Project Logs

Collapse

• Commutation with hall feedback

  kevarek • 02/23/2022 at 08:59 • 0 comments

  So I have learned that static analysis of step vs hall code will not result with usable table for commutation. Here is the table that is actually useful:
  

  
  Same table as above but sorted by step (sixstep algorithm):
  
  Example of waveforms during OK CW operation (this was data source to create tables above) as acquired synchronously with switching frequency (50 kHz, sampling time at 90 % of switching period):
  
  

• Sensor connector pinout

  kevarek • 02/13/2022 at 13:18 • 0 comments

  Pin 1: Supply voltage (both 3V3 and 5V tolerant).
  

  Pin 2: NTC thermistor output (with respect to GND). 10kOhm @ 25 °C; Beta unknown for now.

  Pin 3: Hall 1 output. Open drain. Seems to be related to Phase A.

  Pin 4: Hall 2 output. Open drain. Seems to be related to Phase B.
  

  Pin 5: Hall 3 output. Open drain. Seems to be related to Phase C.
  

  Pin 6: Common ground.

• Hall sensor outputs

  kevarek • 02/13/2022 at 13:10 • 0 comments

  Table with hall sensor output voltage relative to phase voltage (aka six step configuration - current step). Zero phase voltage indicates that phase is not driven by transistor half-bridge.
  

  
  

  Visual representation of table above.
  

  
  
  Hall sensors seem to be aligned in such a way that when any of hall outputs change state, commutation event shall occur (sometimes referred as 0° orientation with respect to phase zero crossing):
  
  Im not sure what is the benefit but apparently the other option might be 30° orientation (not relevant for this motor):
  
  Source: Texas Instruments - SLAA695.
  
  

View all 3 project logs