PDA

View Full Version : One for the chiller engineers!!!



acnerd
29-07-2009, 06:05 PM
I spend most of my life on Japanese split systems, but today I had a look at a Ciat chiller on a regular maintenance visit. It's a 70kW chiller with air cooled condenser and 2 recip compressors on one fridge circuit. No faults indicated. R22. Recips are 3 cylinder with one cylinder on each compressor having an unloader. I noticed bubbles in the sight glass, so took the following readings:

One compressor running:
Suction pressure 64 psi
Suction temp. 12 C
Liquid pressure 195 psi (on liquid line)
Liquid temp 35 C
Air on condenser 22 C
Air off condenser 33C
CHW flow temp 8 to 11C
CHW return 12 C

Both compressors running:
Suction pressure 60 psi
Suction temp. 7 C
Liquid pressure 255 psi
Liquid temp 46 C
Air on condenser 22 C
Air off condenser 40C
CHW flow temp 6 to 9C
CHW return 11 C

Because of very small demand on the system, thanks to rubbish UK summer, the chiller will only run for about 5 minutes before reaching setpoint.

Question: is it short of refrigerant (subcooling quite low) or am I not giving it enough time to "stabilize" or am I missing the plot completely?

PS: we have to assume flow rates through evaporator are ok. Condensor coils are clean, half of the 8 fans run, even when 2 compressors running. Let me know if you need more info. Thanks.

El Padre
29-07-2009, 06:29 PM
Hi acnerd,

Have you tried turning the chiller off for an hour or two to get a load, you can then get both compressors to run fully loaded, measure the compressor current before and after it loads up.

Cheers

Gary
29-07-2009, 06:57 PM
I'm thinking we are looking at two separate issues: Charge and load.

Charge: The system has sufficient refrigerant to maintain superheat at the TXV bulb... so we could say that it has enough to currently do the job. But it needs some surplus (subcooling) in order to handle heavier heat loads... so you should probably add a little more refrigerant.

Load: Reaching setpoint in 5 minutes isn't necessarily a problem, but if the compressor is starting more than 6 times an hour, this is not good for the compressor.

Given the light load, I'm wondering why the second compressor started... or did you force it to start in order to obtain readings for us?

Peter_1
29-07-2009, 07:26 PM
Don't forget your measuring fault (resistance of the copper) when you do a contact measurement.

acnerd
29-07-2009, 07:42 PM
I'm thinking we are looking at two separate issues: Charge and load.

Load: Reaching setpoint in 5 minutes isn't necessarily a problem, but if the compressor is starting more than 6 times an hour, this is not good for the compressor.

Given the light load, I'm wondering why the second compressor started... or did you force it to start in order to obtain readings for us?

Thanks for the reply Gary, I thought the superheat values seemed close enough. I did force the setpoint down lower to fire up the second compressor to see if the extra discharge load would clear the bubbles in the sight-glass, which it didn't.

As far as the frequency of starts, I seem to recall that the compressors will start alternatively, and it does have a short cycle protection of no more than 6 starts per hour.

acnerd
29-07-2009, 07:42 PM
Hi acnerd,

Have you tried turning the chiller off for an hour or two to get a load, you can then get both compressors to run fully loaded, measure the compressor current before and after it loads up.

Cheers

Good idea El Padre, didn't think of that!!

acnerd
29-07-2009, 07:45 PM
Don't forget your measuring fault (resistance of the copper) when you do a contact measurement.

This point is new to me Peter; how much of a loss is there through the copper? The digital thermometer I use measures real-time.