I'm looking for reference material, etc. that describes the procedure used for ADC offset calibration in the Universal Motor Control Lab.
Why is it done in this way?
At a high level:
- PWM for all three phases is set to 50%
- The ADC offset value is initialized to a hardware-specific value, which seems to be near mid-scale.
- For 32000 cycles, take ADC readings
- Ignore the first 2000 readings
- For the remaining 30000 cycles, the offset value is set to
0.998001 * current_offset value + 0.001999f * the read ADC value
I understand what is done, mostly, and I am looking more for "why" it is done this way. Why 30000 iterations? Where does the adjustment equation come from?
Thanks!