|
EDIT: OK something weird is going on. Let me think for a bit.
Old Message:
OK, before setting throttle to a usable range, first let's run the pi test while the throttle is zero (well, somewhere in the "dead zone"). With maybe 48v hooked up, type:
run-pi-test
Alternatively, if you know the resistance from phase to phase, we could figure it out analytically.
If it finds good values for Kp and Ki, next run:
run-pi-test2
That will let us look at just how good the convergence is, to let us know if we need to use slightly different values.
After we have good kp and ki (still at 48v or so), type:
run-rotor-test
Hopefully, that will start out slow, gradually spin the motor faster, and gradually slow down. It might take like 5 minutes to run. Once it's done, you should have a good value for the rotor time constant. Do all of this stuff while the throttle is "zero" (GREATER THAN min-regen-position and LESS THAN min-throttle-position).
|