SLC PID STI What If

I have a project that I just took over. There's and SLC 5/05 running several heaters through SSR's in a vacuum application. In the current prog. the ouput to the ssr is controlled by a PID instruction. PV is a thermocouple input. CV is a SSR 25A. Anyway the PID is set for STI and a loop update time of .12 (which I believe is 120mS). My average scan time is 5mS. There is no STI file and, if I'm understanding this correctly, the PID is being processed each time it is scanned. Will this create an issue? If so what is it? Should I place it in a STI file as suggested?

If there is no STI file and the PID is configured for STI then you are correct, the PID executes the loop every time it is scanned. However, the PIDs could be in a conditional subroutine which is called at a periodic rate, in which case, even though its not expresely an STI subroutine, the configuration would be the right one to use.

As for whether you should correct it or not? That depends. How good is the control right now? You will have to retune the loops is your correct it. 120ms is a pretty fast loop for temperature control - If you correct it and retune, you might want to consider a longer loop time.

Originally posted by IndiaJones:

Will this create an issue? If so what is it?

Click to expand...

Having the scan time and the PID loop update time different will change the meaning of your time based tuning constant, Ti (Reset) and Td (Rate). This probably isn't a huge deal if you tune the loops by trial and error. It is a much bigger issue if you tune using test data and tuning correlations.

As Alaric said, if iot works acceptably now I wouldn't necessarily worry about it. Running a PID faster than you need to just wastes processor capacity. It doesn't really cause a problem. However, if you ever do plan to tune using tuning correlations keep in mind that the numbers you come up with are only valud if the PID instruction is called cyclically at the specified rate.

Keith