PID in Micrologix 1200

A

Thread Starter

Arunprakash

Hi guys,

I am using micrologix 1200 for flow control. But it gets fluctuating, just give me a suggestion to tuning the parameters.
 
G

George Buckbee

Tuning may be only part of the answer here. PID algorithms in PLC's are almost always improperly implemented. Be sure to consider the
following:

1. Is the PID block execution time fixed or variable?

2. Is the actual execution time linked to the (configured) block execution time?

3. Are the inputs and outputs properly scaled?

4. Are the configuration bits for the PID block properly set?

5. Has someone implement other "helpful" logic that makes changes to the purpose of the PID? (Limit blocks, multipliers, etc.?) This happens all too often.

At a previous ExperTune User Conference, Rob Zaun of Rockwell gave an excellent presentation on this topic. You can download the slides from our web site at:

http://www.expertune.com/r2p.asp?f=AList&d=Feb11&l=articles/UG2007/PIDControlsPLCEnviron.pdf

After you have confirmed the configuration, then you can look more closely at the tuning.

If you'd like more training on the topic, please take a look at this course:

http://www.expertune.com/r2.asp?f=AList&d=Feb11&l=trainingEX-501-5E.htm


-George
 
B

Bob Peterson

IME, loop tuning is over rated as far as getting things running in a reasonable way.

Most loops in most plants do not have to run real well to be adequate and spending a lot of time on them to try and perfect the tuning is a waste of time.

I agree a lot of PLC and DCS loops are very poorly implemented. Sometimes I wonder how the people that put them in manage to get them to work at all.

Sometimes the "helpful" logic added to the PID structure is a consequence of the way things are physically and cannot be avoided, although often times the way it is handled is not ideal.

And a lot of people seem to think PID loops can solve any problem, when the control problem at hand is not something that is going to respond well to a standard PID loop approach.

A huge part of the problem with PID loops implemented in PLCs and DCs's is that the instructions provided with them have been full of errors and misstatements, and bad advice on how to implement them. They also tend to be full of features poorly documented but quite useful if used in the right way. I have to give AB some credit here. Over time they have dramatically improved what was originally very bad documentation on the PID loops implemented in their PLCs.
 
Top