Electrician Talk banner
1 - 20 of 21 Posts

·
Registered
Joined
·
1,692 Posts
Discussion Starter · #1 ·
2 water pumps have recently went crazy.. no loop changes nor vfd..
Example: Sp 65 psi pid maintains 55 doesn't kick on until PV drops below 55 even when the CV is at 100% ...literally shuts down shortly after stabilization at 55psi, or goes into overide... I have seen but not pinpointed details when it goes into overide mode.. (but if your not looking you miss the exclamation point showing overide mode.) That's the start I'll explain more later on..
 

·
Registered
Joined
·
1,692 Posts
Discussion Starter · #3 ·
Transmitters. I was thinking transmitter, and changes them out already. One thing that makes no sense is it not trying to zero the error % in the pid. Even if a pump is told to meet an unreachable pressure, motor should run balls to the wall to try and make 0% error. Am I not right? Note: both pumps are not linked logically in any way...which is why I say program.... can someone explain when and why a loop goes into overide mode?
 

·
Registered
Joined
·
438 Posts
Transmitters. I was thinking transmitter, and changes them out already. One thing that makes no sense is it not trying to zero the error % in the pid. Even if a pump is told to meet an unreachable pressure, motor should run balls to the wall to try and make 0% error. Am I not right? Note: both pumps are not linked logically in any way...which is why I say program.... can someone explain when and why a loop goes into overide mode?
You said before that the CV is going to 100%. If so then the loop is indeed trying to go balls to the wall to reach the setpoint. If the actual drive output does not agree with the Loop CV then that is an issue. When all is well the Loop CV is linked to the drive command/speed reference directly.
 

·
Registered
Joined
·
1,692 Posts
Discussion Starter · #5 ·
I see on the vfd reference speed 60 hz. But it is not running. It does not turn on until it drops more than 10 psi below SP. When it turns on, it tries to maintain 10 below SP. When its less than 10 psi from the SP it turns off. Really short On/Off cycle, which makes me worry about equipment damage.
 

·
Registered
Joined
·
1,692 Posts
Discussion Starter · #8 ·
The first pic was prior to our current issue, which was intermittent. The second picture was while it was acting normal..
 

·
Registered
Joined
·
1,692 Posts
Discussion Starter · #9 · (Edited)
I believe the drops in the yellow cv was when it went in overide mode for a split second before going back in program. Note i think the pics are of two separate pumps , but they both did the same thing, I just grabbed two random photos taken the same day.
 

·
Registered
Joined
·
438 Posts
A loop goes into override when the override input's value is lower than the CV. I don't know what drive or controller you're using but you should check to see if there's a value i. The override input parameter. Easy way to bypass it into put 100 into it.
 

·
Registered
Joined
·
438 Posts
A loop goes into override when the override input's value is lower than the CV. I don't know what drive or controller you're using but you should check to see if there's a value i. The override input parameter. Easy way to bypass it into put 100 into it.
 

·
Registered
Joined
·
1,692 Posts
Discussion Starter · #16 ·
Talked to Allen Bradley and they said if our programming company fails to fix it, they can send someone out free of charge to "assess"...
 

·
Registered
Joined
·
3,109 Posts
Transmitters. I was thinking transmitter, and changes them out already. One thing that makes no sense is it not trying to zero the error % in the pid. Even if a pump is told to meet an unreachable pressure, motor should run balls to the wall to try and make 0% error. Am I not right? Note: both pumps are not linked logically in any way...which is why I say program.... can someone explain when and why a loop goes into overide mode?
When measuring DP with two discrete pressure transmitters you're asking for all sorts of problems, i.e. induced noise, ground loops, bias errors, transmitter errors etc. You should really just use one DP transmitter.

If slamming more than one pump on at the same time to control the same variable you just make the problem worse. If you designate a "lead" pump and others that kick in if the 1st can't handle it, you would be much better off.

Also, check your PID parameters. If the proportional band "P" is too narrow you're running On-Off control and just not taking advantage of your VFD and might as well just have a DP pressure switch and a simple on-off contactor controlling the pump(s).

Too much Intergral "I" and you will drive your controller into "reset wind-up" resulting in output away from set-point, and take forever to recover.

If you use too much derivative "D" control will be very poor. I only use it for temperature applications.

As previously stated, if you can't control it manually (or measure it) all bets are off.
 

·
Electron Flow Consultant
Joined
·
3,309 Posts
When measuring DP with two discrete pressure transmitters you're asking for all sorts of problems, i.e. induced noise, ground loops, bias errors, transmitter errors etc. You should really just use one DP transmitter.

If slamming more than one pump on at the same time to control the same variable you just make the problem worse. If you designate a "lead" pump and others that kick in if the 1st can't handle it, you would be much better off.

Also, check your PID parameters. If the proportional band "P" is too narrow you're running On-Off control and just not taking advantage of your VFD and might as well just have a DP pressure switch and a simple on-off contactor controlling the pump(s).

Too much Intergral "I" and you will drive your controller into "reset wind-up" resulting in output away from set-point, and take forever to recover.

If you use too much derivative "D" control will be very poor. I only use it for temperature applications.

As previously stated, if you can't control it manually (or measure it) all bets are off.

For that application described, I wouldn't use any derivative.
 

·
Registered
Joined
·
1,692 Posts
Discussion Starter · #19 ·
No DP .....Two completely different systems. .not lead lag. more like potable and process water pumps. I know how to tune a loop. The process engineer doesn't so his butthole quivers anytime I touch the PID. Thought I screwed it up or something. Called the programmer now everything is default from start up and still jacked.
 

·
Registered
Joined
·
3,109 Posts
You probably know this but:

How do you shut the system down? And, start the system up?

If you just cut power off to the pump(s) but leave the controller powered up (and in auto) you will get Integral or Reset wind-up. Which will give you full (or no) output for a long time, depending on if it's a pump up or down system.

You must start up in manual get the system stable, then switch to auto.
 
1 - 20 of 21 Posts
Top