• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 63 subscribers
  • Views 360 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:

BQ76930: TS2 incorrect reading when 10K resistor not install

BQ76930: waking up upper part with puls on TS2

Michael Hutmacher2
Prodigy 20 points
Part Number: BQ76930

We learned, that the BQ76930 consists of 2 similar chips in one housing.

Because of this the wake up with pulse on TS1 waking up the hole chip, works also with pulse on TS2 NTC Input on upper part.

On first designs we had no additional caps in parallel to TS2 NTC which lead to odd behavior of complete chip.

With longer NTC wires  connected to TS2 the BQ  in shutdown can wake up only the upper part, by flicker EMC noise couppling to NTC Wires

This can be proofed by measuring the voltage on CAP2

the lower part of the chip stays off, regout is not powered.

If a pack is staying for longer time in this condition the pack is getting unbalanced by the currents drained from upper part.

The adding of 4n7 over to TS2  improved situation.

Starting the chip on TS1  and shutting down by software brings the complete chip in shutdown again.

But we still find a few batteries with only the upper part active.

What might be the course of these findings ?

  • 0
    TI__Guru* 81034 points

    Hi Michael,

    The threshold voltage to trigger a boot to the device ranges from 300mV to 1V and the time for it to be held high ranges between 10us and 2000us. So you need to guarantee that there is no pulse above 300mV as long as 10us reaching this pin to ensure the upper device does not turn on accidentally.

    The addition of the 4.7nF capacitor is something that is used in some of the TI reference designs, so it is a good step. The purpose is to filter noise for this type of situation. Is there any way to reduce the long wire length on the NTC to reduce the noise coupling? 

    Best regards,

    Matt

  • 0 in reply to Matt Sunna
    Prodigy 20 points

    Hi Matt,

    thanks for your reply. This confirms our thoughts. With the 4n7 at the TS2 input  we could not reproduce the single boot of the upper part of the BQ in the engineering environment.

    But in seldom cases we find batteries from prodution with End of LIne Tests passed, which are unbalanced because of this.

    So we are searching for what might cause this, even the capacitor is installed.

    As additional Information. There is a single connector for cell voltage measurement and the NTC going to TS2.

    During production this might be plugged in that the upper part of BQ is wakened up. But in the next Step at the EOL Test the BQ is wakened up by Pulse on TS1, all the tests are done, And after this the complete BQ is shut down by microcontroler.

    With this procedure it is proofed, that complete chip is in shutdown.

    So our concerns are, is there another possibilty for the unwanted wake up of upper part, or is there another problem on manufacturer side, or even in storage or transportation.

  • 0 in reply to Michael Hutmacher2
    TI__Guru* 81034 points

    Hi Michael,

    One other possibility is that the VC5X voltage is driven up it might pull current from the internal signals and could possibly wake up the upper part. This is not something we have seen before - if the caps are close to the IC, I don't think it would be possible to drive VC5X up suddenly. 

    I cannot think of anything that would wake up the upper device after the chip exits production other than noise coupling...