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.

TPS3813I50DBV application question

Other Parts Discussed in Thread: TPS3813

Dear Sir

Good days,

Here have one question that when we add 470uF capacior in VCC5V_MAIN, then reset will have plus output.

may you pls advice us what parameter i loss?

 

                                                               Best regards

                                                                            Bogey

  • Hello,

    I would recommend increasing the pull up resistor to 200k. Could you please clarify what you mean by “reset is plus output”. Does this mean RESET stays high? The waveform seems to show RESET going high and low. Is this waveform showing the failure? Could you also describe the effect on WDI?

     

    Best Regards,

    David

  • Dear David

     

    Pls refer as  below normal waveform.

     

    After VCC5V_MAIN + 470uF capacitor. thr Reset will keep having plus wavefor,.

    And another question is, when Vcc rising, RST have small spike(use red mark). is this symptom normal?

                                                                           Best regards,

                                                                                              Bogey

  • Hello,

     

    Increasing the pull up resistor to 200k should solve the problem. The small spike on RST as Vcc is rising is normal. This occurs before the pass element turns on.

     

    Best Regards,

     

    David

  • Dear David

    Good days,

    May we check with you what relation between pull up resistor and Vcc capacitor?

    and  as below circuit, we change R102 to 100Kohm, R382 to 1Mohm, add 470uF capacitor in Vcc5V_Main.

    sometimes affter Vcc5V_Main on but RSL pin will keep low....May you pls have some suggestion to us?

                                            Best regards,

                                                        Bogey

  • Hi Bogey,

    The Vdd capacitor is meant only to ensure best stability of the threshold voltage.  We recommend a 0.1uF capacitor.  Is there a reason you are using such a big capacitor?

    Also can you share the purpose of R382?

    Very Respectfully,

    Ryan

  • Hi Ryan

    Good days,

    Pls refer as below:

    1. why need add 470uf capacitor in Vdd?

     

    For reduce power ripple and stable power voltage .

     

    2. Why add resistor R382?

     

    For Vgs signal clear. It function like tps3813I50 R2 (no floating) .

                                                            Best regards,

                                                                         Bogey

  • Hi Bogey,

    It looks in your last waveform that WDI is not pulsing.  This will cause the device to time out.  Is this correct?

    Very Respectfully,

    Ryan

  • Hi Ryan

    May we check with you, if we have no need 5V monitor(due to add 470uF is to keep 5V voltage)request. just need watch dog function. we want change to 3.3V(TPS3813K33). is other issue we need notice?

                                                                            Best reageds,

                                                                                           Bogey

     

  • Hi Bogey,

    TPS3813K33 is in the same family and will operate like the TPS3813I50.  The difference will of course be the threshold.  The watchdog functionality will operate the same.

    Very Respectfully,

    Ryan

  • Hi Ryan

    Good days,

    Enclosed files are TPS3813K33 circuit and measured waveform(Ch2 is VCC; Ch1 is RST Pin)

    Sometimes RST will have clock output when power on.is our design have some problem?

                                                            Best regards,

                                                                                 Bogey

  • Hi Bogey,

    To answer the much earlier question about the pull-up resistor value:

    The pull-up resistor cannot be too small as the RST/ pin can sink only a certain amount of current (around 4mA in your case for a logic low of 0.4V). A 1kOhm pull-up means your minimum output logic level would be 460mV if you didn't have the 10k resistor, as you are not able to pull enough current through the RST/ pin to drive RST/ low. The following application note explains a bit more on this:

    http://www.ti.com/lit/an/slva485/slva485.pdf

    Onto your problem:

    Your signal on RESET/ is at 1kHz. This does not fall within any of the timing specifications for the TPS3813 with your current configuration, so I expect the signal is coming from somewhere else. Do you have anything operating at 1kHz?

    One other thing - can you remove R382 and make R102 a 10kOhm resistor? This will make the driving requirements for the RESET/ circuit a bit easier, and could help solve the problem.

    Regards,

    David

  • Hi David

    many thanks for your reply, it is useful for us......

    and due to my customer already P.P 1K pcs, and just find 2 pcs fail.

    may you advise us why ?

     

                                               Best regards,

                                                              Bogey

  • Hi Bogey,

    As this seems to be an isolated failure, I would suggest a quick check by doing an A-B-A swap. Remove the failing unit from the failing board and put the failing unit onto a known good board. This will tell you if the part carries the failure - than take the known good unit from the good board and put it on the bad board to see if the board is causing the failure. Then put the bad unit back on the bad board to make sure the problem is still there.


    Regards,

    David

  • Hi David

    We have do ABA swap and symptom follow failure IC. May you suggest us to do RMA process?

                                                             Best regards,

                                                                            Bogey

     

     

  • Hi Bogey,

    Could you answer the following?

    Your signal on RESET/ is at 1kHz. This does not fall within any of the timing specifications for the TPS3813 with your current configuration, so I expect the signal is coming from somewhere else. Do you have anything operating at 1kHz?

    One other thing - can you remove R382 and make R102 a 10kOhm resistor? This will make the driving requirements for the RESET/ circuit a bit easier, and could help solve the problem.

    Regards,

    David

  • Hi David

    Good days,

    we check customer, here have no 1KHz noise. icheck DS, here seems no explain how to calculate pull up Resistor way. we have remove R382 and make R102 a 10kOhm resistor. the issue will solve.

    But for customer original design, they have run 1K  pcs, but just 2 pcs fail...

    So we are not sure it is ic issue or design issue.

                                                                 Best regards,

                                                                                  Bogey

  • Hi Bogey,

    Not sure if I understand you fully - did changing R102 and removing R382 fix the problem?

    If that is the case, the reason for the limitation on the pull-up resistor is the drive strength of the output device for RESET. As per the EC table, with VDD at 6V the device can pull down reset with at most 4mA of current to 0.4V. This equates to a minimum resistor to Vdd of 1.5kohm, but to be on the safe side you should try and avoid this limit by an order of magnitude, hence the 10kOhm.

    Regards,

    David

  • Hi David

    Not sure if I understand you fully - did changing R102 and removing R382 fix the problem?

    The answer is yes,

    TPS3813I50DBV DS page 3, if Iol over 4mA, then what symptom will happen?

                                                    Best regards,

                                                                          Bogey

     

     

  • Bogey,


    I think that the part is detecting that too much current is going through the output device to try and bring down the output, so the part shuts down the output to protect the output device and tries again.

    Regards,

    David