Low Delta T Syndrome

[gogisha]

What is the easiest fix for the Low Delta T syndrome in a typical primary/secondary chiller plant set up.

Reading in one of the ASHRAE Journals it was suggested to control chiller from the secondary supply water temperature rather than from the primary, if the supply water in the secondary is gettting higher than the maximum allowed temperature for 5 minutes then start another chiller. This might make sense since the DDC system will not see low delta T on the primary and start another chiller when the first chiller is not loaded 100% .

Opinions?

 

[Noway2]

When you say "low delta T" I assume by this you mean that you are getting poor (low) return temperatures. More often than not, this is a problem at the building level and can be caused by several factors such as stuck valves, dirty or broken air handler coils, etc and too often the default (and incorrect) response is a call for more chilled water. Having poor return temperatures will cause your production to run inefficiently resulting in running extra chillers, which sounds like your trying to avoid.

At our facility, we use two different control methods, depending on the situation (time of year and other factors). One method we use is based upon differential pressure, which is an indication of load. The second method, which sounds a lot like what your considering we call "load factor", which is in essence a calculation of percent of full load. Under this method, we won't start another chiller until both the current "load factor" is at 110% AND the supply water temperature is above a threshold.

 

[gogisha]

Controlling it with dP is not an option in our case since the engineer that designed one of the buildings on our campus didnt provide tertiary pumps and we are using secondary pumps to lift the water to the highest level of the building and we can't go below certain dP settings on the system which is to high. Adding tertiary pumps in this building would be a possible solution but we would like to try least expensive things first.
When you say calculating percent of full load I assume that you mean chiller load. Correct?

 

[btrueblood]

Gogisha,

Sounds like a perfect place for pressure-independent control valves.

 

[Noway2]

When you say calculating percent of full load I assume that you mean chiller load. Correct?

Yes, calculated chiller load. Note, that this doesn't necessarily mean the machines full load by the manufacturer's standards. For example, we've found that some of our constant speed machines run much more efficiently in terms of kw/ton when "maxed" at about 85% full load capacity. Hence, for control purposes in determining when to bring a second machine on line, we could calculate the 'load factor' relative to this reduce capacity value.

our campus didnt provide tertiary pumps and we are using secondary pumps to lift the water to the highest level of the building

I understand what you are saying. Also keep in mind that there is a difference between static (head) pressure and differential pressure, the former being controlled by your water makeup and the latter being a function of pump speed (flow). The ability to deliver water to the highest building will be a function of head, which is a function of both the pump head and the head (pressure) at the suction side which is influenced by the make up system.

 

[DRWeig]

Some really good relevant work on this subject by Gil Avery (Fellow, ASHRAE):

http://files.engineering.com/getfil...d5-46af56e56aea&file=Chilled_Water_by_Gil.pdf

http://files.engineering.com/getfil...1-4dae-9fee-b3af847903bf&file=ChillersMax.pdf

http://www.achrnews.com/articles/print/get-the-most-out-of-variable-speed-pumps

If you investigate btrueblood's suggestion, be cautious of the extra pump head needed to keep the pressure-independent valves in their control range. They can work well if designed in from the start (though I'm not a big fan of them, energy-wise). In retrofit of an existing system, you need to make sure that your existing pumps and piping can produce the head needed at the control valve location.

I think you'll find your answers in Gil's papers above.

Best to you,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies

 

[SAK9]

As a first step why not calibrate all temperature sensors and set them to design temperatures.Also have all coils cleaned.That will go a long way in reducing chilled water demand.

 

[DRWeig]

Good first steps, SAK9. If the system once worked, it should work again. Look for failed (or disconnected) chilled water valve actuators, cranked-open balancing valves, and other quick-fixes that may have been performed by staff.

Best to you,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies