Jump to content



Featured Articles

Check out the latest featured articles.

File Library

Check out the latest downloads available in the File Library.

New Article

Product Viscosity vs. Shear

Featured File

Vertical Tank Selection

New Blog Entry

Low Flow in Pipes- posted in Ankur's blog

Steam Side During Fire Scenario


This topic has been archived. This means that you cannot reply to this topic.
5 replies to this topic
Share this topic:
| More

#1 Guest_Guest_*

Guest_Guest_*
  • guestGuests

Posted 08 December 2003 - 05:14 PM

When accounting for the steam side of an exchanger during a potential fire, do we assume that the fire is heating water vapor and have to relieve that?

Or do we have to assume some quantity of water liquid in the shell and do it as a boil-off calculation?

Thanks!

#2 pleckner

pleckner

    Gold Member

  • ChE Plus Subscriber
  • 564 posts

Posted 08 December 2003 - 08:07 PM

Steam systems will be protected with their own relief valve. Relief valves on steam feed into the exchanger shells are not required, in my opinion, unless the shell is designed for a lower pressure than the header. And the engineer who does that should be shot!

OK, but what about the blocked-in case. Ahh, now you're talking. And if the steam side is blocked-in, it is probably because the unit was taken out of service and any steam in the shell will condense. So, the answer is, size for boiling water. How much? Obviously, the steam will not fill the entire shell when concensed but heck, just use the entire shell for the wetted surface area and get a little conservatism out of it. Just round low rather than high during your calculations.

#3 Guest_Guest_*

Guest_Guest_*
  • guestGuests

Posted 09 December 2003 - 02:14 PM

Phil,

Thanks for the response. This exchanger is a "bayonet" style inserted through into the side manway to heat a rather large tank and the steam is inside the tubes.

The system appears fairly typical in that the supply is a regulator followed by a relief device set for the MAWP (175 psig) of the exchanger. Downstream of this is a temperature control valve, a block valve and then the exchanger. The condensate outlet of the exchanger is a block valve and then a trap to the condensate header to a pumping trap.

The relief device review stated the scenario as steam side of exchanger (tube) becomes water full and then if there is a fire, we need to relieve for the expanding water (I was just glad it was NOT deemed to be two phase flow too because they were assuming it to be liquid full!). I consider this to be double jeopardy, trap failure and fire at the same time.

Heat Transfer will be limited due to a fire (tubesheet only of the bundle) and the wetted surface area is basically very small. Even if I assume this as the wetted surface area, there is minimal water mass inside the exchanger.

I am inclined to state that the most credible scenario is failure of the upstream regulator and document that if these upstream PSV's are adequate, the exchanger is protected and end of story. I agreed with your earlier statement but I seem to have misplaced my Glock.

#4 pleckner

pleckner

    Gold Member

  • ChE Plus Subscriber
  • 564 posts

Posted 09 December 2003 - 07:44 PM

I've got something very similar going into a 50% Caustic tank I'm designing right now.

My comments:

ASME (Paragraph UG-133(d)) and API, really only require protection against internal failure. This implies a tube rupture scenario. With the bayonet heat exchanger, there is no fire case. The exchanger tubes are not exposed to fire, just the tank is. The correct conrolling scenario is either by-pass control valve or control valve (regulator) failure. In addition, if the tubes are designed for the design pressure of the steam header, that isn't a credible scenario either. You're left with a thermal expansion case and this is pretty easy, a typical 3/4" x 1" relief valve will usually do; and no calculations are necessary.

Now, a relief valve on the tank would be a different story. In this case, fire may be a credible scenario and so would the tube failure.

One thing about double jeopardy, a very controversial subject indeed. For you to have double jeopardy, you need two unrelated scenarios occuring at the same exact time. The trap could have failed at 8:00 (closed? usually they just loose steam) and the fire could have started at 8:01. Although these are unrelated, they do not happen at the same time. I would not consider this to be double jeopardy.

#5 Guest_Ben Thayer_*

Guest_Ben Thayer_*
  • guestGuests

Posted 10 December 2003 - 01:46 PM

Phil,

I respect your comments based on other posts and I know that "double jeopardy" is an area of disagreement but....

I read API 521 not as things happening at the exact same moment but if the two conditions are occurring at the same time.

API 521 states ...."The simultaneous occurence of two or more conditions that could result in overpressure will not be postulated if the causes are unrelated."

I think that there is a difference between two simultaneous conditions and two simultaneous events. For the post regarding a trap failing closed and fire, I would consider this to be double jeopardy if there is some reasonable method of determining the trap is not working in a reasonable time frame. If the heater controls fail on the tank and the midnight readings for the tank temperature are observed, I would argue not to worry about a fire happening at the same time.

If the bad trap is going to be found out when they do the 5 year inspection program of the trap, this would not be a good assumption. I think that a reasonable time period for discovery of the first failure is warranted and that adding another "condition" during the very short duration of the first "condition" is not warranted and adding a fire at 8:01 after the first condition occurred at 8:00 is double jeopardy and should not be considered.

If the trap fails in March 2002 and the fire occurs in December of 2003, that is a valid scenario of two things wrong at the same time. But one minute apart or one second apart should count as double jeopardy in my opinion.

I realize that this is a topic where two reasonable people can disagree

#6 pleckner

pleckner

    Gold Member

  • ChE Plus Subscriber
  • 564 posts

Posted 10 December 2003 - 08:39 PM

Yep, we're going to disagree on this one.

For one thing, if there is any real chance for loss of life, I throw double jeopardy out the window. I have to be able to sleep at night. I'll ask myself, would I feel safe being there during start-up or operating the unit with this design? If no, just do the calcs. It usually doesn't take all that much to just do the calcs and see where it leads you. Afterall, who is getting hurt by taking the conserative route? Maybe someone will (get hurt) if we don't.




Similar Topics