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.

TLV62569: VIN over voltage issue

Part Number: TLV62569
Other Parts Discussed in Thread: TLV62565

Hi Sir,

This is our circuit, we have some failure rate from the market, we are trying to debug, currently we doubt it seems there is a potential overshoot voltage on VIN to cause IC damaged.

Based on the measurement, you can see the VIN has overshoot and follow the PIN3 SW node, it could be through the TLV62569's body diode and cause VIN(max) higher, the overshoot voltage duration is <10ns,

sometime we can see it over 6V on VIN, is this expected or this is an issue ? If this is expected, does TI has spec or criteria of this kind of behavior on VIN ? If this is an issue, it seems a TLV62569 potential issue,

please provide your comment and suggestion.

thanks.

Conditions : Vin : 5.2V, Vout : 3.3V and 1.5V, Load : 0.4A@3.3V, 0.2A@1.5V

CH2 : TLV62569 VIN (U3.4)

CH3 : TLV62569 SW (U3.3)

CH4 : +5VE

CH1 : +5VE

CH2 : TLV62569 VIN (U3.4)

CH3 : +3.3VE

CH4 : TLV62569 SW (U3.3)

  • Hello Vincent,

    Thank you for coming to us with your queries.

    The SW pin voltage overshoot is within specs as long as it is limited to <9V with a pulse duration of <10ns (as indicated in the datasheet). For the case of VIN, can you share waveforms where it exceeded 6V? Please include the SW pin voltage in the plot.

    Please provide the following items below for better understanding of the issue.

    1. At what condition does the IC fail (steady-state, power-up, power-down, low or high temp)? Does it happen with the end customers?
    2. Are you able to duplicate the IC failures in your lab?
    3. The schematic diagram shows TLV62565 instead of TLV62569. Which device is correct? Can you share the schematic diagram of the 3.3V rail also?
    4. Are 3.3V and 1.5V rails used TLV62569? Does both ICs fail in the same time?
    5. Can you share a screenshot of the PCB layout of 3.3V and 1.5V rails?

     

    Best regards,

    Excel

  • Hi Excel,

    Thanks for your reply, i answer your questions below,

    1. This issue was return from market (our end user), The IC fail at steady-state, but have power current different in sleep mode and operating mode, we

        have confirmed there is no heat or temperature issue.

    2. So far, we are not able to duplicate same issue at our lab.

    3. The original design was used TLV62565, but we changed to TLV62569 since end of last year due to material shortage issue, both are compatible, we

        just add TLV62569 as 2nd source in BOM but not change the schematic. Currently, the fail issue is on TLV62569, attached file is our power diagram.

    4. We are using TLV62569 on 3.3V, 1.5V. 2.5V and 1.1V, the fail qty. of IC are not the same on different boards, for example we got both 3.3V and 1.5V fail

        on some board, but some board only fail at 3.3V. The higher failure rate is on 1.5V, and we got TLV62569 Vin pin short to GND on failure IC.

    Attached files are schematic, 3.3V & 1.5V layout and waveform captures.

    TLV62569 sch_layout_waveform

    BTW, we have followed TI's document slva494a.pdf to do measurement.

    There is a little urgent, please help to provide comment and suggestion.

    thanks.

  • Hello Vincent, 

    Thanks for sharing the related files.

    The schematic diagram and PCB layout look fine. And I don't see any strange behavior on the shared waveforms.

    Can you share the Vin waveform at the startup phase? There might be voltage overshoot that exceeds the absolute max rating.

    By the way, can you revert to TLV62565? It has a higher absolute max rating of 7V compared to 6V in TLV62569.

    Best regards,

    Excel

  • Hi Excel,

    We can confirm there is no overshoot at the startup phase of Vin, the only concern from my view is Vin's ripple (or overshoot) is following SW's transition

    and very close to Vin max. rating, please help to provide some explanation on below questions,

    1. Why Vin is following SW's transition ? is it expected ?

    2. Because the SW's overshoot is high at PFM mode during transition, the Vin is close to 6V rating and may have chance to over it even in very short

       period (1~3ns), is this still safe or a concern ? why ?

    We are asking more detail is because we need to evaluate how is the risk on the market, because we have 50K pcs product on the market now, if this is

    expected (Vin is following SW's transition) and safe, we have to turn to other direction.

    thanks.

    CH1 ; 5V source

    CH2 : Vin of U3(TLV62569)

  • Hi Vincent,

    Ian Wang will contact you to arrange a concall tomorrow to discuss your queries in details. And we hope that is fine with you.

    Best regards,

    Excel

  • Hi Excel,

    Yes, i got message from Ian and i've sent webex invitation, if you did not receive the invitation forwarded from him, please inform me, thanks.

  • Hi Vincent,

    I received the meeting invitation. Talk to you tomorrow.

    Best regards,

    Excel

  • Hi Excel,

    Appreciate your file and support, we have moved to other direction now, i will close this case.

    thanks.

  • Hi Excel,

    Need your help again, we have one more question from customer, i would like to have your comment.

    Currently, we got 10pcs failure TLV62569 IC, we measured the resistance of VIN pin to GND and SW in to GND, we got below results.

       - SW pin has short to GND on 10 of 10 pcs failure ICs

       - Vin pin has short GND on 2 of 10 pcs failure ICs, (which means there are 2 pcs failure ICs has damaged on both Vin and SW pin)

    Because the most of damaged are on SW pin, and based on our discussion on last meeting, the possible root cause could be from outside of AC power

    (power environment), customer is asking,

      1. If there is a spike from outside of AC power and through board to TLV62569 Vin, will it just cause SW pin damaged but not Vin ?

      2. Is there any other possible root cause you are thinking that could make the SW pin damaged but not Vin ? 

    My idea is, it is possible, because if Vin's voltage is higher (from outside of spike), the SW pin voltage will be higher and also plus overshoot by FET

    switching itself, both Vin and SW pin AC max. rating voltage are the same (max. 9V, <10ns), the SW pin could have higher failure rate than Vin.

    ex. According to our measurement, the FET overshoot itself is around 1.72V, if there is a 8V spike input to Vin, the SW pin could have 8V + 1.72V = 9.72V

          In this case, the Vin should be fine if the spike during is less than 10ns, but SW pin will be damaged because it is already over 9V max. no matter the

          duration is still < 10ns.

    Please help to provide your comment, thanks.

  • Hi Vincent,

    An increase in Vin will eventually result in a higher SW pin spike that can potentially exceed the absolute max rating of 9V and 10ns. Hence, I agree with your assessment.

    If an overvoltage occurs in the VIN (> 9V) of TLV62569, it will damage the weakest part of the IC. And it is also possible that a part of the low-side FET circuit being biased by Vin failed first, causing the SW and GND pins to short. Therefore, the failure mechanism does not show shorted Vin and GND pins in some parts.

    I hope I was able to answer your queries. If you need additional info from our side, you can email Ian, and we can discuss it directly via email.

    Best regards,

    Excel

  • Hi Vincent,

    Do you still have further questions about this topic? We can arrange a concall if needed.

    If not, kindly click the "Resolved" button to close the thread.

    Thank you.

    Best regards,

    Excel