• State TI Thinks Resolved
  • Locked Locked
  • Replies 1 reply
  • Answers 1 answer
  • Subscribers 39 subscribers
  • Views 345 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.

PMP10143: Recommendation for over 20KHz

May
Genius 9880 points
Part Number: PMP10143
Other Parts Discussed in Thread: UCC28880, UCC28911, UCC28910

Hi,

Customer is working on a design based on PMP10143, according to them it is working very well. 

However,their load is only 250mA on 3.3VDC causing a Fsw of about 4.5kHz which is audible. By modifying C9 for 10nF and R5 for 4.7K, they've been able to raise the Fsw to 9.3kHz which is better.

Now they wondering if there is a way to raise it more without too much modifications, ideally over 20kHz. Do you have an almost drop-in replacement controller which have a fixed frequency but adjustable PWM?

Regards,
Maynard

  • 0
    TI__Expert 5975 points

    Hi Maynard,

    The controller UCC28880 works in hysteretic mode, therefore it's comparing the voltage on pin FB with programmed thresholds.

    That means at light load it will anyway enter audio range. Nevertheless, it can enter audio range in two different ways: burst mode or continuous switching.

    These two modes can be selected by changing the amount of the ripple voltage on capacitor C6, and at the same time removing the pole in the compensation network R6 / C11 by shorting R6. If you short R6, you need to increase C9 in order to keep the power supply stable (I suggest to check out the stability, after each modification, by analyzing the bode plot). The more the ripple on C6, the more continuous the converter switches (and more audible at light load but not at full load). In your case, I suggest to remove C9 and increase C11 but at the same time check out the stability. The advantage of removing C9 is that the switching frequency ripple on C6 will not influence the loop compensation, and therefore the converter should keep working in burst mode. The advantage of burst mode is that in each burst the switching frequency is high (typical is 50 KHz) and the burst frequency is pretty low (hundred Hz, therefore not much audible). 

    We have two other HV FET integrated IC: UCC28910 and UCC28911; they have the advantage of reducing the peak current at light load when they enter audio band, therefore the noise is not audible. BTW these ICs are not pin to pin compatible with UCC28880.

    Another way to avoid audio noise is to varnish / glue / pot the transformer, to avoid vibration. Furthermore you may want to remove the ceramic capacitors C7 and C2 and use only electrolytic ones, to avoid piezoelectric effect, which become loudspeakers in audio range.

    Best regards,

    Roberto