Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations waross on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

Variable flow chilled water system

Status
Not open for further replies.

visitor

New member
Mar 28, 2002
117
0
0
US

Hi friends,
I am currently designing a campus cooling system of about 8000 tons.I intend to use distribution pumps of following capacity:

a)2nos varaiable speed drive pumps each of 3350 gpm@ 200 ft head(eqvt to 2x2000tons based on a delta T of 14.4F)

b) 1 no constant spped pump of 6700 gpm@ 200 ft head(eqvt to 4000tons based on a delta T of 14.4F)

How should I control the speed of Variable speed pumps when both variable speed drive and constant speed drive pump are running? I see that in such an event both pumps will be developing different heads.How will it affect overall ssyetm operation?

Any advice will be appreciated.
 
Replies continue below

Recommended for you

I am an engineer for a large building complex.
For my chillers i have to maintain a designed flow of water.
If you want to use varible speed pumps you should go through a heat exchanger.Then you can use return water temp to control water flow.
hope this helps.
 
I am of the opinion that all pumps will be running at the same head since the run parallel against the same headers. Its just a matter of balancing to a new opearting point (common for all pumps). Increasing the speed of the variable speed pumps will raise the said balancing point.
 
It's been my experience that when a constant speed pump is running the variable speed pumps have to operate near full speed to match the system pressure created by the constant speed pump. We have run chilled water systems designed with more than one operating pump by operating the pmps in unison. For example, the lead pump comes on line with it's VFD and when it comes to full speed, the second pump is energized with the two pumps running in unison to match the load. I haven't used three pumps, but assume that would be the same. When the first two pumps come to full speed the third pump is energized and all three pumps operate in unison.
 
I agree with demar. I believe pumps should be the same capacity if they are designed to operate in parallel. If the 6,700 gpm pump is running and a D/P sensor somewhere in the loop decides more pumping power is needed, ramping up one of the smaller pumps might cause it to operate against a shutoff head for some time before matching the outlet pressure of the bigger pump. The time spent with the smaller pumps ramping up enough to a point of overcoming the head enough to crack its discharge stop check is time spent wasting electricity.

For parallel pumping to a common loop, your pumps should all be of equal capacity. Control sequence should be as follows - Start and run one (say P1) to control loop D/P. When P1 reaches >95% for 5 minutes (give or take), reduce P1 to minimum speed, start P2. Let P1 and P2 ramp together. When P1 and P2 together reach >95% speed for 5 minutes, ramp down P1 and P2, start P3. Then P1, P2, P3 ramp in parallel. Unload the pumps by reversing the sequence when pump speeds drop to a given setpoint, say <45% for 5 minutes.

If the pumps are supplying different loops, different sized pumps are fine, but parallel ops (I believe) mandate equal sizing.
 
You can get away with parallel pumps of different flow capacities, but their heads must match as closely as possible. You can't prevent parallel pumps from running at the same head, since they take from a common suction header and discharge to a common discharge header. The surest way to match heads of multiple pumps is to use identical pumps, if they are to run in parallel.

Another arrangement, which I like, is to have unequal pumps, sequenced according to load, running one at a time. This allows the smaller pump to be selected with a lower head to match the system requirements at lower loads. Saves big bucks on pumps and on motor energy also.
 
Chasbean1 is correct. Now a days controllers are available in which data viz, shutoff head of pump, duty point, efficiency and maximum flow rate are fed. These are very useful when multiple pumps are operated for one purpose. A differential pressure transmitter installed at the last usage point between supply and return headers will take care that sufficient water is reaching the last system as well as optimising energy consumption. ITT Bell&Gosset has such a type of system. These controllers ensure that pumps are always running at optimum efficiency. But automatic back pressure creating equipment like 2 way proportionate valves need to be used along with this system.
 
Status
Not open for further replies.
Back
Top