Cooling methods for closed loop ground source heat pump system 5

[kgood81]

I have just been handed a project involving ground source heat pumps, something I have no experience in, and could really use any advice any of you have to offer. Here are the facts:

Problem: High loop temperatures in a closed loop ground source heat pump system. The temperatures of the fluid (25% antifreeze, 75% water) are regularly reaching and maintaining at 100 deg F in the summer. On several occasions the temp. has been high enough to cause all heat pumps to shut down until fluid cools. The lowest the loop temp. has ever been is 58.5 deg F. The client is specifically looking for a way to reduce the loop temp. (cooling towers have been suggested).

Specifics: The system is located about 100 miles south of the canadian border in Montana, where winter temperatures regularly drop to -20. The system was installed in 1999, has 56 vertical wells (unknown depth at this point), 4 circulation pumps moving 280 gallons/minute (pumps are reported to be balanced to within 1/2% of each other), and 25 seperate heat pumps ranging from 3/4 to 12 ton capacity(Climate Master brand). System also has a dehumidifier installed on fresh air intake.

Question: Does anyone have any suggestions for cooling methods for the system and/or general advice on these types of systems? I have only worked at this company for a month and would really love to do good on my first engineering project. Any help would be greatly appreciated. Thank you in advance.

 

[MintJulep]

(pumps are reported to be balanced to within 1/2% of each other)

Reported by who? How long ago? Did they provide any measurements that you could verify now?

What are the pipes made from? What is the actual (not design) chemistry of the fluid in the closed loop?

What is the temperature of the ground water?

Did the system ever work?

 

[MikeHalloran]

Just off the top of my head, it sounds like most of the wells are plugged, clogged, airbound, or otherwise not doing their part ...

OR

that the system was balanced by closing and then 'cracking' a bunch of variable restrictors, so the pumps are running nearly deadheaded and the wellfield flow is nowhere near what you think it is and the pumps are just adding heat.

OR...

a thousand other things could be wrong.

With 56 thermal sinks, 25 heat sources, and 4 circulation pumps, all presumably parallelled, there's no way that system could _stay_ 'balanced', if it ever could _be_ balanced, absent a bunch of active balancing elements and control loops that you didn't mention, or a full time attendant whom you didn't mention and a bunch of manual balancing means that you didn't mention and associated measurement points from which you gave no data.

You want next week's Lotto numbers too?



Mike Halloran
Pembroke Pines, FL, USA

 

[kgood81]

Thank you both for taking the time to reply to me. The information I have so far comes from one telephone conversation with the building maintenance supervisor. I hope to have the enineering report/manual on the system to look at before too long.

The building supervisor did say that the system has worked, but that they have had several problems with it. He couldn't elaborate further on what he meant by the system being "balanced to within 1/2%". From your posts, I gather that someone may have been blowing smoke at him. I will probably repost when I have more information.

From your posts, I gather that there could be some major problems with the system, including but not limited to:

1) Blockages in the loop which would eliminate one or more wells

2) Excess pressure in the system due to restrictions in the line from valves at improper settings

Thank you both for your help. I will repost my question when I have more information on the system. If you have any other suggestions about the system or methods of cooling the loop, I would like to hear them so I can begin doing research now. Otherwise, I hope I will hear from you with my next post.

Thanks again.

 

[MintJulep]

3) Valves block with corrosion crud due to improper chemistry

or 999 other things.

Basic method of attack:

Find out how the system was supposed to work (design).

Find out the state of the system when it worked (hopefully there is a commissioning report).

Identify what changed.

Mike, if you could just send those lotto numbers to my e-mail...

 

[StoneCold]

While I am no heat pump expert I am familiar with complicated problems. The first thing to do is verify every "fact" that you are using to examine the problem. Temps, flows, loads, pipe sizes, number of sinks. Then you will have a true list of facts. Then simplify the problem. Maybe turn off all the heat pumps except the biggest one. Then what happens? Turn off all the pumps but one. What happened? Switch pumps. What happened?

This type of appoach should give you a gut feel for what is happening and hopefully steer you towards the answer you need.
Most of the time the answer is a simple thing. If you are thinking that it is a complex interaction of ten different variables you are probably wrong. There may be more than one problem but one of the issues is probably causing 99% of your problem. There are exceptions but I find them to be few and far between.

Goodluck

StoneCold

 

[MikeHalloran]

StoneCold is absolutely right; if the system ever worked, then it's probably just one simple thing wrong, or it was at one point.

First thing I'd do after trying to get actual facts out of the building super (good luck with that) is find every adjustment, and look for amateur tracks and clean spots.



Mike Halloran
Pembroke Pines, FL, USA

 

[kgood81]

Thank you guys for the advice on you have given so far. I am feeling both more panicked and more confident, oddly enough.

One thing though. It sounds like there is always the possibility that the system was just poorly designed and there are not enough wells to dissapate the heat the pumps are putting into the loops. If this is the case, what are the options for adding a cooling element to the system? The client suggested cooling towers, but I am not having much luck finding technical information about them on the internet (so far I have come up with Legionairres Disease), and I really have no mental image of what exactly a "cooling tower" is.

Doing some quick calculations, assuming the mixture is 75% water and 25% Ethlene Glycol and leaving pressure out of it entirely, I figure the cooling element needs to dissapate about 4,076 kj/min for every 1 deg Celcium drop in temperature of the loop.

Any suggestions of cooling tower websites to get a good feel for what they are and how they operate? Any other suggestions for coolling elements used with this type of system?

Mint Julep, Mike Halloran, and Stone Cold, thanks for taking the time to help me out, and double thanks if you have any help for the above question.

 

[MikeHalloran]

A cooling tower is a swamp cooler for a process. Greatly oversimplified, it's an updraft fan into which water is sprayed. The net effect is that some of the water is evaporated, and the part that isn't completely lost is cooled, and pumped back into the process.

At your latitude, it's a headache you don't need. It will freeze up and destroy itself every winter.




Mike Halloran
Pembroke Pines, FL, USA

 

[owg]

I like MikeHalloran's suggestion that the balance valves are pinched. I came across this problem once, one shift balanced using one valve, the other shift balanced using the other valve. The system gradually shut down. I wasn't smart enough to see it but someone else was.

HAZOP at

http://www.curryhydrocarbons.ca

 

[Ebels]

I'm not an engineer but I do install GSHP systems where I live. (Michigan)

The guys above have listed some good advice as far as researching/documenting what exactly is happening. Sounds to me like you are being asked to be a troubleshooter instead of an engineer. That's OK, trouble shooting is fun.

I think the first thing I would do is contact the folks at ClimateMaster (good product) and see if they have any documentation on this job. An installation that size usually involves some factory input at the design level. Maybe they have some design parameters documented that will save you a lot of leg work determining what the system should be doing in the first place. (Flow, Temps, delta Temps etc.)

Other than that it sounds like a fun puzzle to me.