SLC500 PID bipolar method

I have read some article here about bipolar method for PID that I have to read it but I could not find any manual for this one. Could some one can help me or guide me ? Thank you very much. My machine can not be at setpoint.

Hello mathisa, and welcome to the forum.

I recently created a PID application that I would consider bipolar. It controls a heating and cooling unit with an analog heat signal and a digital (time proportional) cooling signal. I simply scaled the PID output from -100 to +100 and used the positive half for the heating gas valve command, and the negative half for the time proportioned cooling comand.

My application still needs some enhancement, but it does a pretty good job of controlling room temperature.

One thing I noticed is that when I first put the SLC in run mode, the PID starts out at a 0% cv which ends up being -100 (100% cooling).

I think I need to add a manual mode to start the thing off at a center point. My machine also needs a couple of wiring changes since there are situations in which the PLC is calling for heat but the burner controller can be delayed by other factors not monitored by the PLC.

Please post more details about your application. There are some really good PID experts that frequent this site who can definitely help you!

Paul

There is an AB Tech Note on this if you have access. Number 21677. Okie PC, they solve the 0% CV problem by moving 8191 to Word 17 of the PID on the First Scan. This is described as the Integral Accumulator. This assumes that Integral is used, otherwise they advise pre-loading the PID Bias. This is Word 6 I believe. The SLC500 PID will not accept a Min CV below 0, so you have to use Math or a Scale Instruction to control your Outputs.

Stu....

Thanks Stu. I am using a 5/02 which is different from the higher end SLCs, but I have a test rack I can play with to try out the method you describe.

Stu probably writes about these documents...

http://www.plctalk.net/qanda/showthread.php?p=345767#post345767


Lare

Bi-Polar PID (Caveat)

My $0.02 worth.

I think you will want to predetermine if your application is Bi-Polar "Balanced", prior to electing to use a SINGLE PID instruction to cause +CV and -CV outputs.

This could be an example:
Tank Level Actual is the Process Variable (Pv)
Tank Inlet Proportional Valve 0.00% ~ +100.00% Control Variable (Cv) Adds volume to the tank
Tank Outlet Proportional Valve 0.00% ~ -100.00% Control Variable (Cv) Removes volume from the tank

If the inlet valve and the outlet valve have the same response time, and same capacity to add or subtract volume at the same rate, given the same command value, then a single Bi-Polar PID should work.

BUT: In a typical Heat/Cool where heat is applied via one method (electric resistance, infra-red, or gas fired), and the cooling is a completely different system such as, (cooling water, cooling fans, or refrigeration, then I would classify this type of system as "un-balanced". The Heating slope and thermal lag, are going to be vastly different than the Cooling Slope and thermal lag, and I would not expect one single PID with one set of PID terms to be able to effectively control both heating and cooling.

I would endorse using two separate PID instructions tuned specifically for heating or cooling and enable the appropriate PID depending on which side of the fence that the Pv is on.

Happy New Year...

Thank you everyone for your kindly help.
I will try the way that you are recommended.