• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 63 subscribers
  • Views 388 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.

WEBENCH® Tools/TPS53353: default ceramic output capacitor option results in 0V output

Seth J
Prodigy 60 points
Part Number: TPS53353

Tool/software: WEBENCH® Design Tools

Please see https://webench.ti.com/appinfo/webench/scripts/SDP.cgi?ID=43656B8B11997FFF. This is the default 12Vin, 5Vout @ 20A TPS53353 WEBENCH configuration except with the ceramic output capacitor option chosen. The resulting Steady State simulation shows 0V average output. The electrolytic option correctly simulates 5V out. Is the TPS53353 capable of using ceramic output capacitors? The datasheet covers using low ESR ceramic output caps and WEBENCH correctly includes the additional ripple feedback components, but still fails to give the proper simulated output. Thanks.

  • 0
    TI__Mastermind 35165 points

    Hi Seth,

    Yes, the TPS53353 is capable to use all ceramic output capacitors. It looks like this design also has an issue with the startup simulation. I will need to look into this to figure out if it is related to the simulation model or with the component selection when ceramic output capacitors are selected.

  • 0 in reply to Anthony F
    TI__Mastermind 35165 points

    Hi Seth,

    I am still looking into this with the WEBENCH team. I think this is tied to this design having a very small amount of ceramic capacitors and large external ripple injection amplitude. In the meantime, would you like me to recommend component values to use for this application?

  • 0 in reply to Anthony F
    Prodigy 60 points

    Hi Anthony,

    Thanks for looking into this for me. I was searching for the suggested ripple feedback network component values from the simulator, so while it's being looked at it would be nice if you could make those recommendations. The output of this stage will feed into another set of lower-voltage buck converters so accuracy and ripple aren't of as much concern for this particular application. I'd like to use a 820nF, 24A max inductor as I'm already using one elsewhere. Thanks.

  • 0 in reply to Seth J
    TI__Mastermind 35165 points

    Hi Seth,

    I recommend using at least ~150 µF total capacitance on the output. With the 5V output it may take up to 5x 47µF caps to meet this due to dc bias derating. In this application it sounds like you will have some additional output capacitance due to the input caps of the lower-voltage buck converters. This should be included as part of the total output capacitance.

    For the ripple injection network, I suggest 3.9k, 22nF across the L and 1nF to the VFB pin. For the VFB divider I suggest 68.1k top and 10k on the bottom.

    If you have more capacitance than the ~150 µF I recommended, we could tune this ripple injection circuit a bit more to reduce the amount of ripple injected if you'd like. Just let me know the total amount of capacitance you plan to have on the 5V.

  • 0 in reply to Anthony F
    Prodigy 60 points

    Hi Anthony,

    Thank you very much for your suggestions. I was planning on having a combined total of 132uF (6x 22uF) input capacitance for the downstream converters, and 3x 47uF caps on the output of the 5V converter. However the downstream converters won't necessarily all be placed in the same area so some impedance may be added from that longer path, and I'd like to reduce the 5V converter's reliance on those input caps accordingly. I can add two more 47uF caps to the 5V output to bring the local total up to your recommended minimum. How much could the suggested ripple feedback component values change by adjusting the output capacitance? I'm not as worried about increased ripple on the output for this design since it will only be used to feed lower voltage converters. I just want to make sure I meet the minimum requirement for ripple feedback for the converter to operate properly. Is there an upper bound to the amount of ripple feedback or output capacitance before the circuit fails to operate? 

  • 0 in reply to Seth J
    TI__Mastermind 35165 points

    Hi Seth,

    This morning I went through and tried to make a more organized step by step process to help with this. See image below and attached excel sheet in case you want to try some different things yourself.

    The procedure I have made limits the injected ripple to a max of 10% the reference (60 mV) to give some margin to OVP. If you use more capacitance than the minimum, the ripple injection resistor can be increased to reduce the injected ripple. Reducing the injected ripple is not required, but reducing it will improve transient response and lower the offset it adds to the DC output voltage. When injecting ripple into FB, the FB divider needs to be adjusted to account for the injected ripple.

    If you look at the excel, something you may notice is I used an N value of 3 instead of 4 recommended in the datasheet. It is ok to do this as long as you are accounting for the dc bias derating of the caps. The recommendation of using 4 in the datasheet assumes you are not.

    I am going to work with the WEBENCH team to add this procedure into WEBENCH as well.

    The additional 132 uF is not too significant to have a big enough impact on stability. There would need to be at least 10x or more Cout before seeing a significant impact.

    I think I've covered everything but let me know if you have any more questions.

    /cfs-file/__key/communityserver-discussions-components-files/196/Cout-and-Ripple-Injection-calculations-Suggestions-with-820nH.xlsx

    Anthony