• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 62 subscribers
  • Views 391 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.

TPS548A20: Only stable with MLCC?

SimonF
Expert 3410 points
Part Number: TPS548A20


Hi Team,

the TPS548A20 has description "Supports All Ceramic Output Capacitors" and there is a note in datasheet (page 22):

"TI recommends only using ceramic capacitors for Auto-skip mode."

http://www.ti.com/product/TPS548A20

So does it mean electrolytic capacitors are not recommended?
If I use POSCAP as output capacitors, what does it mean for controller's function? Can I expect for example a worse performance?
Is DCAP3 control suitable for ceramic capacitors only in general?

Best Regards,

Simon

  • 0
    TI__Guru**** 191445 points

    Simon,

    Use of ceramic output capacitors is a key feature for DCAP3 control.  DCAP control family is adaptive on time and hysteretic off time.  It requires a ripple component at the FB node for the hysteretic detection.  DCAP3 uses internal ripple injection to supply the required ripple rather than inductor ac current times output capacitor ESR, so low ESR (MLCC) capacitors can be used.  You may possibly be able to use certain low ESR polymer types provided the ESR zero is well above the crossover frequency.  For electrolytic capacitors, you should look at regular DCAP control.

  • 0 in reply to JohnTucker
    TI__Expert 3410 points
    Hi John,

    thank you for your help!
    I'd like to use 2x Panasonic EEFCX0J151R + 1x MLCC 22uF as output capacitors.

    The Panasonic capacitor has ESR 15m ohm.

    industrial.panasonic.com/.../EEFCX0J151R

    Can I use this capacitor configuration with the DCAP3?

    Vin = 12V, Vout=3.3V, Iout = 10A, Fsw = 500kHz min (most probably 500kHz or 600kHz).

    Best regards
    Simon
  • 0 in reply to SimonF
    TI__Guru**** 191445 points
    Simon,

    That gives an ESR zero at 70.7 kHz. That is probably high enough. Unfortunately, there are no average model tools available to simulate this either in pspice or Webench. The only way to verify is to modify an EVM and measure the loop. I think it has a high probability of working.