Cooling large volumes of gas

[NewfieEng]

Here is my situation. We are testing compressors in our shop that are destine for field use. To test we run the compressor through a closed loop with nitrogen as the compressing gas. As you can imagine the gas in loop heats up very quickly and we run into High temperatures on our compressor cylinders. The gas is cooled from discharge temp (250-300 degF) to 120-130 degF but that still creates high temps in our cylinders. We would like to run our tests for longer, but can't avoid the overheating problem.

My question is: where could I find a refrigerator type cooler that would handle high pressures (1500psig) and cool large volumes (10-20 MMSCF/D) of flowing nitrogen from about 130 degF to less than 80 degF? Who would supply these type of coolers? Any suggestions?

 

[MintJulep]

How many and how frequently do you test?

A chiller for that type of volume and pressure will probably be a good chunk of change.

Could you go open-loop or partially open and inject some liquid N2 into the stream? Capital cost would be way less.

If your testing is not continuous throught the day, you might want to look into thermal storage systems rather than a chiller with the capacity to keep up.

 

[insult2injury]

The cheapest solution to increase test length is probably to build a larger nitrogen volume into the system, depending upon the size and economics.

How is the gas currently cooled from 300 to 120?

I2I

 

[BigInch]

I suppose there is some reason why you can't use dried air.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[MortenA]

Have you considered using a turbo expander on the pressure let down side? This will increase the temperature drop compared to a valve.

Best regards

Morten

 

[NewfieEng]

To answer some questions. We test about twice a week for 4hrs. We would like to test for longer. We test compressors anywhere from 100HP units to 1300HP units. We currently use air cooler fan type cooler (like a giant automotive radiator). At best it will only cool our gas to about 20-30degF from ambient (100DegF).

I like the idea of liquid N2 or a turbo expander, where could I shop for a turbo expander? We have thought about increasing the volume of the test loop but I am concerned that the volume needed to reduce the heat substantially would be too large to be feasable.

 

[pmover]

since you are located tx(?), try testing at night or early morning when ambient temps are lower. at best for air cooler/exchanger, lowering gas temp to within 10°F of ambient is about the best that can be acheived without having a larger exchanger surface area. larger exchanger means more $. just curious, have you investigated the performance of existing air cooler? it may be fouled. i would first investigate the existing air cooler's performance. seems like it is either undersized or fouled.

is there a nearby cooling tower that you can tap into? need to negotiate usage cost with owner.

investigate evaporative type cooler (not too good in high humidity locations).

like MintJulep states, a refrigerant type will be cost prohibitive. so forget it.

meanwhile, good luck!
-pmover

 

[thermcool]

I was thinking you might get some decrease in the temperature of the compressed air by cooling the compressor cylinders using forced convection air flow.

This can decrease the temperaure and thus extent of the cooling required for the compressed outlet gas from the cylinder.

 

[MikeHalloran]

Try misting water on the air side of the radiator, at least at the hot end.




Mike Halloran
Pembroke Pines, FL, USA

 

[MortenA]

Theres this company:

http://www.hsturbo.de/produkte/turboexpander/index_en.html

and of-course:

http://www.mafi-trench.com/

Best regards

Morten

 

[unclesyd]

Comeback with your physical setup as is and compressor operating conditions, and desired test parameters. The additional information could open up some more possibilities.

 

[NewfieEng]

Some good suggestions here. The big problem is not the lack of cooling in the existing cooler, it is the continuous increase in the suction temps to the cylinders. For expample. We start out with about 80 degF on our suction and as the the compressor runs it slowly approaches 120 degF. which in turns raises the discharge temp. Our coolers are design to drop the discharge temperature from around 250-300 down to 120-130 and no further. If we could just drop the temps from 120 down to 80 or less we could, in theory, run continously.

I am considering a valve system that slowly introduces fresh cool N2 into the suction side and slowly releases the hot high pressure air off the discharge side when the compressor gets a larger enough diferential built up. So for example when my compressor got the suction down to about 50psig and the discharge up to about 350psig. slowly start letting fresh N2 into the suction and releasing hot N2 off the discharge. So eventually I could just be compressing all fresh N2 and releasing it to atmosphere, instead of recirculating it. Any thoughts from anyone on this idea?????