• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 229 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.

TPS53219A: switch phase node issue

Tommy Tzeng
Genius 4805 points
Part Number: TPS53219A

Hi Team,

currently testing 0A~12A,phase node waveform is ok, but at 13A~17A began to appear double plus(switch phase node) and jitter state, may i know what kind the possible cause?

please kindly refer the attachment the schematic and waveform test.SCHEMATIC1 _ 077_+V5A_+V5S.pdf

RX621C_+V5A.docx

  • 0
    TI__Genius 16760 points

    We are reviewing and will get back shortly.

  • 0
    TI__Mastermind 48200 points

     

    The TPS53219A uses TI's D-CAP mode control, which is derived from Constant On-Time (COT) control.  COT control  relies on the sensed output voltage (FB pin for the TPS53219A) rising during the On-time and falling during the Off-time.  Since output capacitors with very low ESR create a phase-shift between the On/Off time of the switching node and the rise and fall of the output voltage, they can require additional sensed output voltage ripple to stabilize the switching frequency and avoid the 2 switching frequency operation that you see in your waveforms.

    The switching frequency appears to be 750kHz.  At 750kHz, the 6.04k + 100nF ripple injection circuit adds 6.4mV of ripple at feedback.  Due to the 100pF C1351, this injected ripple represents only 1.6mOhms of ESR at the sensed output.  While that should be sufficient to stabilize a loop with 2x 560uF + 2x 10uF output capacitors, the low ripple injection amplitude will make it sensitive to noise.

    I would recommend:

    Decrease C1341 from 100nF to 47nF or 33nF to increase the injected ripple voltage and provide a more stable switching frequency.

    Since D-CAP mode control is a ripple valley regulation control, this increased ripple will increase the average feedback voltage, and thus output voltage in regulation.  With a 600mV reference voltage, changing C1341 to 47nF should increase the output voltage approximately 1% and decreasing C1341 to 33nF should increase the output voltage approximately 2%  To compensate for this change, you can update the value for R914 replacing the 0.6V reference voltage with 0.606 for a 47nF capacitor or 0.609 for a 33nF capacitor.

  • 0 in reply to Peter James Miller
    Genius 4805 points

    Hi Peter James miller,

    got it,thanks for your great support, but i have a question. but the output has use POSCAP also need to be application like DCR compensation?

    because R1519&C1340&C1340 has no componebt on the board, is option.

  • 0 in reply to Tommy Tzeng
    TI__Mastermind 48200 points

     

    It really depends on the phase-angle of the output voltage ripple, and the amplitude.  TI recommends a minimum of 10mV of ripple at the feedback node to minimize the jitter and maintain a stable switching frequency.  It's likely that under heavy load there just isn't enough ESR ripple on POSCaps that you have to overcome the switching noise, so it's triggering an On-time to increase the ramp amplitude.

    If the existing circuit is using POSCaps and is only going into this mode at 13A, try sizing the R-C for about 5mV pk to pk injection to add to the ESR ripple.

    Reduce the C to the output node of the inductor to 22nF and size R:

    Cc1 = (Vin-Vout) * (Vout/Vin) * 1/Fsw * 1 / (5mV x 22nF)

    That should get you 5mV of pk to pk ripple from the injected RC.  That will add to the ESR ripple at FB and help stabilize your switching frequency without slowing down the transient response.