PID Control: Basics

S

Thread Starter

Student

when an error is created between PV and SP, does all the modes P, I & D act simultaneously on the error or one by one.

Most texts explains the nature of PID output as the response being P initially (does it mean I and D are silent initially) and when the error does not diminsh (offset) 'I' acts (does it mean that P and D are silent now)?

But the single line diagrams explaining PID action, tell that error is acted upon simultaneously by all the three control modes. so why doesnt the 'I' action contribute initially at the time of error creation or why doesnt the P action contribute at the time of 'I' action.

Would a system be able to maintain setpoint with 'I' action alone? With 'P' action alone ofcourse there would be an offset.

Best Regards,
Student
 
To explain in brief how a 3 Term (P.I.D.) Controller functions under all variables would be near impossible the functions are:
P = Proportional
I = Integral
D = Derivative

There are numerous Web sites that you can review.

Today there are too many Engineers and Technicians that forget the basics of Process Control. Keep it as simple as possible maintaining plant stability.
All Electronic controllers will have the ability to perform all 3 functions but the type of process you are controlling will determine what's needed.

Look at the following Web site and some of the associated links

http://www.engineeringtoolbox.com/process-controllers-d_499.html
 
The three terms are simultaneous and the output is the sum of the P, I and D terms. It is the nature of the integration to occur over time, because it is a time function (it is integrating SP-PV over time). And the derivative term responds to a change in PV over time.

There are vast amounts of simulations, descriptions, etc. on the Web describing the PID behavior.

Bill Code
 
G

George Buckbee

P, I, and D all act together. In practice, however, many different forms of the PID equation are used. The specific form of the PID calculation depends on the vendor, controller type, and other factors.

Listed here are a few references that you may find useful:

A basic PID Tutorial:
http://www.expertune.com/r2.asp?f=AList&l=tutor.html

A comparison of the various algorithm types:
http://www.expertune.com/r2p.asp?f=AList&l=artCE87.html

-George

George Buckbee
Plant Performance Supervision with PlantTriage
http://www.expertune.com/PlantTriage.html
 
B

Bruce Durdle

Hi Student,

While it is true that, in the "standard" model of a PID controller, the error signal is fed to all three terms at the same time, the nature of the three terms means that the effects can be separated in time. The integral term will generate an output as soon as an error is fed into it. However this output term is very small initially and builds up with time. So immediately after the onset of an error the integral term has an output that is effectively zero. After some time, the integral contribution to the total output will have built up to a significant level. This adds to the output from the proportional block, and the two contributions will be equal (for an open-loop controller with a step change in error) after the Integral Action Time = 1/reset rate. Similarly, the D term is significant only while the output is changing and has no effect once the change in error has stopped. So all modes act simultaneously on the error, but the effect of each control mode differs depending on the time after the change.

Cheers,

Bruce
 
Top