• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 3 answers
  • Subscribers 63 subscribers
  • Views 720 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.

BQ76940: I have few problems when bringup my PCB, please help me!

nguyen phong
Prodigy 100 points
Part Number: BQ76940

I have a project with BQ76940. It monitor 14 cell, maximum voltage 59V.

 

I have 2 problem:

- Some time, CHG pin (pin 2) was dead. I don't control it (DSG was still normal). Why did it die?

- About SCD threshold, when IC meet SCD threshold, mosfet DSG was dead. I used 3 DSG mosfet, it only dead 1 mosfet (2 mosfet was still live).

  I add resistor 100 ohm in DSH pin to mosfet and capacitor 47nF D->G but i was still dead.

Please give me some advice.

Thanks

  • 0
    TI__Guru 54715 points
    Hi Nguyen.
    For the CHG pin damaged, check for a short across D24 or R65, or a low value at R65 which would allow excessive current into the CHG pin. Check the bad board for errors, measure on a good board looking for overstress conditions.
    For the discharge FETs, when FETs are paralleled and one turns off, the load shifts to the other. Some FETs when paralleled can oscillate during switching causing heating and failure. A small resistance in each gate compared to the common resistance or a ferrite bead is sometimes suggested to avoid this situation. There can also be layout effects with either routing or power dissipation differences. Once one FET fails shorted the stress is releived on the others which are not damaged. You might consult with a colleague, instructor, reference book, or your FET supplier's representative for how to handle parallel FETs or anything specific with that part number. Measuring the FET switching signals with lower currents may show a condition which needs to be corrected to survive at higher currents.
  • 0 in reply to WM5295
    Prodigy 100 points

    Thank for supporting.

    I will check again.

    About Rsense, now i used 3 resistor 3mOhm parallel (=1mOhm). Follow EVK, i use filter RC, R=100Ohm, C=0.1uF but OCD threshold is not true.

    I set 50A but it met OCD at 30-34A

    Please give me some advice.

  • 0 in reply to nguyen phong
    TI__Guru 54715 points

    Hi Nguyen,
    Be aware and check the tolerance of the resistors. If the lot is running in the high end of a large tolerance, less current will be needed to reach the IC trip point.
    Check the routing of the sense wires from the current path to the filter for the IC pins. The circuit will sense to the current path. If you have resistance on the board it will add to the effective sense resistance. Also consider the solder resistance.
    Check your register settings for PROTECT3. Early data sheets has a typo in the reset value causing some users to set the reserved bit 3 to 1. This causes protection at a lower voltage (current). Reserved bits should not be set. Be sure to use the latest data sheet.

    Also be aware of the tolerance on the current protection thresholds in the data sheet.  OCSCALEERR may be 10% although not typical.

  • 0 in reply to WM5295
    Prodigy 100 points

    Hi

    Thanks you

     To do some other function I add 1 resistor 100K (R83) like picture, Is there a problem?

  • 0 in reply to nguyen phong
    TI__Guru 54715 points
    Hi Nguyen,
    R83 creates a leakage path around the FETs, so the FETs are not able to stop current flow. While R83 reduces current, it may not sufficiently prevent overcharge or overdischarge. Overcharge is usually more of a concern, but check your system requirements.