A good thing to check for before you start breaking your head: is PID really needed?? Maybe a P will suffice, or maybe a variable rate integrator... but for smaller applications, PID can be overkill

The best thing for you to do would be to sit with the motor and try to control it manually. The response and your experience will give you a pretty good idea what sort of control is needed. If that is not enough, the next step would be to try various modes (P, then I, then D, an maye PI) individually and check the effect of each.

In my project, i undertook some healthy experimentation and came up with a set of values that I put into a lookup table. I found that a bunch of if-else conditions based on the lookup table managed to control a weird and inconsistent process very well, whereas conventional algorithms would have failed.

In short you will need a lot of trial and error and intuition on your side. The trick is to program your experience with the mechanism into the microcontroller.

kundi