• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 64 subscribers
  • Views 261 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.

TPS53647: TPS53647 Bandwidth issue

Kao Mick
Intellectual 330 points
Part Number: TPS53647

HI SIR

We use TPS53647 to design 0.8V for LAN system.
The overshoot & undershoot requirement is 3%. In order to achieve this specification, we adjust the frequency response.
The following are the adjusted parameters and test results


Bode plot
phase margin: 62.7 degree
gain margin: 44.39dB (@ 997kHz)
bandwidth: 224kHz
Transient improvement was approximately 3%.
Transient meets specifications due to faster response, but is the bandwidth too high?
Gain margin cross 0db is 997kHz, is the frequency too high?

 


  • 0
    TI__Genius 11095 points

    Hello,

    I do not see an issue here. TPS53647 is a variable frequency controller. As the loop is perturbed the switching frequency modulates. So there is not a hard rule about having the crossover at 10-30% of the switching frequency. The gain margin of -13dB looks acceptable to me. 

  • 0 in reply to Matt Schurmann
    Intellectual 330 points

    We have cot mode buck ic from other brands.

    Their IC was suggested that the phase crossing frequency of 0 degree (Gain margin) should be lower than 1/2 of the switching frequency, and the output ripple will not oscillate.

    In this case, the frequency of TPS53647 Gain margin is 997khHz.

    We try to adjust the compensator, and reducing the bandwidth will not make the frequency of Gain margin lower than 1/2 of the switching frequency.

    Do we need to consider the high-frequency output ripple in this DCAP+ Buck IC?


     

  • 0 in reply to Kao Mick
    TI__Expert 5830 points

    Hi Kao,

    I'm not sure what's the frequency setting of the application here. The bandwidth now is ~184.9k should be <1/2 of effective switching frequency(2x switching frequency). Also, you should not see high-frequency output ripple by controller itself because the gain margin is 16dB enough.

    Thanks

    Chasel