (didn't remember exactly where i saw) indicated that it is possible a fire could pretty destoy the exchanger from the shell side in.

I may be mistaken, but these are the assumptions which i have taken after reading many posts.

[fallah]

think not to have a fire case in case if liquid is in shell side. As liquid is capable of taking more heat and even requires more heat for converting from liquid to vapor, so shell side material is saved for more time with PSV of it when compared to gas in shell side.

Considering fire case for a heat exchangers (e.g. for its shell side) dosen't depend on if liquid or gas is in the shell side.But in the case of the liquid,as you mentioned,distruction of the shell side would be postponed.

[fallah]

Heat transfer from the fire through the shell and into the tubes can be significant. I have also seem where the heat from the fire to the shell limits the heat transfer, i.e. the shell fluid was capable of transferring more, but could not due to the heat from the fire being limited by the shell area.

In the case mentioned by post originator (particular exchanger consists of vapor on both sides) heat transfer from the fire through the shell and into the tubes can not be significant with respect to the case with liquid in the shell.

Your statement as "I have also seem where the heat from the fire to the shell limits the heat transfer, i.e. the shell fluid was capable of transferring more, but could not due to the heat from the fire being limited by the shell area." seems being an obvious phenomena in any exchanger subject to fire.Please explain more if my understanding isn't completely right.

Edited by fallah, 03 September 2009 - 02:50 AM.

[kpavan]

think not to have a fire case in case if liquid is in shell side. As liquid is capable of taking more heat and even requires more heat for converting from liquid to vapor, so shell side material is saved for more time with PSV of it when compared to gas in shell side.

Considering fire case for a heat exchangers (e.g. for its shell side) dosen't depend on if liquid or gas is in the shell side.But in the case of the liquid,as you mentioned,distruction of the shell side would be postponed.

Fallah,
Thanks for your reply. Here my intention is only about fire case consideration for tube side not for shell side. That's true, fire case consideration doesn't depend on shell side fluid, but shell side fluid will effect tube side fire consideration.

Edited by kpavan, 05 September 2009 - 03:44 AM.

[fallah]

Fallah,
Thanks for your reply. Here my intention is only about fire case consideration for tube side not for shell side. That's true, fire case consideration doesn't depend on shell side fluid, but shell side fluid will effect tube side fire consideration.

Kpavan,

Thanks for your clarification.

One new point: Although fire case can be credible for tube sides of S&T heat exchanger in some cases e.g. liquid in shell side, it has not been mentioned any point about this matter in API 521 sec. 5.15 (External pool fires) other than those points submitted regarding Air cooled exchanger (sec. 5.15.7).What is your opinion in this regard?
Of course,I have involved in calculation of relief load for tube side of the S&T heat exchanger (e.g. depropanizer reboiler) in fire case without considering the heat may pass through surface area of tubes even though there was liquid in the shell.

Actually in the case of liquid in the shell most of heat from fire consumed as latent heat of vaporization (especially in almost pure component) and when all liquid to be vaporized (gas expansion case) even though the gas temperature would be highly increased,because of low heat transfer capacity of the gas, the transferred heat would be low in comparison with the heat to be directly transferred via wetted wall of tube side(head,channel,piping pot,....).

Anyway, even though seems as per above the heat to be transferred via liquid in the shell to tube side content in fdire case may not be considerable in most cases,it is worth to evaluate the rate of such heat transfer in any case to prevent undersizing the relevant PSV.

Edited by fallah, 05 September 2009 - 12:46 PM.