Carry over liquid in vacuum pump

[wdd0422]

Gentlemen....We have a process which strips methanol from a reactor. During the reaction the system vacuum is brought down to about 4 torr and pulled through 3 condenser stages. The first condenser is cooled with regular city water, the second condenser is cooled with 40 degree f chiller water and the final condenser is cooled with -25 degree f chiller water. Finally it goes through a knockout pot, then a canister filter to the vacuum pump (Busch Cobra 0300B liquid cooled, screw type) This is a fairly new process but it has been running. We recently added the low temp chiller loop on the final stage in a attempt to lower the amount of methanol that was being vented to the atmosphere. The problem is since Thursday we have gone through two vacuum pumps because methanol got past the dropout pot and into the screw chamber on the vacuum pump and bent the screws. We were recently getting some liquid in the dropout which is something we had not seen before (LL limit switch in pot went off) but it was drained and did not go off before the pump failures. I was suspecting the final condenser thinking that since it was being cooled more it was putting more liquid in the system but am told this is not possible because of where it is in the system. There are some design problems such as the pump is on the bottom floor and the condensers are on the third floor but it has been running like this off and on for the last year. The only variables are the final condenser is now cooled to -25 f (Before it was on the 40 F loop) and it has been considerably colder lately. There is another pump on the way (a very expensive pump) and they are putting some horseshoe piping about 5' high right after the knockout pot to try to catch any liquid that gets by. I hope this makes some sense, sorry for the long post, any advice or comments would be appreciated.

 

[CH5OH]

if I understand system correctly, you have superheated methanol,which you cool down in 3stages and keep under vacuum condition with a screw type compressor?
I can't picture the installation but I give it a shot anyhow...
it seems more logical to drain condens of from the condensors rather than trying to catch the droplets in the outlet piping (risc of droplets getting sucked out the knock out drum before they collect at the bottom)
this assuming the methanol is on the shell side.

 

[Compositepro]

I don't understand how your system can work at all. It sounds like you are trying to condense methanol at 4 torr. If you condenser is cold enough it is possible but you must not only condense the methanol but maintain its cold temperature so it does not boil-off again. Any air flow through the vacuum system will strip the methanol though and result in poor or no recovery. I've seen many such systems where the design was very poor and unworkable. You'll have to provide a lot more detail to get any useful help.

 

[Compositepro]

I forgot to mention, for a system as you describe, the most likely place for liquid methanol to collect is in the vacuum pump or at the discharge of the pump. The methanol vapor will condense as it is being compressed. This, alone, may be what is damaging your vacuum pump.

 

[Chance17]

You did not mention any V-L seperation equipment.
At 4 torr, the gas velocity must be very high.
Sounds like an evaluation of V-L seperation is needed.