Issue with Restarting Motor After Power Cycle in High Sensitivity Observer Mode (Sensorless)

[wow]

Hello,

I am currently using High Sensitivity Observer (sensorless) mode for speed sensing in my project.

When the motor is running and I turn OFF the power, the motor continues to rotate due to inertia.

If I turn the power ON again while the motor is still spinning and attempt to restart the motor, the control becomes unstable. Even if I completely stop the motor and then try to run it again, the control remains unstable.

It seems that when the MCU is powered on while the motor is still moving, there is an initialization issue affecting motor control.

Q:
1. Is there a way to handle this issue and ensure stable motor control after a power cycle?
2. Are there recommended methods to properly reinitialize the motor state when power is restored while the motor is still spinning?

Any guidance on how to avoid this issue would be greatly appreciated.

Thank you!

[Gael A]

Hello wow,

You are trying to do what we call an On-The-Fly start-up. Be aware that this feature is not available on HSO, and that it should be manually implemented. You can look at the STO-PLL implementation to get inspiration from, but be aware that if you use an STDRIVE101, the implementation will not be trivial.

Now, why does it behave like what that ?
Well, first you should now that after each power cycle, the FW performs an Offset Calibration, which must be done when no current is flowing inside the motor. This is obviously not your case since the motor is still spinning, so your offsets are wrong, and no control can be achieved this way. Moreover, both PolPulse and RsDC start-up methods will behave poorly when the motor is still spinning, so even with correct offsets your On-The-Fly start-up will be difficult. We recommend waiting for the motor to stop spinning before any start-up, or implementing the OTF yourself.

[wow]

Hello @Gael A ,

Thank you for your detailed explanation.

In our application, the motor fan continues to spin even after the power is turned OFF due to airflow. Given this situation, is there any possible workaround to handle the issue?

For example, would it be feasible to perform offset calibration only when the motor is completely stopped? This way, the offset calibration would only be done during installation and not during normal operation.

Additionally, are there any other recommended approaches to manage this scenario effectively? Any advice would be greatly appreciated.

Thank you!