Question about a P and I control

bigrthanur

Member
Join Date
Mar 2007
Location
Ohio
Posts
85
I am having a brain block about using control with only gain and reset If i was to set up my controller to work optimal. I would think i would want the controller to look at the error as much as possible. But i am trying to think back to basic process and forgot?

ANyway my question is i set my reset high(to look at the error more often) Wouldnt my controller lets say heat control in a water tank, Work better or more smoothly?

ANy info would help

Thanks
 
Check the units.
Reset is normally expressed as minutes per repeat.
So entering larger numbers makes the controller check the error less often.
 
yes that is correct but my question is if i check the controller more often will it smooth out more easily or will the temp fluxuate more?
 
Minutes per repeat? Yuk.

That is old terminology. Integrator gains are normally experessed as time constants.

bigrthanur, making the integrator time constant shorter doesn't make the controller 'look' at the error more often. This is done by making the controller update the PID loop more often. The sample time is divided by the time constant and mulitplied by the gain to find the true integrator gain. This means the integrator gain gets smaller as the same time is smaller. This way less gets added to the integrator when it is update more often keeping the over response to the error pretty much the same as long as the sample time is 10 times shorter than the minimum of the plant time constant and closed loop time constant.
 
Old termanology, or i think you are misunderstanding what i am asking. (or i am lost somewhere and not getting the basics?)

"I" standing for Integral or Reset i thought to mean how often the product controller will look at the error.
Error i understand it as the difference between the set point and the actual value
 
Last edited:
bigrthanur"I" standing for Integral or Reset i thought to mean how often the product controller will look at the error.[/QUOTE said:
No, how often the PID 'looks' at the error is dependent on how often the PID routine is called. The PID is called when a timer times out or from an periodic interrupt. You must select the sample period to be fast enough for your application. The integrator time constant is one of the tuning parameters.

Now y'all know why I don't like the term resets or repeats. It is misleading. What is repeating or reseting? Think in terms of time constants. A PID with a time constant of 1 second will take 5 time constants or 5 seconds to wind up to within 1% of the final value. That is simple and gives one a gut feel for how to adjust the integrator time constant.
 

Similar Topics

Hello all. I stepped into an established system to integrate a Fanuc Robot to tend machines that are using CLX's. I have seen many a machine use...
Replies
11
Views
1,073
Hi, I'm looking to see if there is a solution for an application I'm attempting. I have a PF525 running a pump. It is primarily hard-wired...
Replies
4
Views
1,505
Been working with PLCs for a couple of decades, but almost 100% DirectLogic. Have a customer who wanted me to make a couple of simple changes to a...
Replies
3
Views
1,114
Are you able to use the drop cable (node connector portion) of a t tap to connect to one side of another t tap to make a three way cable...
Replies
2
Views
1,520
New to the EcoStruxure programming software (I'm an AutomationDirect guy) and I've been banging my head against the wall trying to figure out how...
Replies
2
Views
1,014
Back
Top Bottom