it seems to trip the desat and turn off the controller IC, right ? Any idea whether the controller IC was in drive 2 or tripping out of 2 into 1, which
then caused the desat ?
anyways, there are some things not totally optimal:
Code:
j) maximum shutdown error current, fixed: 54.0 A
k) maximum shutdown error current, proportional: 45.0 A
the fixed part under j is a bit high, I would lower it if possible. Reason is that it is used as a cheat when running in drive_2 under hall sensors.
The halls only give very crude phase info, 60 degree steps (360/6). When the motor starts moving the controller starts to see error current (as the
60 degree phase grid means the phase is never 100% perfectly the same as the motor). In drive 2, once the error current goes over the fixed limit
of option j, it goes 'f#ck them halls, I go sensorless' . It stays in drive 2, but it is cheating and not only running on hall info. Therefore, don't make
option j too high.
In drive 2 when the motor is standing still or running slow, it can be forced to rotate fast (like when you do a slide with a hub motor in your bike, and
you release the brake). The controller must detect this and go to drive 1 (to sync with the now all of a sudden fast spinning motor).
The detection for this (so basically, when it conks out of drive 2 back to 1) is in the recovery menu.
Code:
p) current to check: total current
q) fixed part: 36.0 A
r) proportional to throttle current, factor: 150 %
s) current filter 50% step response time: 5.0 msec
Since I don't know what is better you can choose between total current level or error current detection. The level with which it compares is based on a fixed
part and the throttle current. The measured current is filtered to remove spikes. Try playing around here, especially since your frequencies are so slow I would
increase option s to 10 or 20 msec. Maybe increase the fixed part, though especially the 150% with a pinned throttle will already give quite a high limit.
Something else that may help: you could reduce the rate at which the controller can increase the voltages (and thus the current). This is controlled in the
control loop coefficients,
Code:
amplitude control loop
h) 1st order: 60
i) 2nd order: 3.0000
j) 3rd order: 0.0000
change h and i by multiplying with the same factor (such that the ratio stays 20), for instance try 40 and 2 for h and i respectively...
P.S. how can the controller IC be responsible for tripping the output stage desat ? These are totally unrelated, no ?
