Refrigerant level control

[DJCherre]

Folks,
We have several shell & tube kettle-type heat exchangers where propylene refrigerant is in the shell, process in the tubes. The propylene is flashed to either 70 psig (35 degF) or 6 psig (-40 degF), depending on the exchanger. The level indication/control of the refrigerant is lousy, so they all run in manual. What we currently have is dP level indication. In one instance both impulse lines are insulated & steam traced and in another they're not insulated & are hapazardly steam traced. We recently replaced dP with guided wave Radar & insulated the stilling well. It's better, but still not under control and still spikes up & down a lot. So, I wonder if anyone has any suggestions for refrigerant level control in south Louisiana.
Thanks,
Dean

 

[ilan7]

Dean
We are using similar liquid (Pentane) in close loop system where it boiled in one heatexchanger and condensed in another one. We need to keep the level at the boiler which is tube and shell heatexchanger by controlling the pump supplying the liquid. The best solution I found is the DP transmitter. Due to the nature of the liquid and the low pressure measured, it must be equipped with two diaphragm seals each with 2” or (Preferable 3”) 150 or 300# flanges. Best performances we got from Rosemount, Smar, Yokagawa, DP with integral Diaphragm seals.
ilan

 

[ilan7]

Dean
We are using similar liquid (Pentane) in close loop system where it boiled in one heatexchanger and condensed in another one. We need to keep the level at the boiler which is tube and shell heatexchanger by controlling the pump supplying the liquid. The best solution I found is the DP transmitter. Due to the nature of the liquid and the low pressure measured, it must be equipped with two diaphragm seals each with 2” or (Preferable 3”) 150 or 300# flanges. Best performances we got from Rosemount, Smar, Yokagawa, DP with integral Diaphragm seals.
We have one unit not far from Morgan City LA.
Regards
ilan

 

[DJCherre]

By diaphragm seals are you referring to what we call "pressure pads" or "remote seals"? Where the pressure-sensing diaphragm is on the end of a tube-like gizmo that you slide into the flange, right up to the process. That sounds like a great idea, since the big problem we're having is the flashing/condensing of propylene refrigerant in the impulse lines. The seals would eliminate the impulse lines.
Thanks for the response!
Dean

 

[ilan7]

Dean
As you said, I mean the remote seals. as you have low pressure, see that the minimum diaphragm size is 2" or 3".

See

https://www.smar.com/products/sr301.asp

for more information.

Regards
ilan

 

[hacksaw]

you should have filled impulse lines in this service, not filled with liq. refigerant, but with non-freezing liquid

the remote seals work too as long as you mak provision for valved isolation from the process for maintenance.