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.

DRV8343H-Q1EVM: Became unable to drive the motor

Part Number: DRV8343H-Q1EVM

Hello Team,

I'm Yuta Kurimoto FAE from TIJ.

Our customer has been evaluating the DRV8343H-Q1 EVM board and they faced some issue.

Case 1:

During the sensorless motor rotation, they intentionally shorted the phase A and phase B. Then the GUI represented phase fault.

After that they cleared the fault, and they connected the each phase correctly.

However they can't drive the motor anymore. They are not sure the reason. Do you have any idea?

Case 2:

Before starting the motor, they removed the 30A fuse of each phase. Then they started the motor correctly.

After that they attached the fuse again, but they can't drive the motor anymore. They are not sure the reason. Do you have any idea?

I will visit customer tomorrow and will have experiment with customer.

So I'd like to know what the reason is assumed before the experiment.

Thanks,

Yuta Kurimoto

  • Kurimoto-San,

    Thanks for posting on the MD forum!

    I have sent your post to the applications expert who will get back to you today.

    Regards,

    -Adam

  • Hi Kurimoto-san,

    For Case 1, does the nFAULT pin go low when attempting to drive the motor? Can you double-check the 30A fuses? If the motor phases were shorted together, it may have resulted in an over-current event, which could blow the fuse or possibly damage a MOSFET.

    For Case 2, can you clarify if the customer is driving a 3-phase brushless motor, or are they driving a different kind of load? Normally, these fuses connect SHx to DHx, which is necessary to drive a 3-phase BLDC.

    Thanks,
    Garrett

  • Hello Garrett,

    I had experiment with customer yesterday.

    For case 1, Let me close this question. As I saw their EVM board, the fuses were OK but some of the FETs were damaged. They set the VDS as Disable.

    So I recommended them to set the VDS value for lower threshold.

    For case 2, I still don't know the reason. They drive the 3-phase brushless motor by sensorless control.

    They removed the 30A fuse of Phase C ⇒ They enabled the driver and push the starting motor button ⇒ The GUI showed phase fault ⇒ They implemented 30A fuse again ⇒ They enabled the driver ⇒ The GUI showed phase fault before pushing the starting motor button

    I found the phase C low side FET was shorted. The setting of VDS was disabled. The setting of MODE was independent MOSFET.

    I tried the same operation with my EVM, but I could not reproduce it.
    Thanks,
    Yuta Kurimoto
  • Hi Kurimoto-san,

    The independent MOSFET mode/firmware shouldn't be used to spin a motor; it is intended for driving independent loads, like brushed DC motors or solenoids. I am worried that if they used independent mode with the fuses installed, it may have caused shoot-through if they turned on the high side and low side at the same time, damaging the low-side C FET (especially if VDS protection was disabled).

    Can you double-check with the customer which firmware they are loading on the EVM?

    Thanks,
    Garrett

  • Hello Garrett,

    Customer used sensorless firmware.(Somehow the sensorless firmware is pre-loaded in the MCU so they can't choose other firmware.)

    Thanks,

    Yuta Kurimoto

  • Hi Kurimoto-san,

    The EVM comes pre-programmed with the sensorless firmware, but can be re-programmed with the other firmware options if needed. I think the issue is likely related to setting the "MODE" pin to independent MOSFET mode. The sensorless firmware is intended for use with 6xPWM mode, and using independent MOSFET mode will bypass the internal shoot-through protections.

    Was there a particular reason the customer was using independent MOSFET mode?

    Thanks,
    Garrett

  • Hello Garrett,

    No, There is no particular reason. The reason is that the independent MOSFET mode was default setting at the shipment.

    I told customer to use 6*PWM mode and to enable the VDS threshold. Then customer understood.

    I heard they had tested the board on the cushioning material which is contained in the EVM packing. It might be a root cause.

    Thanks,

    Yuta Kurimoto