• State Resolved
  • +1 person also asked this people also asked this
  • Locked Locked
  • Replies 4 replies
  • Subscribers 30 subscribers
  • Views 175 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.

DRV10983EVM: Burning Mosfets when powering Eval-Board

Frank Mammen
Prodigy 80 points
Part Number: DRV10983EVM
Other Parts Discussed in Thread: DRV10983

Hello everybody,

I'm facing a problem with DRV10983-Evalboard:

Started with Board No. 1:

Connected all needed wires and checked connections and polarity to make sure nothing goes wrong when powering up (24Vdc from linear PSU).

When plugging in 24V the Current Limit of the PSU comes up and stays a 3 Amps. After checking the three Halfbridges of the DRV10983 it showed up that four of six Mosfets are shorted.

 Changed Eval-Board

Started with Board No. 2:

Connected all needed wires and double checked connections and polarity to make sure nothing goes wrong again when powering up (24Vdc from linear PSU).

No current limiting and smoke when powering up. BLDC tried to turn and stopped. Disconnected 24V made some checks and powered up again.

Same damage as on Board No.1, after checking the three Halfbridges of the DRV10983 it showed up that four of six Mosfets are shorted.

Any ideas what could cause this problem?

Best regards

Frank

  • 0
    TI__Mastermind 24405 points

    Hi Frank,

    DRV10983 is rated up to 28V so applying 24V to the EVM shouldn't cause any problem. You mentioned that four of six MOSFETs are shorted. DRV10983 is an integrated FET device. How did you conclude that the 4 MOSFETs are shorted? Can you share the picture of the EVM board?

    Regards,

    Vishnu

  • 0 in reply to Vishnu Balaraj
    Prodigy 80 points

    Hi Vishnu,

    The DRV10983 didn't show any smoke because my PSU came up with current limitation. I checked the Mosfets using a Diode-Tester:

    + Testpin  connected to DRV10983-GND and checking  with respect to the phase outputs (U, V, W).

    - Testpin connected to DRV10983-VCC and checking with respect to the phase outputs (U, V, W).

    Four of the Mosfets showed a short instead of ~ 0.5V.

    But it might be possible that an Overvoltage is caused by the Inrush-Current. The inductance of my powercable is appr. 1µH, the Bypass-Cap placed on the Evalboard shows 4.7µF.

    LTSpice-Simulation shows an VCC-GND voltage of ~ 50V, need some Caps for damping the resonance.

    If so, you should redesign the Eval-Board.

    I let you know.

    Regards

    Frank

  • 0 in reply to Frank Mammen
    TI__Mastermind 24405 points

    Frank,

    DRV10983 EVM has been in the market for more than 5 years and we haven't seen any such issues. Also, we test all our EVMs before it leaves TI. I don't think this is related to the EVM design. Can you connect a Zener diode (rated less than 28V) across Vcc and GND and see if this helps resolve the voltage spike issue?  

    Regards,

    Vishnu

  • 0 in reply to Vishnu Balaraj
    Prodigy 80 points

    Vishnu,

    I made some more tests by reducing the Bulk-Capacitor.

    Starting with 2200uF parallel to VCC-GND:

    As can be seen, VCC is coming up smooth and no Overshoot.

    I did this tests down to Cbulk = 4,7µF and got an Overshoot up to 28,9V:

    Going further down with Cbulk would for sure burn the Chip so I stopped reducing Cbulk.

    I'm using 220µF for now and it seems to be OK.

    Regards

    Frank