• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 63 subscribers
  • Views 133 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.

BQ25798: IC catching fire

Dave Besson
Prodigy 30 points
Part Number: BQ25798

Tool/software:

I'm using the configuration at figure 10-1 in the datasheet, 10.2 Typical Application p. 130., single input source, no optional FETs used (VBUS, VAC1 and VAC2 are connected together)

BQ25798 connected to a 4S battery pack (14.8V)

Power input from a USB-C PD Adapter set at 12V by the BQ25798 (Measured 11.8V at input caps)

Output of the BQ25798 (VSYS) is connected to a MOSFET that is used to turn the circuit on and off with a push button (push on, hold off)

Circuit works perfectly, until the following conditions are met:

Device is turned off.

Power adapter is plugged in, BQ25798 starts charging the battery and the LED turns on.

When the charging cycle is finished, LED turns off.

Device is powered on, instantly the BQ25798 catches fire and keeps burning until I unplug the battery.

If the battery is not fully charged the device turns on normally

If the power input is disconnected the device turns on normally

When both conditions are present (battery is fully charged and the power input is present) powering the device burns the BQ25798. This happened 4 times so far on 2 different devices (both using the same power input / battery setup).

Is it a hardware problem?

Is it a config problem?

  • +1
    TI__Guru**** 166626 points

    Hi Dave,

    Wow!  Haven't seen this before.  Can I see a full schematic?  Did you add a TVS diode on BAT as recommend in the datasheet and EVM user guide?  The BAT/BATP and SYS pins are only rated to 20V.  That MOSFET on SYS may be causing transient that spikes up the voltage, esp at load release.

    Regards,

    Jeff

  • 0 in reply to Jeff F
    Prodigy 30 points

    Hi Jeff, thanks for your quick reply

    I have an 18V TVS at both the USB-C power input and at the battery output, but nothing on the VSYS pin. First I thought my TVS at the battery was the culprit because the max voltage of 16.8V (measured 16.85V) was pretty close to the 18V TVS threshold, shorting the power through the BQ so I removed it for the second test, and the BQ fried again.

    I didn't think about that MOSFET switching on the VSYS pin though, this could be the problem. I'll try with a TVS there as well. This means that the higher voltage from the power source and/or the battery rushing in would cause the transients to be just high enough to clamp that 20V pin. I could also try a soft start on my 5V regulator to lower the inrush. Let me do a few more tests on Monday and I'll post the results.

    Unfortunately I cannot post the schematics on a public forums, but send me an email at the address on my account, and I'll reply privately with the schematics.

  • 0 in reply to Dave Besson
    TI__Guru**** 166626 points

    Hi Dave,

    I will sent a friend request to you in this forum so we can communicate privately.

    Regards,

    Jeff

  • 0 in reply to Jeff F
    Prodigy 30 points

    Hi Jeff

    I'll post here so others can benefit from this info

    Here's what the turn on sequence looks like on the scope.

    There's the Battery voltage DC offset of 16.5V at the time of measurements, I didn't use a fully charged battery just in case. Fully charged battery measures about 16.85V.

    You can see a first spike at ~1.6V and then a second one at ~1.3V which suggests that the 18V TVS I added on VSYS is doing its job.

    16.5V (VBat) + 1.6V (spike) = 18.1V

    To me, this is a good enough indication that the first spike may reach much higher if this wasn't for the TVS, and then toast the VSYS pin if it goes above the 20V threshold of the pin, especially with a fully charged battery. Just to be sure, I will test with a lower voltage battery, like 13-14V to see if the spike still reaches 18V and if so, this will conclude your hypothesis but I'm pretty sure you nailed it.

    Thanks again for your help!

  • 0 in reply to Dave Besson
    TI__Guru**** 166626 points

    HI Dave,

    I hope this fixes your issue.

    Regards,

    Jeff