• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Subscribers 63 subscribers
  • Views 588 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.

TPS565201: Output bulk capacitance

Brian Rupp
Prodigy 50 points
Part Number: TPS565201
Other Parts Discussed in Thread: TPS54528

Hello,

I am modifying an existing design to incorporate a switching regulator instead of a linear. The existing design has a number of large electrolytic capacitors distributed across the board (almost 200000uF in total). I am trying to be as minimally invasive as possible with this design in order to avoid causing any unexpected issues, but I would be surprised if this amount of capacitance didn't lead to instability on the TPS565201.

The datasheet states that "The inductor and capacitor for the output filter must be selected so that the double pole of Equation 3 is located below the high frequency zero but close enough that the phase boost provided be the high frequency zero provides adequate phase margin for a stable circuit." It then suggests a maximum of 68 uF, which obviously I am well above. Is there a way to prevent this extra capacitance from affecting the regulator (perhaps an output LC filter)?

Additionally, I am concerned that this amount of capacitance will trip the short circuit detection on the regulator. The datasheet does not provide a maximum load capacitance to avoid this either. Presumably an LC filter would help here as well, but I am concerned this may cause other issues.

Thanks!

  • 0
    TI__Expert 4720 points
    Hi Brian,
    Yes we suggest to use the "Recommended Component Values" in Table 2. Please note the output cap is suggested to use ceramic cap. May I know why do you need such a large capacitance?
  • 0 in reply to Neal Zhang
    Prodigy 50 points

    Unfortunately the engineer who made that decision is no longer available so I am not sure why so much capacitance was used. However, the board is quite large and has numerous high current switching loads, so presumably these are used as local storage to avoid large voltage drops across the board.

    If possible I would like to avoid removing them as I don't know the repercussions of this change and don't really have the capacity to test it at this time, so I was hoping there was some way to decouple the regulator output from the rest of the circuit (LC filter).

  • 0 in reply to Brian Rupp
    TI__Expert 4720 points
    Hi Brian,
    The hiccup time of TPS565201 is 1.8ms. It is quite a small value. At startup, if the output voltage is smaller than the soft start voltage or Vfb for 1.8ms, the divice will enter short circuit protection mode. So I'm afraid it can't be used for such a big output cap.
    However, you can put the large cap at the input of TPS565201. It will have the same function as local storage.
  • 0 in reply to Neal Zhang
    Prodigy 50 points

    Hi Neal,

    I see that hiccup value now, that definitely would be much too short in this instance. If I wanted to avoid altering the existing board, could I perhaps use something like the TPS54528 which has an adjustable soft start?

    Thanks for your help,

    -Brian

  • 0 in reply to Brian Rupp
    TI__Expert 4720 points
    Hi Brian,
    Yes TPS54528 can solve the short circuit protection problem. You can add a bead between the ceramic output cap and the large electrolytic capacitors, to avoid the large electrolytic capacitors' effect on loop response.
  • 0 in reply to Neal Zhang
    Prodigy 50 points

    Great, thanks! Would a power line bead (such as a Murata BLM18KG260TN1D) be sufficient or would a larger choke be preferable?

  • 0 in reply to Brian Rupp
    TI__Expert 4720 points
    Hi Brian,
    I'm afraid the impedance of BLM18KG260TN1D is too small to avoid the large electrolytic capacitors' effect on loop response. The impedance of bead need to be larger than 1 Ω at the target bandwidth. For safety, please choose the impedance larger than 1 Ω at 50 kHz. Or you can choose a bead then measure the loop response from bench test.
  • 0 in reply to Neal Zhang
    Prodigy 50 points

    Got it, I'll grab something bigger then. Thanks again for all of your help!