STSPIN32G4: No errors, but no PWM output

[ClyneModa]

[PN]: STSPIN32G4
[VERSION]: v6.4.1
[TOOL]: MC Workbench / Motor Pilot
[DETAILS]:

I am working with a custom PCB design which was largely follows the EVSPIN32G4 schematic. I defined a custom board in MC Workbench and generated firmware, and can upload the firmware to the STSPIN without issue. Motor Pilot detects the board, and I can command START/STOP and read values/status.

When I press START for either Torque or Speed control, the board enters some ALIGNMENT stage (only the first time) followed by RUN. No behavior is observed with the motor. No faults are ever reported by Motor Pilot. I can STOP and START motor control again and still see no change in behavior.

With a scope, after START I see a single pulse on the U phase's gate for around 22 microseconds, then it falls to 0V and remains there. I did briefly run the firmware under GDB, and I see that MOE is set during RUN.

Using a 48V frameless motor, running from a Li-Ion pack (bus voltage currently 56V), SIR680LDP-T1-RE3 FETs. VCC measures a steady 8V, and the motor's phases idle at 5.45V. Have tested sensorless and encoder feedback; Pilot does read RPM from encoder when I manually spin the rotor. After initial START, current measurements appear to calibrate and rest right around 0A during RUN and IDLE.

Can anyone suggest what settings or portions of my circuit I should be looking at to resolve this issue? The lack of documentation on what some of the Workbench settings actually do has made this testing process cumbersome; this is beginning to impact project schedule.

MC Workbench file is attached. I'm happy to share any other design details as needed.

[EXPECTED BEHAVIOR]: PWM activity on the gate drive pins GLSx and GHSx. Any observable behavior in the motor.

[HOW TO REPRODUCE]: Design board following EVSPIN32G4 schematic, generate basic firmware, try to START the motor in Motor Pilot.

[GMA]

Hello @ClyneModa,

PE15 nFAULT pin is not described in the project (driver protection).
Refer to the STSPIN32G4 datasheet for the nFAULT description and to the EVSPIN32G4 board description for the driver protection feature definition.
You can check if nFAULT is triggered when you start the motor.

[ClyneModa]

Hi @GMA,

Thank you for the suggestion. I connected my debugger to the board and watched the status of GPIOE_IDR during firmware execution. nFAULT is low and READY is high during RUN state, so there was a fault being triggered.

nFAULT indicates thermal shutdown, VCC UVLO, or VDS monitoring by default. Thermals have not been observed to be a problem, and VCC UVLO is ruled out since READY is high, so VDS monitoring was the issue here.

For now, I have disabled VDS monitoring by shorting SCREF to 3.3V. This led to Motor Pilot detecting Over Current faults when I START the motor, which I fixed by reducing my ADC sampling time to ~500ns. The motor does not spin yet, but there are no faults and I hear the motor produce a loud tone; my issue now must be with alignment or something beyond the scope of this post.

 

I would suggest that ST make the status of these flags (nFAULT, READY, WAKE) available in Motor Pilot. It's frustrating that the list of detectable faults in the program is incomplete. I also see there is a STATUS register in the STSPIN which could indicate the above faults, but the generated firmware does not appear to access it.

Any form of guide from ST on how to bring up a motor with their MCSDK would be immensely helpful as well. Many of the settings in MC Workbench are unclear on their meaning, and the program's documentation is extremely lacking in detail.

[GMA]

Hello @ClyneModa,

Driver protection error is reported in the MC Pilot interface if feature is enabled:

An associated comment is added in the next MCSDK delivery user manual to describe the error source.