Results 1 to 9 of 9
  1. #1
    Join Date
    Jan 2002
    Location
    Fort Worth\Dallas, Texas
    Posts
    1,731
    Post Likes

    Cooling Tower Optimization

    Im curious how many people are doing tower optimal control, and for those that are what strategies are you using?
    Go Rangers!

  2. #2
    Join Date
    Aug 2006
    Posts
    123
    Post Likes
    We do this alot. We control tower fan speed to maintain a calculated CWFT setpoint. CWFT setpoint is calculated based upon outside air wetbulb + an approach temperature of something like four degrees. Need to set limits on the setpoint based upon chiller information, a common low end is 65 degrees.

  3. #3
    Join Date
    Jan 2002
    Location
    Fort Worth\Dallas, Texas
    Posts
    1,731
    Post Likes
    Thanks for the info Travis.
    Go Rangers!

  4. #4
    Join Date
    Mar 2005
    Location
    Pacific Time Zone
    Posts
    4,292
    Post Likes
    Doing a similar thing - only when you have everything with Econo-coils you can run the loop a lot cooler.

    BTW - Is CWFT - Condenser Water Fan Temperature Setpoint?
    "How it can be considered "Open" is beyond me. Calling it "voyeur-ed" would be more accurate." pka LeroyMac, SkyIsBlue, fka Freddy-B, Mongo, IndyBlue
    BIG Government = More Dependents
    "Any 'standard' would be great if it didn't get bastardised by corporate self interest." MatrixTransform
    My 5 yr old son "Dad, Siri is not very smart when there's no internet."


  5. #5
    Join Date
    Apr 2007
    Location
    San Diego, CA
    Posts
    1,389
    Post Likes
    Quote Originally Posted by crab master View Post
    Doing a similar thing - only when you have everything with Econo-coils you can run the loop a lot cooler.

    BTW - Is CWFT - Condenser Water Fan Temperature Setpoint?
    Maybe Condenser Water From Tower? as opposed to CWTT

  6. #6
    Join Date
    Sep 2004
    Location
    SoCal
    Posts
    464
    Post Likes
    We like to control tower fan speed from head pressure inputs to a PI control loop. We have a building that includes a 300 ton evap condenser with 2 separate DX refrigeration circuits. Pressure transducers installed on the liquid line are wired to a couple Circon GPC controllers which send the values to an iLON which selects the highest signal and transmits back to the Circon controller. The GPC controller uses a PI function and sends the speed signal to a Yaskawa VFD that has a LON card installed. It works really well and holds head pressure on mild days within +/-5 PSI of setpoint.

    Recently we added a flow meter on the makeup water line and added the makeup water flow gpm to the network. We log the data every 15 minutes and were amazed when there were a few days this summer where the tower consumed close to 5,000 gallons of water a day! We can also detect if the float valve leaks during shut-down hours. Next phase will be to add TDS and PH sensors.
    Cheers,
    lb
    A hundred million nodes - it's a LON story.

  7. #7
    Join Date
    Oct 2009
    Posts
    4
    Post Likes

    Optimization?

    Quote Originally Posted by tshort View Post
    We do this alot. We control tower fan speed to maintain a calculated CWFT setpoint. CWFT setpoint is calculated based upon outside air wetbulb + an approach temperature of something like four degrees. Need to set limits on the setpoint based upon chiller information, a common low end is 65 degrees.
    That's not really optimization though, is it? More of a strategy, but it doesn't really optimize the system.

    Trane has a strategy, and I could tell you what it is, but then I'd have to kill you.

  8. #8
    Join Date
    Oct 2009
    Posts
    4
    Post Likes
    Quote Originally Posted by lonboy View Post
    We like to control tower fan speed from head pressure inputs to a PI control loop. We have a building that includes a 300 ton evap condenser with 2 separate DX refrigeration circuits. Pressure transducers installed on the liquid line are wired to a couple Circon GPC controllers which send the values to an iLON which selects the highest signal and transmits back to the Circon controller. The GPC controller uses a PI function and sends the speed signal to a Yaskawa VFD that has a LON card installed. It works really well and holds head pressure on mild days within +/-5 PSI of setpoint.

    Recently we added a flow meter on the makeup water line and added the makeup water flow gpm to the network. We log the data every 15 minutes and were amazed when there were a few days this summer where the tower consumed close to 5,000 gallons of water a day! We can also detect if the float valve leaks during shut-down hours. Next phase will be to add TDS and PH sensors.
    Cheers,
    lb
    Quick question - is this being done on a chilled water plant with centrifugal compressors? I think that you would end up with a lot of surging.

  9. #9
    Join Date
    Sep 2004
    Location
    SoCal
    Posts
    464
    Post Likes
    Quote Originally Posted by SubCoolWasTaken View Post
    Quick question - is this being done on a chilled water plant with centrifugal compressors? I think that you would end up with a lot of surging.
    No, this is a DX plant with (2) Carrier 5H120's. On another job we did similar head pressure control of tower fans with a Carrier Screw 30GT chiller running R134A equipped with LON translator card. Discharge pressure from the chiller was sent onto the network and a freely programmable LON device did the rest.

    At the end of the day, it's the head pressure that counts! Not too high and not too low!

    Cheers,
    lb
    A hundred million nodes - it's a LON story.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  

Related Forums

Plumbing Talks | Contractor MagazineThe place where Electrical professionals meet.
Quality Home Comfort Awards