lowering the condensing temperature reduces chiller's capacity?

[capaula]

I have Trane 2000 tons centrifugal duplex chillers, and I am having problems getting them to reach full capacity.
I am resetting the cooling towers temperature. Could this be contributing to the capacity shortage?
I am short on 300 tons. The chillers were designed for 85F condense water, and right now they get around 78F.

 

[mauricestoker]

I've encountered the same problem, ended up discovering the seondary was controlled by temperature which the coils demnaded. We were pumping at 42* F, loop control was set for 44* F.

As PretendFarmer mentions, is demand being satisfied?

 

[capaula]

This is part process cooling and most HVAC.
I run 39F in summer because it dehumidifies a lot.
I need a dew point of 44F in the end of the dehumidification coils.
We are talking about 850,000 of make up air, (100% OSA units).
The dehumidification coils were designed for 40F. In the rest of the year I reset the SP, it can go up to 42.5F.
Tertiary pumps are in part of this system, 6,500,000 CFM on recirculation units, doing 2F delta T (air side) but most of the load is on the secondary system. There are NO 2 way control valves at the individual buildings to hold back the return water until the it warms to 55-56F.
There are no 3 way valves, each coil has its control valve (proportional Baumann butterflies).
I have no real balancing valves on the condenser water and chiller water pipe before the chillers. We managed to balance it with 18" butterfly valves before each chiller, which was a nightmare.
There are actually hundreds of coils on this system. I have to check the coil arrangements.
In fact I am dying to change this to variable primary flow, but have no way to justify this financially yet...

 

[capaula]

Now I am suspicious that the temperature sensors in the chillers are drifting.
I will check those guys one by one.

 

[pretendfarmer]

capaula,

If you can believe your flow meters and temperature sensors, you are bypassing almost 4100 GPM of nice cold chilled water you just made and are dumping it into the return line, warming it up.
In order to maintain the setpoint you need, you need to chill that mixed chilled water return temperature of 39ºF and 56.3ºF flow down to 39ºF. What is the mixed chilled water return temperature at the point #1 on the attached PDF?

 

[KiwiMace]

So, just to confirm, the chillers were originally specified at 39degF CHW supply temp to get the 2000tons?

I think the rule was 3% capacity lost per degF reduced sp? This works out to 300 tons, which is what you say you are missing.

 

[ChasBean1]

pretendfarmer,

The 4,100 gpm being “wasted” is also cooling the return which means less chiller work. The temperature at Point 1 should be 51.7°F (I called it 52° earlier for errors’ sake). The data is all there – 3,100 gpm primary pumps x 5 running is 15,500 gpm; 4,100 gpm in the decoupler means 11,400 gpm in the secondary.

4 secondary pumps at 68% speed delivering 11,400 gpm implies 2,850 gpm per pump and about a 4,200 gpm capacity per pump.

Some of this isn't directly metered. To me the data begged the question of flow measurement accuracy. Temperature measurement is usually pretty good…

 

[pretendfarmer]

capaula,
To me, running additional chiller(s) to accommodate the additional 4100 GPM being bypassed qualifies as wasteful and inefficient. You not only have to to run the additional chiller(s), but you have to run the additional chiller pump(s), additional condenser water pump(s), and probably additional tower fan(s).
Just because you have pumps and chillers rated by nametag information at 3100 GPM does not mean that is where the pumps are actually operating on their curve. That is why I believe knowing what the actual versus calculated temperature at point 1 is important. Trust but verify.
I also believe eliminating uncontrolled bypass flow will allow your chillers to fully load as they are designed to do. That way you will be able to satisfy your chilled water requirements with fewer chillers. The calculated payback of meeting your load requirements with fewer chillers will quickly convince you to change over to a vari-prime type piping system. All you need to do is to install a valve in the bypass, shut it, and let the system run as is currently designed.
You mentioned that you have some tertiary pumps piped in your system at the coils (hopefully not as much as 4100 GPM). These tertiary pumped loads should be checked to make sure their flow rates will provide minimum chiller flow protection.
This simple system modification will allow you to operate in both vari-prime and primary secondary and then you can convince yourself (and others) which mode works best for you.

 

[marcoh]

Have you had the numbers reviewed by Trane to see what the chillers should be doing at the flow and temperatures you are experiencing, rather than specified design conditions?

Why do you want to change to a variable primary system? Primary/secondary, especially for a large system, should be easier to control. If you install variable speed drives on the primary pumps you should be able to realise the same energy savings.