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.

TPS54320: Power unstable issue

Part Number: TPS54320

Hi,

We are using TPS54320 to step down a 12V input to 1.8V with a 1A nominal load.

Both TI WEBENCH® Designer and our loop test result show we have enough Phase and Gain margin.

While the 1.8V will randomly power off, it seems to be very sensitive, could you help check our design to see what could be the root-cause?  

Here is the schematic and power loop test result.

Here is the waveform when unexpected power off happen, channel 1(yellow) is 12V input, channel 2(purple) is 1.8V output, channel 3(blue) is Enable control, channel 4(green) is SW waveform.

When power down happen, we didn't any overvoltage or undervoltage in the output.

After power off, TPS54320 try to power up for several times, but still failed with following waveform.

  • Thanks for your question, and detail on the issue. It looks like there is a lot of noise on the EN input. Can you verify that with a solid EN the issue is still there? Also, can you verify the load conditions where this happens (light load or full load or in between)?  What were the conditions for the Bode Plot you provide?

    I will investigate and provide you with further inputs tomorrow.

  • After looking at the compensation on your design, as well as the Bode plot, it does not seem likely to be stable.  Without seeing the SW waverform zoomed in, I am not able to tell if that is the case.  However, even if it is not stable, I would not expect behavior like you see. The loss of the output, and the slow start-up and unstable output after start-up is not typical of the device. Please look at the design for a 12Vin to 1.8Vout design at https://webench.ti.com/appinfo/webench/scripts/SDP.cgi?ID=22B3E3B43613FDB9, and see the compensation chosen there. It is for a similar switching frequency and slightly more output cap than you have in your design.

  • Hi Steve,

    Thanks for your quickly support.

    The EN noise comes from SW, after we have better Probe grounding, the noise disappeared, 

    We have measured the inductor current, it is lower than 1A for normal condition. Since this issue didn't happen again until now, we can't provide the inductor current during power off.

    The Bode Plot I provide is 0.7A output current.

    thanks.

  • Hi Steve,

    Do you see any potential risk for our compensation design and Bode Plot? Could you help explain more detail?

    Except 3*22uF output capacitor, we still have more than 200uF capacitor in the end device.

    Not sure if they will affect the loop compensation?

    thanks.

  • Thanks for the additional information. Are you saying you have not seen the issue again?

    With additional output capacitance, the crossover frequency will be lower with the same compensation.  It should also improve your phase margin.  I recommend that you re-simulate in Webench with the total output capacitance you have in your system and re-optimize the compensation for that condition.

  • Hi,

    thanks for the support.

    It seems our issue was caused by poor soldering of compensate components, since  few components are manually soldered when we do hardware basic test.

    after re-soldering these components, the power off issue didn't happen again.

    thanks

  • Hi,

    That is great news, thanks for sharing the update.