unstable flow loop

dear all,

i have a flow control loop with a DP transmitter and a valve with an actuator type samson 5824, and I use a simple progme with pid controller (FB41)and  CPU 313c .in a manual mode his works fine and in automatic loop become unstable .you can see attached file  my programe and a manual for my valve.

Hello kam;

I am afraid we do not see your program in your attachment, only the manual for the 5824 valve positioner.

One initial question, while we wait for your program: how do you plan on connecting a LON actuator to a S7-300 PLC?

Hope this helps,

Daniel Chartier

Hello Kam;

You can have a look at this , one of many FAQs and discussions in the forum on FB41:

What should you watch out for when calling and assigning parameters to the controller functions in function blocks (S)FB41, (S)FB42 and (S)FB43?

https://support.industry.siemens.com/cs/ww/en/view/18459342

Siemens also proposes tutorials for Step 7 and TIA Portal that can help out beginners or programmers who are going about a new task.

Have a look here for an Advanced Programming tutorial for S7-300 (with Step 7 v5.x) covering PID controls:

http://w3.siemens.com/mcms/sce/en/advanced_training/training_material/classic-modules/Pages/default.aspx?tabcardname=advanced%20programming

Look at Module B3 : Control Engineering with Step 7; includes an examination of FB41 with example

.The tutorial documentation is available in English and French.

Hope this helps,

Daniel Chartier

Hello kam;

If you control loop output is jumping from 0 to 100% continuously, it generally means that you have tuned the parameters of the PID control very badly.

For example, you stated earlier that you use a Gain value of 20. Do you understand what that means, what effect this has on the control output?

The main calculation of a closed-loop system is the error (of the system), the difference between the setpoint requested and the actual measure of the process variable. If you have say a temperature control, and you request a setpoint of 31C, and your actual room temperature is 30C, then your error is 1C (31-30=1).

The PID equation starts with an immediate reaction (no delay) from the P (gain) parameter. If your proportional gain is set at 20, then the control output will start off at P(gain)*error, or 20*1C=20C. You will be asking the control output (a heater) to do everything it can to bring the output temperature 20 C higher than it is now (50C), so the control output will go to 100%. Then the room temperature will become higher than the setpoint very quickly, the error becomes negative, and the control output becomes -20 C, and the room temperature must become 20 C lower very quickly, so the heater closes down (goes to 0%), and the cycle continues... Sound familiar?  As Hamid Hosseini told you earlier, your P gain is much too high...

You have to start by understanding how the different parameters of a PID loop influence the control output before you start selecting the various parameters of the loop control, that is a basic requirement.  Then you need to read up on various methods of tuning (selecting adequate gain parameters for your loop) the control loop.

Have a look at these documents for example (there are many, many more on the web); the first shows a method of tuning the look manually, through trial and error; the second uses various mathematical methods, based on a step test of the loop (setting the loop in manual, jumping the setpoint value by a known percentage, say 10%, and examining the change in the measured value that follows the step change. You need something to capture the change, such as an HMI trend or a recorder, in order to use these methods accurately.

http://innovativecontrols.com/blog/basics-tuning-pid-loops

http://www.chem.mtu.edu/~tbco/cm416/tuning_methods.pdf

Hope this helps,

Daniel Chartier