• State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 64 subscribers
  • Views 1010 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

UCC27201A-Q1: IC failing with a burning hole

Richard
Prodigy 70 points
Part Number: UCC27201A-Q1
Other Parts Discussed in Thread: UCC27201A, , UCC27211A-Q1

We am experiencing repeatable failures in a new design using the UCC27201A (and -Q1 version).  The failure method is consistent; a spot on the chip turns bright red and and burns a hole through the top.   I've went down several avenues attempting to figure out what is causing the failure and I am hoping you can help me focus the issue based on the location of the burn-through.  I have attached an image of the chip - my apologies for the resolution but the microscope was busy and I hope this will be sufficient to identify the general location.

FET's: Large  SOT-227 VS-FA72SA50LC with a 255nC/10nF gate.  
VDD : 12v
Power: 48v
Rgate(hi)=12Ohm, (Diode+3Ohm for pull-down)
Rgate(lo)=3Ohm
LO protected by 16V Zener
HS protected by 75V TVS
Boost capacitor is various: from .1uF to 2.2uF (mostly .1uF).  No correlation found to failure

48V is always on to the FET's before the 12V+5V for Driver+CPU starts up.

The issue has been difficult to reproduce on the bench and capture but is occurring pretty regularly during prototype testing.  The failure always occurs within the first second or so on power-up.  Power seems good although it's possible that HIN/LIN are being driven (5V) before VDD is all the way up to 12V.

The photo should appear after my name.

Thank you!

Rick

UCC27201A

  • 0
    TI__Expert 3200 points
    Richard,

    An AE should be contacting you soon.
  • 0
    TI__Intellectual 2910 points
    Hi Richard,

    I regret learning that you are experiencing these issues in your application. I don't know what part of the driver chip resides in that corner of the device, but I will look into it.

    In the meantime, do you have a schematic or schematic portion you can share? This will help the analysis. Also, if you have oscilloscope captures of the HI, LI, and HB and HS signals during startups when the failure occurs, please provide them as well.

    Regarding HI/LI being driven while the 12V supply is still rising, this should be fine. What will happen is that the UVLO feature of the UCC27201A-Q1 will keep the outputs off until the UVLO threshold is satisfied (typ ~7.1V).

    I will start looking at this from my end and let you know what I find out.

    - Daniel
  • 0 in reply to Daniel Ruiz
    Prodigy 70 points

    Daniel, Thank you for your help.  Unfortunately I do not have a trace yet as it's been very difficult to reproduce so far.  I assumed that undershoot had something to do with it but so far have not seen anything short spikes to -2 on HS.

    I've included the relevant schematic below.  The board is simple, just a small pcb with the driver that sits on top of two SOT-227 fet's.   The boards are paired or tripled to drive a motor.  Note that in the schematic the mosfets have two source connections, 1 of which is selected to be the return current for the driver circuit.

    Thank you for the UVLO explanation.  I can avoid jumping down that rabbit hole now.

    Thank you again,

    Rick

  • 0 in reply to Richard
    TI__Intellectual 2910 points
    Rick,

    Thanks for providing the schematic and additional information. It looks like the spot where you see the damage is right at the Low-side output driver pull-up MOSFET. There could be something holding the LO pin low at times when LI is high and the device is trying to drive the LO node to VDD.

    A few things to try:
    - Try testing without the zeners from LO to VSS. If the issue goes away with this, then try again with a higher voltage zener (18-20V)
    - Ensure there is no overlap between HI and LI in a high-logic state -- a dead time needs to be provided to avoid shoot-through conditions
    - Ensure that neither HI or LI is going more negative than -1V, as this is beyond the device spec and may result in undesired state
    - A small resistance and capacitance can be used to clean up the edges of HI and LI. ~10-100ohm and 10-100pF usually does the trick
    - You could try the UCC27211A-Q1, which is more robust to negative transients on HI/LI and HS

    Please let me know what you find out.

    - Daniel