Steam trap drop in flow from reboiler fouling? 2

[mjpetrag]

Wanted to test my theory with the engineers here. I have a distillation column with a thermosyphon reboiler at the bottom. We have 250# steam going through the tubes of the reboiler and product on the shell. Recently, we think we may have fouled the product side in a matter of hours. As soon as this happened, the steam valve went wide open and the steam flow dropped off. On the discharge of the reboiler for the 250# condensate, there is a ball float steam trap.

What I think may have happened is the fouling is causing the steam to not transfer heat and condense, so that the level in the trap isnt rising as much with condensate. Therefore the float valve isn't opening as much as it did before, and we are possibly steam locking the system.

Does this make sense or have I missed something?

-Mike

 

[Latexman]

It could be, but fouling usually occurs over a period of time much longer than a few hours. Unless you have experienced this before, I'd look elsewhere first, like a defective trap. The floats usually fail in the closed position.

Good luck,
Latexman

 

[mjpetrag]

After a boilout of the column, the steam flow comes back. It's an operational upset that we speculate causes rapid fouling. The traps have been replaced and the problem lingered.

-Mike

 

[Latexman]

Sounds more like decomposition being that fast.

Good luck,
Latexman

 

[StoneCold]

Sounds like you should reduce your incoming steam temp/pressure if possible.

Regards
StoneCold

 

[ione]

Have you got the possibility of monitoring the differential pressure across the steam trap?

 

[mjpetrag]

Yes, we're going to put pressure and temp gauges upstream and downstream of the trap so we know where we're at. I will post updates as they happen.

-Mike

 

[ione]

If steam locking is a possibility, consider the use of a steam trap with a 'steam lock release' valve incorporated. It is basically an internal needle valve which allows to bleed away steam, which would otherwise prevent the thermosyphon reboiler from being properly drained.

Another thought. Rather than having an issue on the product (shell) side, you could have a problem on the steam side. If your reboiler has experienced flooding phenomena on the steam side during its operation, corrosion issues could have arisen thus jeopardizing the heat exchange performance.

 

[Latexman]

Can you tell what the product is? That may give an idea of whats going on chemically, if any. Knowing the general chemical family would help if you don't want to mention specifics.

Good luck,
Latexman

 

[mjpetrag]

It's the product of a organic peroxide + acid.

It's a once thru reboiler design. Weird thing we tried was raising the level of the column bottoms, which dropped the steam flow off. Then we lowered the level in the column bottoms and the steam picked up. It actually keeps picking up very slowly.

It's very counterintuitive

-Mike

 

[Latexman]

I don't think I've ever (32+ years experience) seen a thermosyphon reboiler with steam in the tubes and process on the shell. Also, I've always chosen to put the "dirty stream" in the tubes. Easier to clean that way. Of course, we are all products of our experience, a lot of which was influenced by our predecessors, and the companies I've been with practiced the reverse. Are the tubes running vertical or horizontal?

For thermosyphon reboilers (aka natural circulation calandrias) with steam on the shell and process in the tubes (vertical), liquid levels in the column are generally maintained at a point corresponding to the location of the top tubesheet of the calandria. Greater capacity can be obtained by lowering the liquid level, at least to a certain point. Beyond this point, the top portion of the calandria is starved for liquid so that heat transfer results primarily by heating the vapor. The vapor normally contains sufficient liquid so that the vapor is not superheated. However, poor heat transfer results because the heat transfer occurs in the vapor phase rather than as a result of boiling. The increase in capacity obtained at lower liquid levels is modest. The decrease in heat transfer at levels below the optimum point is dramatic and rapid. Lower liquid levels may require additional surge capacity in the column. Liquid circulation rates are reduced at lower liquid levels with increased fouling and higher maintenance costs the typical result, especially if the top portion of the calandria is not
adequately wetted.

The optimum liquid level depends upon the system. The optimum can be predicted with confidence, but the final point must be established by observation because it is not always
possible to accurately predict pressure drop in two-phase systems, the effects of product composition, and the piping arrangement for the system. For most systems, the optimum point
results when the column liquid level corresponds approximately with the mid-point between the two tubesheets of the calandria.

Looks like you are seeing this phenomenon in your equipment too.

Good luck,
Latexman

 

[mjpetrag]

Sorry i mis-typed. It has steam on the shell and product on the tubes. It is a vertical exchanger.

Maybe that is what is happening. We'll keep adjusting until we get it right or just boil it out and start over.

Thanks for the replies!


-Mike

 

[sheiko]

mjpetrag,
I believe the book "A working guide to process equipment" by N. And E. Lieberman discusses this kind of problem in chapter 11 "Steam and condensate systems".

"We don't believe things because they are true, things are true because we believe them."

 

[mjpetrag]

awesome reference sheiko. thank you!

-Mike