This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

DRV8353R: 1X Mode Motor Operation and Noise- Reg

Part Number: DRV8353R
Other Parts Discussed in Thread: DRV8353

Dear Team,

We are testing our motor controller based on DRV8353RH in 1x mode. Motor has 27 Poles and hall sensor placed 120 degrees apart.

We could observe noise and oscillations while the motor is running. Also, the noise is more in while rotating in one of the directions when compared to the other direction. 

We have tried with 25kHz and 40 kHz PWM with 20%. Board consumes around 1 Amp @ 48V, Motor RPM is around 210.

When we increase the dulty cycle to 40%, the current consumption is increased to 2.4 Amp and the noise is high

Please support us on to resolve this for the smooth motor operation.

With Best Regards,

B.Saravana Vikram.

  • Hi Saravana,

    It sounds like your hall sensors are not in the proper place for your motor to commutate correctly. You should look at the voltage waveform of the phases for the forward and backward direction and compare the two. I have seen this happen once before, I fixed it by feeding the hall sensors into a MCU and then delaying or advancing the hall sensor signals based on the direction and how fast the motor was spinning. 

    Regards,

    Michael

  • Dear Michael,

    Thanks for your valuable reply.

    Initially, we thought of the same. The motor we are using, has marking of the Hall Sensors.

    Still, in order to double check, we have tried to rotate the motor with all 6 possible Hall Sensor (A,B,C) combinations.

    Motor has not run for few combinations and got struck in few other combinations. Could there any other reasons for this issue?

    With Best Regards,

    B.Saravana Vikram.

  • Hi Saravana,

    Can you try to get a scope capture of the motor rotating in both the forward direction and the reverse direction?

    I don't think your halls are out of order I think they might be just a little bit off(a few degrees) causing the motor to commutate more efficiently in one direction than the other direction. 

    Regards,

    Michael

  • Dear Michael,

    Please find the snaps of the Captured Phase Waveforms in both the directions.

    With Best Regards,

    B.Saravana Vikram.

  • Hi Saravana,

    What phase is this? can you capture a whole cycle of the motor commutation? which is in the forward direction? which is in the reverse?

    Regards,

    Michael

  • Dear Michael,

    This is the Phase-A. waveform First waveform with Peak Voltage of 61.2 is reverse direction and the other is forward direction.

    Please find the waveform of Phase - A as requested. 

    Reverse:

    Forward: 

  • Hi Saravana,

    Why is the phase of A  going so negative? it seems like it is going to -10-15V there should be a diode preventing that. it looks like your device is correctly commutating the reverse direction but not in the forward direction. Can you share the schematic? I want to make sure the connections to the DRV8353 are correct.

    Here is an example of what a properly commutated phase waveform will look like. look at the yellow line:

    You can see that the voltage is going a little bit negative but only about .7V. You can see in the High Z period you can see the BEMF of the motor. 

    Regards,

    Michael W.

  • Dear Michael,

    Please find the attached Schematics. Mostly we have followed the reference Schematic provided by TI.

    We have the provision to use 2 MOSFETs in Parallel in the H-Bridge section. Currently one MOSFET out of the two  is mounted. 

    Regarding the diode, would you propose to use diodes in each of the phase to Ground.

    AIO-MC-BLDC-D-100V-50A-1.rarAIO-MC-BLDC-D-100V-50A-1.pdf

  • Hi,

    your Mosfets should be good enough diodes in the reverse direction. I am just trying to figure out why your phase goes so far negative. It might be because the GND of the phases is bouncing. can you probe the gnd by the fets and see if it is bouncing?

    Regards,

    Michael

  • Dear Michael,

    Sorry,it is our bad. The negative voltage on the Phase is due to Ground connection to the oscilloscope probe.

    Please find the attached Waveform in one of the Directions. Still the issue persists. Please provide us your feedback.

    With Best Regards,

    B.Saravana Vikram.

  • Dear Michael,

    Further to the above, please find the captured image of Hall Sensor - A waveform with back emf captured between A and B phases.

    It looks like, the waveforms are not in Synchronization. Please provide your views and let us know how we can overcome

    this.

    With Best Regards,

    B.Saravana Vikram.

  • Oh, that makes sense! Hmm it looks like your Phases A is not properly going to High Z in the region highlighted below or the other phases is not properly PWM their phases. Can you capture all three phases in the same scope plot?

  • Dear Michael,

    Thanks for your extended Support.

    We could not capture the waveforms today as requested. We shall share it by tomorrow.

    Meanwhile, how do we address this misalignment? Whether it is possible to adjust the commutation

    by routing the Hall Sensors to our MCU and control thru Firmware or DRV8353 can handle this?

    With Best Regards,

    B.Saravana Vikram.

  • Hi,

    Sorry, I missed the message about the BEMF and Hall sensor alignment. 

    The waveform of your "BEMF" and hall sensor do not look correct. How are you measuring them? You need to remove the motor from the driver and them groung one phase to the probe GND and then measure one of the other phases like the diagram below.  Also capture the Hall sensor waveform.

    The offset needs to have the MCU and firmware compensate for it if there is an offset.

    Regards,

    Michael

  • Dear Michael,

    Sorry for the delayed reply. We have got another Motor to test with our Driver. We have measured the Back emf and hall

    sensor Signals. It looks fine. Still the commutation is not happening properly. Please find the individual phase waveforms.

    We could not capture all the phase waveforms simultaneously. Please help us to narrow down to the root cause of the issue. Phase_Waveforms.rar

    With Best Regards,

    B.Saravana Vikram.

  • Hi,

    It looks like your hall sensors are alligned with your BEMF correctly but just to make sure you can follow the following FAQ to make sure,

    Can you capture all three phase voltages on your scope to show if the phases are properly going high Z?

    Regards,

    Michael W.

  • Dear Michael,

    Thanks for your extended support. We have tuned the Hall Sensor signals with respect to the Back EMF.

    Initially the Motor phase signals were not in Sync with the Hall Signals. Now the Motor is running smoothly 

    in both the directions.

    With Best Regards,

    B.Saravana Vikram.