• State
  • Locked Locked
  • Replies 10 replies
  • Subscribers 63 subscribers
  • Views 2784 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.

BQ24610: About boostrap diode issue

Yang Ken
Intellectual 500 points
Part Number: BQ24610

Hello Sir,

Our customer met IC damage issue on Pin20(PH) & Pin22(BTST) short together, and we reviewed their schematic then found they used boostrap dioe is general diode not schottky diode, so we recommend to change it and test again.

The result seens good to solve this damage issue, but we have to explain why using general diode will let IC damage on Pin20 & Pin22 shorted?

Could you please help to provide this explanation?

Thanks

Regards,

Ken Yang

 

  • 0
    TI__Genius 14430 points
    Hi Ken,

    The bootstrap diode needs to be chosen as fast recovery device to minimize the charge fed back from bootstrap cap to Vcc(REGN). So we usually use schottky diode as bootstrap diode. Besides PH and BTST, are general diode and REGN also damaged? Is that any other component damaged?

    Thanks,
    Ann Lien
  • 0 in reply to Ann_Lien
    Intellectual 500 points

    Hi Ann,

    Thanks for you promptly reply.

    We got total three damaged parts so far, there is one Pin 18(REGN) shorted to Pin 17(GND) but the diode seems good.

    To replace new part, the PCB works nornally but not sure when it will damage again?

    Thanks

    Ken

  • 0 in reply to Yang Ken
    TI__Genius 14430 points

    Hi Ken,

    About the losses, we could roughly calculate total losses in whole bootstrap circuit. That is why we need to add schottky diode which is better than general diode because of its fast recovery capability. For the real damage root cause, we couldn't figure out since there is no waveform to show what it's going on. But the recovery losses could be the reason why it damaged. REGN, PH and BTST are all damaged represents its high side driver damaged. It may be caused since the recovery losses is over the driver limitation. Using schottky diode could solve this problem and we usually use schottky diode as bootstrap diode. I think it would be fine and not be damaged again if you change general diode to schottky diode.

    Thanks,

    Ann Lien

  • 0 in reply to Ann_Lien
    Intellectual 500 points

    Hi Ann,

    Thank you again for the reply.

    Could you please teach me how to calculate recovery losses and what's the driver limition of bq24610?

    Do you mean the trr parameter of diode specification? The general diode vs schottky diode seems no more different.(See the following figures.)

    I can capture some waveforms if you need it to judge the damage root cause.(Let me know the test conditions & points.)

    Regards,

    Ken Yang

  • 0 in reply to Yang Ken
    TI__Genius 14430 points

    Hi Ken,

    The bootstrap diode losses  is the sum of the forward bias power loss that occurs while charging the bootstrap capacitor and the reverse bias power loss that occurs during reverse recovery.

    • Forward bias power loss = forward voltage*I_diode. Idiode=Qbootcap/t_low_side_on. 
    • Reverse bias power loss depends on recovery current and recovery time. We could estimate roughly how many coulomb discharge from bootstrap cap to REGN. 

    Could you share the whole schematic to us? We can discuss further if there is any other circuit would cause this issue.

    For the diode, the general diode's trr is almost the same as schottky diode. So maybe trr is not the reason on this issue.

    Could you please measure REGN, BTST, PH before/after this issue? We can discuss with waveform to figure out which one is the very first damaged and causes other circuit damaged.

    Thanks,

    Ann Lien

  • 0 in reply to Ann_Lien
    Intellectual 500 points

    3S CHARGE.pdfHi Ann,

    Thanks for you explain about boostrap diode losses calculation.

    Attached file is the customer's schematic for you review.

    This issue is not easy to duplicate, both can work normal with using general or schottky diode and the waveforms look like similar.(Please see the following waveforms.)

    CH1 : PH

    CH2 : BTST

    CH3 : REGN

    Figure1 => General Diode

    Figure2 => Schottky Diode

      

    Thanks

    Ken

  • 0 in reply to Yang Ken
    TI__Genius 14430 points
    Hi Ken,

    I didn't find design error in your schematic. One thing need to be confirmed, is this issue occurred many times/many board or just a single case? Since you tested general diode again and there is no damaged in your test, I'm wondering if it is a single case. Based on general and schottky diode datasheet, trr/IF parameter are almost the same value. Even based on your test result, they looks no different. Please kindly check if it is a single case for your design.

    Thanks,
    Ann Lien
  • 0 in reply to Ann_Lien
    Intellectual 500 points
    Hi Ann,
    This isn't single case, that's failed field return and we have been recevied 3 pcs failure ICs so far.(Those 3pcs damage symptom are very similar.)
    We had recommended to change the diode from general to schottky then test it again, the customer responses the test result seems good, but we don't know how to explain why change to schottky be able to resolve this issue due to general & schottky testing waveforms look no big different?
    Should I send these damaged parts to FA? Any help from he FA report?

    Thanks
    Regards,
    Ken
  • 0 in reply to Yang Ken
    TI__Genius 14430 points

    Hi Ken,

    My mail address is ann_lien@ti.com. In order to speed up the status, we can discuss it by mail.

    For the debug direction, we thought the diode is damaged in the beginning. But the diode seems good after the damaged, it represents diode maybe is not the key point at this issue. Could you please submit FA report first? We can know if there is any other pin or function is damaged. And we can discuss the further step.

    Thanks,

    Ann Lien