Viewing Error Terms in from PIDE block in RSLogix 5000

While troubleshooting control issues due to windup, it would be helpful to be able to view the error term due specifically to the error term in the PID control. The PIDE block saturates the control output to some max and min setting (user configurable), but I do not believe the integral portion is limited in the same way by default. I have looked into utilizing the Windup parameters in the block, but it would still be useful to view this integral error term. Is there a way to get this value in RSLogix 5000?

I do not believe the integral portion is limited in the same way by default.

Click to expand...

welcome to the forum ...

take a look at the following thread and see if you still feel the same way about the PIDE having an Integral action "windup" issue ...

http://www.plctalk.net/qanda/showthread.php?p=782214&postcount=5

as near as I've been able to tell - the Integral does NOT "windup" as many people think it does ... if you can come up with another test/experiment that would either prove - or disprove - that a windup situation could exist, just post the idea here and I'll see if I can find enough time to test it out for you ...

Ron Beaufort said:

welcome to the forum ...

take a look at the following thread and see if you still feel the same way about the PIDE having an Integral action "windup" issue ...

http://www.plctalk.net/qanda/showthread.php?p=782214&postcount=5

as near as I've been able to tell - the Integral does NOT "windup" as many people think it does ... if you can come up with another test/experiment that would either prove - or disprove - that a windup situation could exist, just post the idea here and I'll see if I can find enough time to test it out for you ...

Click to expand...

I am currently testing the windup issue on our level control system running on a primary and secondary set of pumps. I will report back what I learn.

JPLC7 said:

I am currently testing the windup issue on our level control system running on a primary and secondary set of pumps. I will report back what I learn.

Click to expand...

The original theory arose when a number of tanks overflowed on site. When I looked to the trends to see what was happening, I saw the level rise above the set point of the secondary control without a change in the CV. It appeared to be windup.

The tests I performed today do not agree with what I saw that evening. I did not see any delay between the resting secondary pump when I lowered its set point to that of the primary after leaving it saturated at 0 for about 2 hours.

It looks like I might have some other issue here then...

Ron Beaufort said:

welcome to the forum ...

if you can come up with another test/experiment that would either prove - or disprove - that a windup situation could exist, just post the idea here and I'll see if I can find enough time to test it out for you ...

Click to expand...

I only ran my test for about an hour and a half at the saturated CV point, but it can sometimes be days without seeing the need for the secondary pump to turn on. I noticed that in your test, the time period at saturation was also relatively short. I understand that if windup was occurring, that would have been plenty of time to see at least some delay, but perhaps the time period is not long enough. To test that, if you have the time, possibly run the same experiment, only while saturating the CV for a much longer time. I do not believe that that will result in anything different, but it is a possibility.

Thank you for the quick response and the help here!

I will gladly run the test that you suggested – but I am literally SWAMPED with work these days ... so - it might be quite awhile before I get a chance to experiment ...

here's a suggestion that might be helpful to you ...

set up a separate "test" PIDE and use "dummy" input and "dummy" output signals ... (it doesn't sound like you're trying to actually "tune" the PIDE – so you really don't need feedback for a closed-loop system for this type of experiment) ...

once you have the PIDE set up, then manually type in a reasonable PV signal – and watch what the CV does in response ...

if I understand what you're trying to accomplish (and I'm 99% sure that I do) – then you should be able to manually "saturate" this experimental PIDE for as long as you like - and see first-hand what the results would be ...