We are using NI PID controller for our temperature measurement and control DAS system (NI DAS). PID parameters are presently set to P = 5, I= 420 and D= 0. PID loop time is 3 seconds. There are two source of temperature in our system. One is through T type thermocouple directly connected to our DAS. It gives the accuracy of better than +/-0.5 Deg C temperature above -50 Deg C and scan interval is less than 1 sec. In this case with above settings PID control works fine. We control the temperature using power supplies (from LAMDA). There is another source temperature data which acquired at every 20 seconds with one count error. One count is 0.8 Deg C. In this case PID controller doesn't behave properly. What we observed is when Process variable is above the Set Point and consider total output is 0%. Now if temperature varies and still Process temperature is much higher than the Set Point PID controller gives significant amount of output (remember our D is zero). Please suggest how to tune PID controller in such cases.
So far I am not involved too much into PID controller. I hope i am able to explain my issue properly otherwise please let me for any other information.
In the world of control, your rates of 3 and 20 seconds are exremely slow. Are you running the PID loops at the same rate? Can you increase the frequency of reading the temperature? Can you post your code?
Your PID loop interval should be slower than the slowest process input. Modify your code so that the loop interval is much longer when using the slower temp data.
