Results 1 to 11 of 11
  1. #1
    Join Date
    Sep 2006
    Posts
    60
    Post Likes

    Johnson PCT Programming Logic

    Folks,
    Wanted to pick your brains on how you program with the Johnson PCT/CCT and control multiple DO points . Im referring to when you have a 3 or 4 pump system ( Or any digital points for that matter ) and want to assign the pump numbers to the lead thru lag3 designations either thru a runtime logic block , manually assign a random sequence and force them to rotate or auto rotate based on a faulted device . It sounds from my guys that there is a problem doing it this way... Meaning you cannot assign DO Pump_1 to a software point named LEAD ….

    has anyone been able to control based on this logic ? or anything close to it ?

    It seems like it was way too easy with the old Carrier Best ++ Logic where you could assign an Array for lead lag of multiple chillers or pumps etc......and literally watch the lead lag designation swap before your eyes during rotation execution...

    Any thoughts would be much appreciated..

  2. #2
    Join Date
    Feb 2009
    Location
    Northeast
    Posts
    521
    Post Likes
    I want to see other peoples answers.

    I write my own logic most of the time because the canned JCI sequencing ends up with all the pumps off at some point and the only way to get them to restart is reset from the workstation. (Or place them in hand at the starter/drive. End up with a totally manual system.) Around here on most parallel pump systems we need them running to prevent freezing in the winter. If my pump logic allows them to be all off I have a problem.

  3. #3
    Join Date
    Oct 2009
    Posts
    876
    Post Likes
    Are these pumps on frequency drives? Is there a differential pressure sensor or are these constant speed? The canned Q and A Johnson programs have a "stage up" setpoint to stage on a 2nd pump if 1 pumps can't satisfy the differential pressure set point. The canned logic also rotates pumps based on runtime. If the controller is connected to a network or if you have a PCA controller, you can set a schedule which will rotate the pumps at a given time.

  4. #4
    Join Date
    Jun 2007
    Posts
    119
    Post Likes
    From memory...
    Lead lag would be chosen by program.

    If say user made pump3 lead and pump2 lag and pump1 lastLag... then what would happen if pump3 was failed... and only one pump required.... it would just keep trying to run pump3.
    By letting the program choose lead/lags (it based it on assignment of lead/lag user requests) then if it sees pump3 failed, it would roll to lag aka pump2 etc. believe if all failed it would roll again and jeep trying.

  5. #5
    Join Date
    Sep 2006
    Posts
    60
    Post Likes
    Thread Starter

    Pump Lead Lag

    Quote Originally Posted by energyStarBoy View Post
    From memory...
    Lead lag would be chosen by program.

    If say user made pump3 lead and pump2 lag and pump1 lastLag... then what would happen if pump3 was failed... and only one pump required.... it would just keep trying to run pump3.
    By letting the program choose lead/lags (it based it on assignment of lead/lag user requests) then if it sees pump3 failed, it would roll to lag aka pump2 etc. believe if all failed it would roll again and jeep trying.

    i understand there is a logic activity that functions and swap by what is called Rank
    but thats not really lead lag.

    but what im looking for is the ability to have a graphic with a four block cell named lead_Pump thru Lag3_Pump and have the ability to manually change order in 1324 or whatever order and execute that change or either by runtime or fault execution visually see the order change

  6. #6
    Join Date
    Jun 2007
    Posts
    119
    Post Likes
    From memory,
    I believe it was in fact based on runtime not user request (as other stated).
    But i feel like there was an option to apply it differently... or perhaps i changed the logic to apply based on user requests (or i am mistaken).

    There wont be an array (that i know of) but the ‘ranks’ could be pushed to variables and presented on a graphic.

    Its certainly possible as other manufacturers do it with user requests and rotate on request; in graphical programming.

    Would need analog (or enumarated) variable for each pump to hold it’s rank position. (A good program would check that no two pumps have same rank, and adjust if they do)(random ranks would be a mess to program)(rotating all ranks +1 and roll last to first, would be possible)
    Any of that would take considerable ‘programming’ time, but is do-able. I would start by copying another manufacturers logic.




    As for the pumps turning off at some point, i remember that being an issue.
    We wired one pump off a normally closed relay (pumps were for IT room) so that even if the controller lost power or rebooted we would at least have that pump running.

    I also remember manipulating state tables to keep pumps always on.

  7. #7
    Join Date
    Jul 2002
    Posts
    1,278
    Post Likes
    Quote Originally Posted by billcontrols View Post
    the canned JCI sequencing ends up with all the pumps off at some point and the only way to get them to restart is reset from the workstation.
    OMG I just had this happen! The owner has no front end so he would have been screwed. Why does this happen?

  8. #8
    Join Date
    Feb 2009
    Location
    Northeast
    Posts
    521
    Post Likes
    "Why does this happen?"

    2 Pumps; P1 Lead and P2 Standby. P1 running and status fails for some reason. ( Could be turned off by operator at starter/drive for service) P2 will start and run, P1 will be stopped by the program as long as it is in alarm. Now if P2 fails, P1 does not start because it is in alarm, P2 is stopped because it is in alarm. The only way to clear the alarm is toggle the Pump Alarm Reset point on the BMS.

    Usual result is the owner puts the pump in hand and the automatic system becomes useless.

    Pretty poor real world pump sequencer out of the box.

    I usually let the CCT/PCT program write the pump sequencer through the Q&A. Then delete the Alarm activities and put in my own Alarm activity. Then I make a Lead pump network or hardware point and use the state table to command the mode of the output controller for the pumps based on OA shutdown, lead pump, and alarm status. I usually leave a pump in alarm "running" since most of the systems I work on will not be damaged if both pumps are running, but if neither pump is running on the wrong day the whole building can freeze.

  9. #9
    Join Date
    Sep 2007
    Location
    Australia
    Posts
    216
    Post Likes
    How about having a virtual point (LLMS) and assigning it an integer based on the day of week (if day of week = mon - tues then LLMS = 1, etc (say theres 3 total). But I'd increment the LLMS point directly, (if it's midnight + 1 minute then add one to LLMS, with a 1 minute sample/timer lockout).
    Rather than worrying about clearing the fault, move the whole cycle on (as the that pump wasn't on duty) then worry about fixing it.
    If LLMS > 3 then set it to 1. (LLMS 1 = Pump 1 as lead, 2 = Pump 2 lead etc)
    If LLMS = 1 & Pump 1 is in fault/FTR then increment LLMS by 1, if it's 2 & Pump 2 is in fault/FTR then increment by one etc.
    If all 3 are in fault then I often trigger a 1/2 kilogram block of Semtex that I install free-of-charge in all jobs
    The DDC system... guilty until proven innocent

  10. #10
    Join Date
    Dec 2007
    Location
    Connecticut
    Posts
    1,162
    Post Likes

    Post

    Quote Originally Posted by ccnman View Post
    Folks,
    Wanted to pick your brains on how you program with the Johnson PCT/CCT and control multiple DO points . Im referring to when you have a 3 or 4 pump system ( Or any digital points for that matter ) and want to assign the pump numbers to the lead thru lag3 designations either thru a runtime logic block , manually assign a random sequence and force them to rotate or auto rotate based on a faulted device . It sounds from my guys that there is a problem doing it this way... Meaning you cannot assign DO Pump_1 to a software point named LEAD ….

    has anyone been able to control based on this logic ? or anything close to it ?

    It seems like it was way too easy with the old Carrier Best ++ Logic where you could assign an Array for lead lag of multiple chillers or pumps etc......and literally watch the lead lag designation swap before your eyes during rotation execution...

    Any thoughts would be much appreciated..
    Put your email in your profile please. Or just shoot me an email.

  11. #11
    Join Date
    Sep 2007
    Location
    Kenilworth NJ
    Posts
    1,906
    Post Likes
    Quote Originally Posted by Knife Switch View Post
    If all 3 are in fault then I often trigger a 1/2 kilogram block of Semtex that I install free-of-charge in all jobs
    I totally need to add that to my installs. What a time saver!
    Hmmmm....smells like numbatwo to me.

  12. Likes heatingman liked this post.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
Contracting Business
HPAC Engineering
EC&M
CONTRACTOR