• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 388 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.

LP87644Q1EVM: EVM won't enable bucks & performs orderly and immediate shutdowns

Zareef Huq
Intellectual 2325 points
Part Number: LP87644Q1EVM


Hello,

I have an LP8764 EVM I was working on and I accidentally shorted the switch node for Buck 3 and the output side of C6 while trying to probe the switch node, and the EVM stopped working properly. EN_DRV never goes high anymore, and none of my bucks are turning on. Manually turning them on in the control registers do nothing. I haven't made any other hardware or software changes. Looking through the registers on the PMIC GUI, I'm seeing lots of communication errors. It fails BIST, and raises interrupts for the following errors: 

REG_CRC_ERR, BIST_FAIL, COMM_FRM_ERR, COMM_CRC_ERR, COMM_ADR_ERR, and RECOV_CNT_INT

Looking through the datasheet, there is an error with the SPI frame, I2C is writing to a register it shouldn't be messing with, there is CRC mismatch, and it unsuccessfully tried to recover 16 times (according to the threshold value set for RECOV_CNT). What could be causing all of these problems? None of the INT_SEVERE_ERR bits are set, all the errors above are set in the INT_MODERATE_ERR register. Thank you in advance.

  • 0
    TI__Expert 5111 points

    Hello Zareef,

    The expert who is in charge for this device will be back on Monday and may have more insights into this issue, whether this seems like a damaged device or not.

    Could you please describe in more details the debugging steps you already took ? What happens if you try to clear the interrupts in the Quick-start page ?

    It seems like the register map check fails and that there is potentially issues on the SPI as well.

    Best regards,

    Florian

  • 0 in reply to Florian Gabelle
    TI__Intellectual 2325 points

    Hi Florian,

    I've looked through the registers I mentioned above to find errors, and I've tried flashing the basic NVM template onto it, but that didn't help. In addition, I've tried checking C6 with a multimeter and the capacitor didn't seem like it was damaged. I've tried resetting the recovery count register but it crashes the EVM/GUI and I have to restart the web app and power cycle the EVM. I can try clearing the interrupts and see what happens, but I can't get it out of the shutdown state so the bucks can't be turned back on, even if I set them to turn on in the control registers manually. I have probed VCCA and it still shows 3.3V when plugged into USB (which is how I've always worked with this EVM).

    I have ordered a replacement PMIC if all else fails and I need to put in a fresh device. I've looked through the EVM and none of the passives look visibly damaged. 

  • 0 in reply to Zareef Huq
    TI__Genius 13860 points

    Hi Zareef,

    Just to double confirm, you have been using original NVM for testing? No any non-volatile changes made?

    Have you checked VIO and VOUT_LDO voltages are those present and correct? VIO based on selection either 3.3V or 1.8V and VOUT_LDO 1.8V. From the right side of the EVM you see _S marked test points to check voltage close to device.

    How about communication to IC, is it working, meaning you can read for example register 0x01?

    This sounds device might have got damaged but how is not clear.

    Br, Jari

  • 0 in reply to Jari Niemela13
    TI__Intellectual 2325 points

    Hi Jari, I'm seeing the proper voltages on VIO and VOUT_LDO, and I am able to find 0x01:

  • 0 in reply to Zareef Huq
    TI__Genius 13860 points

    Hi Zareef,

    Looking again interrupts it doesn't make sense you are getting all of this only from BUCK3 short case. BUCK3 short must report BUCK3 related interrupts.

    REG_CRC_ERR means the NVM content is not matching to calculated CRC somehow and device goes straight to SAFE_RECOVERY state from the beginning. Device cannot get out from there without NVM is re-configured.

    Br, Jari