Compact Logix PIDE Calcs

Hi All,

It seems that everywhere I look in the AB literature the calc for the PID in a compactlogix controller for Proportional Band is a simple Kp X E. But in practice it seems like it should be 100/Kp X E.
Does anyone out there know for sure what the proper calc is for the Compact logix PIDE?

Try document Logix-wp008 from here:
http://literature.rockwellautomatio...passedParameter=logix-wp008&submitQueryForm=»

I recently converted a bunch of single-loop controllers to MicroLogix and simply represented the gain as "Band" on the HMI, since that's what the maintenance guys were accustomed to.

But the A-B equations have always used "Kp" as a very ordinary Proportional Gain value, so if you're going to think or document in "Band" terms, convert the value (Kp= 100/Band, right?) before placing it into the PID or PIDE equation.

The difference between Gain and Band, or Rate and Period, or Dependent and Independent are just part of the history of approaches to PID control equations and there really isn't a "right" way.

after carefully rereading your post, I've got a hunch that you're confusing the term "Proportional BAND" with the term "Proportional ACTION" ...

I mention this mainly because Allen-Bradley seldom (if ever) uses the term "Proportional Band" ...

if this doesn't help clear things up for you, it would be a good idea to tell us WHERE in the Allen-Bradley documentation you're running across the term "Proportional Band" ...

we'll be glad to help you sort things out ...

I hope that this helps ...

Sorry Ron, Should have been 'gain' not 'band'
The calcs in the white paper (thanks WIDELTO) I am familiar with and I am using the Independent Gains form. It just does not seem to work like the calcs show. I guess I should ask if anyone knows for sure if this white paper describes the calcs for both Logix500 and Logix5000 versions of the PIDE?

That white paper is strictly for the ControlLogix PIDE instruction in the Function Block language.

It does not address the ControlLogix Relay Ladder Logic PID instruction, nor does it address how the PID controller works inside the PLC-5, SLC-500, or MicroLogix. All of those use a Position form of the PID equation instead of a Velocity form, but as far as I know they, too, use Gain as ordinary "CV = Kp x Error".

You mentioned CompactLogix and PIDE in your original post. What is the actual controller and instruction you are using ?

Remember also that none of the Emulators support PID or PIDE; they don't attempt to keep accurate Time, so they can't perform Time-based instructions.

... if this white paper describes the calcs for both Logix500 and Logix5000 versions of the PIDE?

Click to expand...

there is no PIDE for Logix500 ... only for RSLogix5000 - and only in Function Block Diagrams ...

why don't you post your ACD file? ... we'll be glad to take a look ... (you'll have to zip it first - forum rule) ...

suggestion: do a File / Save As operation and save the file in the L5K format ... this will make it a lot smaller - and much easier for the forum members to work with ...

Hi Ken, Thats exactly what I get out of the AB docs, Kp X E, Yes it is a compactlogix L32E using PIDE instruction in function block code. It seems as though the 'P' input in the instruction may not be using the basic calc of Kp X E but it could also be me and my hard (and small) head trying to figure out why the smaller the 'P' input to the instruction seems to make a higher CV change to the given error. It is a very slow acting loop so I may have been seeing things too. I tend to get 'glazed' over after a while. It almost seems like the calc is 100/Kp X E for the 'P' value in the instruction but if it ain't it ain't and I will get back to it. Thanks again