• State Resolved
  • Locked Locked
  • Replies 9 replies
  • Subscribers 30 subscribers
  • Views 1541 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.

Original question:

DRV8703-Q1: SPI Communication problems

DRV8703-Q1: Chip becomes hot

Klemen Zivkovic
Intellectual 870 points
Part Number: DRV8703-Q1
Other Parts Discussed in Thread: TIDA-01389

Hi!

I was using chip for half hour, moving motor back and forth and than suddenly while playing with PWM settings, DRV8703-Q1 became very hot, and start eating 0.3A on its own.

It seems SPI still works on chip and if I leave it running for 25seconds, I get over temperature warning over SPI.

What could be wrong? how can I troubleshoot? Is this chip so sensitive that some voltage spike from Vm could make it go this way :).

Also since MLCC of 10uF / 60V are quite hard to find these days (https://www.ttiinc.com/content/ttiinc/en/resources/marketeye/categories/passives/me-zogbi-20180302.html - I decided to go without them - could that be reason of failed chip?

regards

  • 0
    TI__Expert 8985 points

    Hi Zhivko,

    We do not recommend you operate the device without the bulk cap and it may be the cause the issue.

    The way to troubleshoot is to check the OCP protection is violated by analyzing if your circuit reaches any of the following OCP thresholds. Dependent on your schematic.

  • 0 in reply to Hector Hernandez Luque
    Intellectual 870 points
    Hi Hector,

    I am getting:
    VDS monitor overcurrent fault
    and
    VDS monitor overcurrent fault on the high-side FET of half bridge 1

    Does this means high-side FET is damaged? How could it damage with voltage only on 24V??
    Is there any way I can confirm this measuring FET on board with voltmeter, without desoldering it?
    My mosfets are: IRFS7534 (Vdss = 60V, Id = 240A).

    regards

    ...
  • 0 in reply to Klemen Zivkovic
    TI__Expert 8985 points

    Hi Zhivko,

    Please verify by checking what setting you have on the Vds_ocp trip written through SPI (or if its chip default).

    Then you can measure with a trigger in your scope with a differential probe from GH1 to SH1 or from GH1 to SH1 (the latter you must take into account the additional voltage below the switch node).

    That way, you will know if you meet the SPI set OCP threshold.

  • 0 in reply to Hector Hernandez Luque
    Intellectual 870 points
    I haven't changed Cds_ocp - so it must be default.
    GH1 to SH1 is 400mV.
    Is your statement correct regarding: " GH1 to SH1 or from GH1 to SH1" - isn't this same in each case?
    Why would chip heat up in case of OCP situation? BEcause as I mention earlier it keeps getting warm - current on power supply show 0.246A bbut only when I connect 3.3V to chip (Vp = 12V is connected constantly).
  • 0 in reply to Klemen Zivkovic
    TI__Expert 8985 points
    Hi Zhikvo,

    To what pins are you connecting the 3v3?

    The chip does not heat up due to OCP, it heats up due to the amount of power requiring to drive your load or handle the settings you have set. Whatever increases the current load on the bridge and violates your Vocp thresholds. Reducing switching speeds, voltage magnitudes,, gate drive current and the load current can assist in reducing all of this heat. If you know how much current you want your load to require, then calculate based on the Rdson of your Fets what Vocp thresholds you should set.

    Regarding my comment, I meant VDRAIN to SHx or SHx to SP2/SL2 (the sense resistor).
  • 0 in reply to Hector Hernandez Luque
    Intellectual 870 points

    I apply 3.3V only to NSLEEP signal.

    Load is disconnected from board while driver is getting hot.
    While I was driving motors before, gate driver was not getting hot, but then not sure after what conditions it is getting hotter.

    OCP values are not changed over SPI anyhow.

    While observing driver getting hotter - I am not driving it with PWM anyhow. PWM signals are all flat 0V.

    regards

  • 0 in reply to Klemen Zivkovic
    Intellectual 870 points

    I just measured voltage on gates of High mosfets - they are both 700mV. PWM's are all flat 0V - so why is gate driver outputting 700mV to high side mosfets?

  • 0 in reply to Klemen Zivkovic
    Intellectual 870 points
    Just soldered new chip on board - and now it works - I wonder what happened to that chip, what could happen to it to output constantly 700mV to gate of high side fets ... Could it be that:
    1.) missing caps are the cause
    2.) I dont have reverse protection in place as suggested in TIDA-01389_Sch.SchDoc (www.google.com/url
    3. Any other idea?
  • 0 in reply to Klemen Zivkovic
    TI__Expert 8985 points

    Hi Zhivko,

    Good to hear. Reason 1 is a possibility. Bulk caps are integral to driver operation.

    Now, another idea is that the soldering of the previous IC to your board was not effective, adding additional resistance and requiring more power to operate.