cancel
Showing results for 
Search instead for 
Did you mean: 

EVSPIN32G4-DUAL + MCSDK 6.3.2: speed oscillations when only one motor is running

IrianJ
Associate II

[PN]: EVSPIN32G4-DUAL
[VERSION]: MCSDK 6.3.2
[TOOL]: Firmware

[DETAILS]:
Hello,

I am working on a dual-motor application using the EVSPIN32G4-DUAL board with MCSDK 6.3.2.

I observe the following behavior:

  • when both motors are running at the same time, the speed regulation is stable

  • when only one motor is running, strong speed oscillations appear

This behavior is observed when running either motor alone. In other words, the issue does not seem to be linked to one specific motor channel.

The motors are custom motors designed/manufactured by us, with the following configuration:

  • 5 pole pairs

  • Rs / Ls identified with the profiler

  • Hall phase shift = 121°

  • Rotor angle measured using Hall sensors

  • PWM frequency = 16 kHz

The attached speed plot illustrates the issue:

  • first phase: only one motor running → oscillations visible

  • middle phase: both motors running → stable behavior

  • last phase: only one motor running again → oscillations visible

The current regulator seems very stable, only the speed loop is going crazy when both motors are running. 

So the surprising point is that the system is more stable when both motors are enabled than when only one motor is active.

[EXPECTED BEHAVIOR]:
I expect the speed regulation to remain stable both in:

  • single-motor operation

  • dual-motor operation

The target is to have similar speed control performance regardless of whether one motor or both motors are running.

[HOW TO REPRODUCE]:

  1. Use EVSPIN32G4-DUAL with a project generated with MCSDK 6.3.2

  2. Configure both motors with Hall sensors

  3. Use the following motor parameters:

    • 5 pole pairs

    • Rs / Ls identified with profiler

    • Hall phase shift = 121°

    • PWM = 16 kHz

  4. Start only one motor and observe the measured speed

  5. Speed oscillations appear

  6. Then run both motors simultaneously

  7. The measured speed becomes stable

  8. Stop one motor again

  9. The oscillations return on the remaining active motor

Thank you for your help.

 

 

 

1 ACCEPTED SOLUTION

Accepted Solutions
GMA
ST Employee

Hello @IrianJ,

Using two Shinano motors and a Hall main speed sensor, the following speed behavior is observed:
image.png 
Starting or stopping a motor does not affect the speed of the other motor.

At low speed, the Hall sensor average speed FIFO depth parameter in the Speed Sensing Config tab is reduced from 16 to 6. This adjustment may also help in your case.

Is there a mechanical or electrical link between your both motors?

 

If you agree with the answer, please accept it by clicking on 'Accept as solution'.
Best regards.
GMA

View solution in original post

2 REPLIES 2
GMA
ST Employee

Hello @IrianJ,

Using two Shinano motors and a Hall main speed sensor, the following speed behavior is observed:
image.png 
Starting or stopping a motor does not affect the speed of the other motor.

At low speed, the Hall sensor average speed FIFO depth parameter in the Speed Sensing Config tab is reduced from 16 to 6. This adjustment may also help in your case.

Is there a mechanical or electrical link between your both motors?

 

If you agree with the answer, please accept it by clicking on 'Accept as solution'.
Best regards.
GMA
IrianJ
Associate II

Hello,

Thank you for your reply and for your suggestion.

On my side, I managed to get the motors working correctly in single-motor operation by drastically lowering the speed controller gains (Kp and Ki). After that change, the speed oscillations disappeared when only one motor was running.

This should not be an issue for our application, even with the reduced speed control bandwidth.

So this is an acceptable workaround for us, even if it does not really explain why this behavior was happening.

Regarding your question: there is no mechanical link between the two motors. The only link is electrical, because both motors share the same DC bus.

I will try to change the FIFO depth parameter but we can considered the topic closed as explained above. 

Thank you again for your support.