• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 65 subscribers
  • Views 2961 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.

BQ20Z45-R1: Short circuit between VCC and Ground

Mike Mrazik
Expert 1255 points
Part Number: BQ20Z45-R1

Hello,

I currently have 3 packs returned that have 20-30 ohms measured between VCC and GND.  This circuit is setup similar to the circuit in the reference design, with a diode and 1uF capacitor.  The only difference is that we have an added 10 ohm resistor inline to create an RC filter with the capacitor.  On all 3 of these units, we found the 10 ohm was open, which is an indication that this was drawing too much current.  Sure enough, we are getting the short circuit on VCC.

What can cause this failure?

Thanks,

Mike

  • 0
    TI__Mastermind 42535 points
    Mike,

    I'll be looking into this. Give me time to check with the hw engineers here to get back to you next week.
  • 0 in reply to Batt
    TI__Expert 5090 points
    Hi Mike,

    When you say short circuit on VCC, do you mean it is grounded due to the 10-Ohm resistor being open?

    If so, then one possibility might be excessive current flowing while the 1uF cap is being charged. If you had a case where the FETs were off, the VCC should be powered through the CHG FET body diode. Say you had a min cell level, such as 2.5V/cell, so 10V on the stack. The voltage at the VCC pin would be ~8.6V, after one diode drop through the CHG FET, another drop due to your series diode. Then if you had a high voltage applied from a charger, such as the 17.4V we have discussed separately, that voltage would flow through the DSG FET body diode and VCC diode, trying to charge the VCC pin to 16.0V. So you could get a transient current starting at ~(16.0V - 8.6V)/10-Ohms = 740mA through the 10-Ohm resistor, exponentially dropping as the 1uF cap charged. Your time constant is ~10usec, so it won't last long, do you think your resistor can withstand this?

    Thanks,

    Terry
  • 0 in reply to -Terry
    Expert 1255 points
    Terry,

    Between the VCC pin and GND, we are measuring 20-30 ohms (varies across 3 boards). The 10 ohm resistors were open on all of these boards as well, but there was no sense in replacing them since the 20-30 ohms were measured directly across the part. We found this because our packs wouldn't communicate, when investigating we found battery voltage on the input side of the 10 ohm, and nothing on the output side. Then we did the ohm measurement.

    Thank you for explaining the transient current across the 10 ohm, that is good data. Our part is an 0603 package, so at first glance we may think the part isn't sized appropriately because that is 5.476W across a 1/10th watt part (740mA x 740mA x 10). However, I pulled the datasheet for a Stackpole part (one of the parts we use), and the 0603 case size can handle 70W for 100uS, so our resistor will be able to withstand the 5.5W for 10uS.

    Thanks,
    Mike
  • 0 in reply to Mike Mrazik
    Expert 1255 points
    Terry,

    I noticed this is marked as "TI thinks resolved". I don't believe that is true, what resolution was made?

    Mike
  • 0 in reply to Mike Mrazik
    TI__Genius 14985 points
    Hi Mike,

    What is the resistance you measure at VCC on good ICs/boards? I'm assuming much higher but just want to make sure.
    If the VCC pin was damaged then Electrical Overstress (EOS) is the most common cause. Is there any chance that a path to the pin was exposed externally and could have seen either an ESD zap or an extreme positive or negative voltage by accident?
  • 0 in reply to dMax
    Expert 1255 points

    On a good board, I measure open loop between VCC and Ground.  

    You asked: Is there a chance that a path to the pin was exposed externally and could have seen either an ESD zap or an extreme positive or negative voltage by accident.

    I'd assume yes, during assembly someone has to solder the wire from the top of the stack onto the PCBA.  The only thing between VCC and the top of the cell stack is series resistance of the traces, the CHG fet body diode, my inline diode, my inline 10ohm to make an RC filter w/ the 1uF capacitor.  What else should be there?  What does TI have internally to clamp ESD spikes?

    Mike

  • 0 in reply to Mike Mrazik
    Expert 1255 points
    dMax,

    Any update on this?

    Mike