• State Resolved
  • Locked Locked
  • Replies 15 replies
  • Subscribers 23 subscribers
  • Views 1529 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

TPS9260x-Q1 / RC fillter to ISP, ISN

Ryuji
Guru 20090 points

Hello,

Our customer is testing the operation of instantaneous interruption of VBAT by TPS9260x EVM.
As a result, short failure was occured by instantaneous interruption.
Thus, I would like to add the RC filter to ISP and ISN.

I found that the pad of RC filter(R39, R38, C24, C7) in the EVM schematics.
However, I couldn't found more information about RC filter of RSNS in datasheet and User's guide.
Could you please let me know the method of  RC filter determination and equation ?
I think that this RC circuit will influence loop stability.

Please see the attached file as instantaneous interruption waveform and RC filter schematics of User's guide.
1738.20141125_TPS9260x Filter.pdf

This post is regarding the following post.
http://e2e.ti.com/support/applications/automotive/f/26/t/383581.aspx

Best Regards,
Ryuji Asaka

  • 0
    TI__Expert 5380 points

    Ryuji-san,

    We are working this case now. Maybe this need do some test in the lab, we will give your answer within a couple of days.

  • 0 in reply to Anda Zhang
    Guru 20090 points
    Anda san,

    Thank you for the reply.
    I understood the situation.
    I'm looking forward to your reply.

    Best Regards,
    Ryuji Asaka
  • 0 in reply to Ryuji
    TI__Expert 5380 points

    Ryuji-san,

    Sorry for replying so late. I was full scheduled these days.

    Because the bandwidth of the circuit is  20kHz, so the RC frequency should be far away from the cross frequency 20kHz. And because R will impact the current accuracy, customer can use a small R (several ohm) and small C (max is 470nF) to filter the noise, make sure the cutoff frequency of the RC is far bigger than 20kHz, but less than the frequency of the noise.

    Also you can try to increase the output capacitor or decrease the slew rate of the input supply. 

  • 0 in reply to Anda Zhang
    Guru 20090 points

    Hello Anda san,

    Thank you for the reply.
    Could you please let me know the recommended cutoff frequency ?

    I think that the our customer noise frequency by instantaneous interruption is about 10kHz or less.
    Thus, the RC filter may be not effective for this issue...

    Also , I think that we can't change the slew rate of the input supply since slew rate might be decided by test conditions of Enduser. 

    Best Regards,
    Ryuji Asaka

  • 0 in reply to Ryuji
    TI__Expert 5380 points

    Ryuji-san,

    You can try a 60kHz frequency. 

    Do you know the test condition of the customer? What is the frequency of the instantaneous interruption? What is the purpose of the test? I want to do some test in the lab.

  • 0 in reply to Anda Zhang
    Guru 20090 points
    Hello Anda san,

    Thank you for the reply.
    I'm requesting the test condition... However, I couldn't get yet.
    I will get back to you after get it.

    Best Regards,
    Ryuji Asaka
  • 0 in reply to Ryuji
    Guru 20090 points

    Hello Anda san,

    Current conditions are the following.

    -Vin: 12V
    -LED: 10pcs series, Vf=3V
    -Iled:500mA
    -Fsw:600kHz
    -OEM requirement is that there is no failure on 2ms of instantaneous interruption.

    I proposed adding the snubber circuit (4.7ohm+10nF) to our customer.
    I found that snubber layout in the EVM.

    Best Regards,
    Ryuji Asaka


  • 0 in reply to Ryuji
    Guru 20090 points

    Hello,

    Short circuit detection by instantaneous interruption is not improved by adding the snubber , adding  the Cout(10uF), and adding the RC filter.
    Our customer use programmable power supply for VIN.
    Then, there is no short circuit detection by 3ms of instantaneous interruption.
    Please see the attached file as waveforms.
    5430.20141206_TPS92601-Q1.pdf

    (1) Could you please advice other  improvement plan?
          For example, adding the reset ic or FET etc...
    (2) Could you please check the reproducibility by EVM ?

    Best Regards,
    Ryuji Asaka
          

     

  • 0 in reply to Ryuji
    TI__Expert 5380 points
    Ryuji-san,

    It can be reproduced on our EVM. Actually it is overcurrent protection rather than short circuit protection. If customer only wants to pass the 2ms instantaneous interruption, customer can reduce the capacitor value on Vin C21 to 2.2uF. By reducing the capacitor, TPS92602 will shut down during the 2ms power off time, because there is a softstart function, so there will be no big inrush current when restarting the circuit. Previously the C21 is 10uF, there is still big voltage on the Vin pin during the power off time, the device is still active, when restarting the EVM there will be no softstart function avaiable, so there is a big inrush current during the startup which will trigger the overcurrent.
  • 0 in reply to Anda Zhang
    Guru 20090 points

    Hello Anda san,

    Thank you for your reply.
    Can we avoid the over current/short circuit protection by reducing C21 capacitor ?
    Since C21 is connected to Vcc pin for internal LDO bypass capacitor, I think that we would reduce C30~33 of 3.3uF.

    Best Regards,
    Ryuji Asaka

  • 0 in reply to Ryuji
    TI__Expert 5380 points
    Ryuji san,

    It is my typo, it is true reducing C21 will help , reducing C30-33 helps little, because the voltage on the Vin will follow the voltage on Vcc when the internal LDO is in drop out region. Reducing the C21 on Vcc pin will speed the voltage drop on Vin.
  • 0 in reply to Anda Zhang
    Guru 20090 points
    Anda san,Thank you for the reply. I understood.
    I will confirm the effective of reducing the C21.Best Regards,Ryuji Asaka
  • 0 in reply to Ryuji
    Guru 20090 points

    Hello Anda san,

    We could prevent the overcurrent protection by redicing the C21.
    Thank you for your support.

    Then, I couldn't find the method of determination and recommended value of C21.
    Could you please let me know the method of determination of C21 ? (we should consider noise frequency ?)

    Best Regards,
    Ryuji Asaka

  • 0 in reply to Ryuji
    TI__Expert 5380 points
    Ryuji-san,

    The value of the cap can be found in the datasheet, 2.2uF - 20uF. Usually a capacitor with min value of 2.2uF is recommended to put on the Vcc pin to keep the Vcc voltage stable. As Vcc is the power supply for the internal control circuit and low side mosfet driver, if the capacitor is too small, Vcc will fluctuate a lot during operation.
  • 0 in reply to Anda Zhang
    Guru 20090 points

    Hello Anda san,

    Thank you for the reply. I understood.
    And sorry for I overlooked the C(vcc) in the EC table.

    Best Regards,
    Ryuji Asaka