• State Resolved
  • Locked Locked
  • Replies 1 reply
  • Subscribers 64 subscribers
  • Views 262 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.

TPS54623: Buffer in FB Loop

Russell Wetherley1
Expert 4255 points
Other Parts Discussed in Thread: TPS54623, OPA189, TINA-TI

Hello e2e.

I have a customer who wants to buffer the FB loop of the TPS54623.  The reason is to reduce current consumption on a highly sensitive / accurate current reading.  Would adding in a buffer like the OPA189 create any issues with phase margin?  This opamp (I think) has a small enough phase shift at low frequencies that the loop phase shift shouldn't go near 180 degrees.  Can you confirm if there would be any issues here?

The gain for the regulator feedback loop  is about 20 or about 26 dB. There is no GBW plot for the regulator amplifier but I’d guess no more than 50 kHz or so?

 The OPA189 is being run as a buffer, at 0 dB the phase shift is about -10 to -15 degrees from what I can see from the open-loop plot. I think it will be OK unless the TPS54623 is right on the edge of about -170 degrees of phase shift.

Thoughts?

Russell

  • 0
    TI__Mastermind 35135 points

    Hi Russell,

    Could you share a schematic (at least a simplified one) how the buffer is being inserted into the feedback loop? My guess it is between the output voltage and the FB divider so that the FB divider is not loading the output?

    My first thought is most likely it will not impact the loop enough to cause stability problems. Most likely the buffer should not start affecting the phase significantly until well above 1-MHz. I ran a quick TINA-TI sim with it configured as a buffer and the results below confirm this.

    To be sure, it should should still be possible to measure the bode plot with the op amp buffer added. Assuming my guess of the location above is right, the 10Ω resistor injection point should be between the output voltage and the non-inverting input of the op amp. If they don't have the right equipment to measure the bode plot, the load step response should also give a good judgement of the stability.