• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 29 subscribers
  • Views 1256 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

DRV8301: Catastrophic failure, transistors and DRV

Larry Stevens
Prodigy 40 points
Part Number: DRV8301
Other Parts Discussed in Thread: TIDA-00281

Hi,

We have a custom board design based on the drv8301kit_D. We have had several failures of the DRV8301 driver in which a hole is blown out of the chip. This usually is accompanied by at least one or more of the MOSFETs failing. The schematic design is almost identical to the drv8301kit_D, the MOSFETS are TI CSD19535KTTT. 

We are running at a 48v bus.

Various motors, absolute encoder sensing so we can get maximum stall torque.

The first failures happened while running off a bench power supply at higher speeds, we assumed that braking action bumped the bus voltage and caused the failure. The last failure happened with the motor locked (only some small play in the transmission so not a lot of motor motion). This did have a high current command ~50A that was sent to the motor.

Any help is appreciated.

Regards,

Larry Stevens

  • 0
    TI__Guru** 114905 points
    Hi Larry,

    You are describing an electrical overstress event. To determine the cause, can you place protection circuits around the system when performing these types of tests?

    If you can add Zeners or TVS clamps, then you can examine the board for possible violations of absolute minimum and maximum voltages. You may not see the violations after the clamps have been added. The fact that the clamp is activating is a clue that this could be a source.

    Another possible method is to run the motor at lower speeds and duplicate the events that led to the failure. This can also provide clues.
  • 0 in reply to Rick Duncan
    Prodigy 40 points

    Hi Rick,

    What parts of the system do you recommend placing clamps on? I'm assuming Gate to Source connections, we already have a TVS clamp on PVDD.

    Thanks,

    Larry

  • 0 in reply to Larry Stevens
    TI__Guru** 114905 points
    Hi Larry,

    I was thinking PVDD, but you have that covered.

    Please examine the SH_x pins for negative voltages. Please let me know how low the voltage is with respect to PGND.

    A scope shot would be ideal.
  • 0 in reply to Rick Duncan
    Prodigy 40 points

    Hi Rick,

    It looks as though the SH_x pins are getting between 10 and 15v below PGND. We didn't run it as hard as the failure points and the transitions seemed to get smaller with high current (although that could simply be due to the commutation point moving slightly). I'll try a Schottky on th e SH_x connections as well as clamping the gates.

  • 0 in reply to Rick Duncan
    Prodigy 40 points

    Hi Rick,

    More catastrophic failure, we placed quite a bit of protection circuitry around the MOSFET gates to protect them against Vgs failure modes. These came from the Automotive 48v 1kw Motor Drive design, TIDA-00281, schematic that uses the same MOSFETS but a different gate driver.  While the MOSFETs don't appear to be damaged the DRV8301 vaporized three pins and blew a hole in the package. PVDD1 pin is gone. This failure looks similar to the one described in this post (e2e.ti.com/support/applications/motor_drivers/f/38/t/200671) but more dramatic"  

    We were also unable to remove the DRV8301 intact from the board so I've attached pics of the damaged section of the die. This chip was sourced through Digikey. We were exercising the circuit quite vigorously at the time with high current (+50A) pulses through the MOSFETS and trying to generate the maximum BEMF that the circuit would receive. The power was being supplied by a LiFePO4 battery pack that was quite capable of sinking any generated current. We currently have 5 ohm gate resistors so we'll increase those to 10 ohm, turn the gate drive to minimum and increase the dead time to maximum value. Any suggestions will be greatly appreciated.

    Larry Stevens