7 replies to this topic

[Art Montemayor]

[FONT=Times]A. Collins:

I've done what you describe on distillation column reboilers and never suffered anything resembling a "water hammer". I also never had a reason to believe that water hammer could be induced or be a product of varying the condensate level in the vertical shell-side of the reboiler. As I recall, I used a condensate pot besides the vertical reboiler and put my level controller on the pot, rather than directly on the reboiler shell. I also used a 1" vapor equalizing line between the top of the external pot and the top of the shell (right under the top tube sheet) to facilitate condensate drainage out of the shell and into the condensate pot.

This arrangement worked rather well for me and I would not expect any water hammer as a result of doing this. Water hammer is the action of high velocity steam (or vapor) pushing a slug of water or condensate down a pipe at a rather high flow rate. The hammer is a result of the high velocity liquid hitting a fitting or a change of direction within the piping arrangement. Sometimes this effect can be so severe that it can break a fitting or a welded component right off of the pipe. I do not believe that reducing the available condensing surface will cause this effect.

I hope this experience helps you out.

[Art Montemayor]

A. Collins:

The purpose of the pot was:

1. It avoids having to cut and weld appropriate nozzles on what may be an ASME coded vessel and, therefore, require an "R" stamp;
2. It also avoids having to put a weight stress on the side of the vertical shell;
3. The side pot allows you to completely drain the condensate if you so wish - it allows for a very wide condensate level range. This is very difficult to do if you attach the level controller directly to the shell.

The reboiler operation won't, in my opinion, be affected by the lack of a condensate pot; however, your range of operation might be limited and you may require additional reinforcement on the shell if you attach the level controller directly on it. The use of a condensate pot is one of practical necessity rather than process requirements. Selection and installation of equipment does not entail solely process needs. Process engineers have to consider the practical aspects of operating and maintenance requirements as well as mechanical strength: clean out access, overhead clearance, operating flexibility, safety requirements, quick and practical adaptability to design codes, etc.

I hope this answers your questions.

[joe lambert]

You can get water hammer in a shell and tube heat exchanger that is partially flooded with condensate. Thermal water hammer happens when small steam bubbles collapse quickly when meeting colder water. This collapse creates water hammer.
A reboiler is different than a shell and tube heat exchanger, and I have seen flooded heat exchangers with water hammer problems many, many times.
For more information on this look at:
http://www.kirsner.o...snerHammer.pdf
http://tis.eh.doe.go...l/bull0099.html

A control valve on the steam inlet should control the outlet temperature, even if there is very little steam into the heat exchanger and it runs under a vacuum condition. With proper air vent, vacuum breaker, and maybe a condensate pump and steam trap, this should not be a problem.

[morali]

[size=7]Thanks for this excellente references. I am working in a failure analysis of a Sulphur Cooler in a Degasser unit (Heavy Oil Upgrading). We found in the last inspection that this heat exchager had both tubesheet deformed (Concave shape to inside of the bundle). The bundle is one pass and it is an integrated bundle-sell heat exchanger, seal weld in both tubesheet.

The original design of this Hx was to have BFW in the shell side to cool down the liquid sulphur that go to loading area. However somebody decided to change form BFW to a heater, putting 50 Psig steam inside of the shell side with no furthe rchange in the hx (Operator drain the condensate one per shift). We suspect this cause a vacuum effect in the bundle and the suffer a severe damage (a lot of leak).

The problem here was change the operation without Management of Change, it happen before, now we have a better control of this modification, but your "chat" will help us a lot in our analysis.
I would like to post some photos but I do not know how do that.

Thanks,
Moraima