The TI E2E™ design support forums will undergo maintenance from July 11 to July 13. If you need design support during this time, open a new support request with our customer support center.

  • State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 63 subscribers
  • Views 430 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.

UCC21520: Short-circuit on three phase full bridge application with SiC.

Fabio Posser
Prodigy 235 points
Part Number: UCC21520
Other Parts Discussed in Thread: UCC21530, UCC21551, UCC21550,

Hi,

I'm facing some issues with the PN UCC21520QDWQ1.
Some random short-circuits are happening, damaging the full bridge.
Unfortunately, I could not get any image until now during the failure, because it is random.
Please check the schematic below.
You have any advice for this or saw this happening with this PN before?
I'm using an insulated power supply for each output, Low and High side.

  • 0
    TI__Expert 8616 points

    Hi Fabio,

    With the 10k deadtime resistor connection, there should be short circuit prevention between the low side and high side outputs. Is the current deadtime configuration enough deadtime for the switches? Would increasing the deadtime prevent these short circuits?

    If increasing the deadtime still does not prevent the short circuits, this could be a case of a dV/dt induced turn-on. This occurs when high dV/dt in the FET causes current to be sent through the Miller capacitance of the FET that the gate driver is driving. This current creates a voltage at the gate, which will accidentally turn on the device if larger than the threshold voltage. 

    Couple ways to prevent dV/dt induced turn-ons from occurring would be:

    • Slow down the dV/dt by increasing external gate resistance
    • Smaller gate loops to mitigate parasitic inductances

    I hope this was helpful!

    Best,

    Hiroki

  • 0 in reply to Hiroki Honda
    Prodigy 235 points

    Deadtime today seems to be fine, besides the datasheet recommendation of minimum 2.2nF in parallel with Rdt (I will fix that).

    Based on test results, seems to not be short circuit due to Miller capacitance charge and we use negative votlage to hold the SiC off.

    Seems that this device is from 2016, it was design to work with SiC Mosfets?

  • 0 in reply to Fabio Posser
    TI__Expert 8616 points

    Hi Fabio,

    Yes, this device is designed to work with SiC MOSFETs that can work with the UVLO options that the UCC21530 provides. If a higher UVLO option is a better fit for the power switches used in the system, I would highly suggest looking into the new UCC21551 device which has UVLO options of 12V and 17V. UCC21551 is also more robust and has improvements on some of the specifications.

    The deadtime decoupling capacitor is designed to help with noise immunity, so this could be the cause of this issue. Let me know if increasing the capacitor did not fix the issue. 

    Best,

    Hiroki

  • 0 in reply to Hiroki Honda
    Prodigy 235 points

    Ok, I will do that.
    For a 10kohms on Rdt you recommend some particular value of filter capacitance or 2n2 as show in the datasheet is fine?
    The new PN, UCC21551 have more immunity and robustness compared with the one I'm using today? It is mensurable the improvement? 

  • 0 in reply to Fabio Posser
    TI__Expert 8616 points

    The 2.2nF capacitor should be sufficient.

    The direct replacement for UCC21520 would be UCC21550 which offers the same UVLO options and a disable pin, with design improvements based on past customer feedback. 

    UCC21551 would be the alternative for higher UVLO options (12V and 17V).