• State TI Thinks Resolved
  • Locked Locked
  • Replies 14 replies
  • Answers 2 answers
  • Subscribers 64 subscribers
  • Views 587 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.

LMZ31710: Delayed start when cold

phil chabot
Prodigy 110 points
Part Number: LMZ31710

I have a LMZ31710 in basic configuration (default SS, no tracking), 12V input , 1V output (around 1A output).

After sitting overnight, it can take 20s or so for the device to start on power up. after that, it starts OK with some hesitation(1second). It used to work OK.  (I have a separate board with same LMZ31710  that does not exhibit that issue)

What troubleshooting steps can be done ? Is the device dammaged ?

thanks

  • 0
    TI__Guru 50120 points

    Hi Phil,

    Is this behavior on the EVM or your board. If it is your board please provide the schematic for me to review. The start-up waveform can be seen in Figure 24 of the datasheet in section 7.3.10. Channel 1 shows Vout transition from 0V to 1.8V in roughly 2ms (1 division). A normal operating device should not take 20s to start up. This abnormal operation may be a hint that the part is possibly damaged.

    1. What application does this LMZ31710 device go into?
    2. Can you also provide a scope probe of VIN,VOUT, VSS, and PH during startup. I want to see what the phase node (switching) looks like during startup.

    Regards,

    Jimmy

  • 0 in reply to Jimmy Hua
    TI__Guru 50120 points

    Hi Phil,

    Can I get an update on this? I haven't heard back from you for a while so I assume you were able to resolve this issue on your end. If you still have remaining questions, please reply back to this message and it will reopen this thread for further discussion.

    Regards,

    Jimmy

  • 0 in reply to Jimmy Hua
    Prodigy 110 points

    Hi Jimmy,

    My apologies for the delay. Here is the schematic. It is our own board. Power is supplied to an FPGA board. I dont have the signal plots yet, but it does start OK with a 2ms or so ramp time. It just starts after 20s or so. we are looking at U22 which provides 1.0V from 12V.

    thanks

    Phil

  • 0 in reply to phil chabot
    TI__Guru 50120 points

    Hi Phil,

    1. Do the other DC/DC (U23,24,25) start up instantly or does it have the same delayed startup issue that U22 experiences?
    2. Do you have any additional output capacitance downstream to the +1V0 rail? Too much output capacitance may result in the part hitting current limit / hiccup mode and prevent proper start up operation because of output inrush current. It may be necessary to increase the SS time.
    3. If you can provide scope plots, this will help with the debugging process.

    Regards,

    Jimmy

  • 0 in reply to Jimmy Hua
    TI__Guru 50120 points

    Hi Phil,

    Can I get an update on this? Were you able to get the scope shots?

    Regards,

    Jimmy

  • 0 in reply to Jimmy Hua
    Prodigy 110 points

    Hi Jimmy,

    I was not able to get some scope shots since I needed to board to stay on for board testing, but I have some news.

    After being left unpowered for a few days, it now seems that the LMZ31710 won't start at all UNLESS I connect the INH pin to 5V, if I disconnect the 5V to the INH the converter shuts off.

    Any ideas?

    thanks

    Phil

  • 0 in reply to phil chabot
    TI__Guru 50120 points

    Hi Phil,

    I just want to clarify that you are only having issues on U22 and not the other rails.

    The INH high threshold is a minimum 1.3V. There is an internal pull up that will enable the part if you leave Pin 30 floating. If 5V is required on the INH pin for the part to start converting, it sounds like the part may have some damage.

    If this is the case, I would suggest swapping the old U22 part out for a new LMZ31710 and retest operational stability.

    Regards,

    Jimmy

  • 0 in reply to Jimmy Hua
    Prodigy 110 points

    Hi Jimmy, thanks for the reply.

    Yes the issue in on U22 only and one board only (one other board works just fine).

    What would cause the failure? I'm a bit reluctant to go for a small production batch with that issue on mind.

    Do you have any data on the reliability of that part ?

    thanks

    Phil

  • 0 in reply to phil chabot
    TI__Guru 50120 points

    Hi Phil,

    Usually for delayed startup issues, it is either the input of the power supply or an excessive amount of output capacitance. Are you operating at near maximum output capacitance? I don't think this is the case since I only see 3 x 100uF + 200uF which comes out to 520uF, unless you have more capacitance on the 1V output rail that connects to the FPGA.

    If you have a PCB layout that you can attach, I can help review it for any potential issues. For the reliability data, I will have to contact the customer quality engineer to see if he has any data on the LMZ31710. When I get more information, I'll get back to you on this.

    Regards,

    Jimmy

  • 0 in reply to Jimmy Hua
    TI__Guru 50120 points

    Hi Phil,

    I've talked to the CQE for this device. Worldwide we are seeing that this device is extremely robust and has no indication of a quality or reliability issue. If you try replacing it with a new device will the failure mode go away?

    Any updates on the total output capacitance on the 1V rail? Can you share the PCB layout?

    Regards,

    Jimmy

  • 0 in reply to Jimmy Hua
    Prodigy 110 points

    Hi Jimmy,

    There is about 1300uF of total output capacitance on the 1V rail. The PCB layout is essentially the one on the datasheet/evaluation module.

    So far we can use the board by connecting the INH input to a 5V supply through a 100Ohm resistor. We will replace the module eventually, but need the board for testing purposes.

    Thanks

    Phil

  • 0 in reply to phil chabot
    TI__Guru 50120 points

    Hi Phil,

    Had the part still been working but only with a slight startup delay, I would have recommended that you increase your soft-start time to avoid any potential issue with the part going into current limit.

    The INH circuit has an internal pull up current source. This should allow the part to be enabled even if INH pin is left floating. However, your description of the device's condition leads me to believe that there may have been some internal damage in the part. Initially  it would start up properly after 20s but now it sounds like the part is no longer operating unless powered with 5V INH. Can you confirm proper switching and operation of the part in your current state? You can check this by probing the PH pin and observe the switch node. If it is operating properly I would expect a nice and clean square wave.

    I would still recommend you replace the module since this issue with the INH pin does not sound within proper operations of the datasheet spec.

    Regards,

    Jimmy

  • 0 in reply to Jimmy Hua
    Prodigy 110 points

    Jimmy,

    PH pin is looking good. output voltage is nice and clean. except for not starting, the converter operates normally.

    We will replace the converter and have a few more boards made to see if we just had a bad converter.

    thanks

    Phil

  • 0 in reply to phil chabot
    TI__Guru 50120 points

    Phil,

    Thank you for the update. Please keep me informed on whether this is still an issue with a new converter on your PCB.

    Regards,

    Jimmy