Help with Motor Parameter Variation in ST MC Profiler

[Mani2]

Hi ST Community,

I'm working with a sensorless BLDC motor and using the ST Motor Control Profiler to extract the required motor parameters, as I don't have access to the exact datasheet values. I'm facing some issues and would appreciate guidance.

What I did:

• Entered motor information including Max Current, Max RPM, and Number of Poles.

• When I entered the actual Max Current, the profiler often gave Speed Feedback Faults or failed profiling.

• To overcome this, I reduced the current in steps from 93% down to 80% of Max Current, and profiling succeeded with reduced current values.

Observations:

• The profiler gave significantly varying results, even though the motor and setup remained the same.

• Not reaching max RPM Specified.

• Some examples of variation:

  • BEMF Constant: ~±3 Vrms/kRPM

  • Friction: 700 to 1450 µN·m·s

  • Inertia: 20,000 to 45,000 µN·m·s²

This creates confusion — which profile values should I consider as trustworthy for tuning my motor with Motor Control Workbench?

Questions:

1. Which current value range should be ideal for profiling if the max current causes faults?

2. How do I decide which BEMF/Friction/Inertia values to trust if they vary this much?

3. Is there a recommended way or best practices to improve consistency in the profiler results?

4. Should I average the values from multiple profiles, or take the one with the least profiling current, or the one closest to the actual motor current?

Any advice or insights would be really helpful. Thank you!

Best Regards

Manikanta

[Mani2]

Hi everyone,
Still looking for guidance on this.

To help clarify and move the discussion forward, here’s some more context and what I’m trying to understand:

• Which motor profile result should I trust among the multiple ones generated with different current values (70–95% of actual max current)?

• I am mainly looking for reliable values of:

  • BEMF constant

  • Friction

  • Inertia

• These parameters are important for tuning PI gains in the Motor Control Workbench (especially Speed Loop PI), so I'd like to get as close to accurate as possible.

• If values like Inertia and Friction vary significantly, should I:

  • Take the average?

  • Choose the one with smooth profiling behavior?

  • Ignore friction/inertia and tune gains experimentally?

Additional Details:

• I don’t have actual datasheet parameters for this motor.

• Profiling at 100% rated current always gives faults (speed feedback or startup failure).

• So I reduced the current in steps to avoid faults and got successful profiles.

If anyone has experience using Motor Profiler successfully without datasheet values, I’d really appreciate your input on how to tune the motor starting from profiler results.

Thanks in advance for any help!

[Mani2]

Any from recommendations from ST?