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.

UCC28C43: Minimum Sync Pulse Width

Part Number: UCC28C43

What is the minimum required sync pulse width and amplitude for synchronization on the UCC28C43D pulse width modulator?

Thanks for your help with this!

Richard Elmquist

  • Richard,

    Unitrode application note U-111 has a detailed section III about synchronisation using the RtCt pin, which is applicable to UCC28C4x devices, among others.

    www.ti.com/.../slua110

    The sync pulse amplitude depends on the programmed free-running frequency, which will dictate how much extra voltage needs to be added to the ramp at the RtCt pin to cross the upper discharge threshold. Since the RtCt ramp amplitude is specified to be 1.9 Vpp, an sync pulse of similar amplitude should always ensure synchronisation - although a lower amplitude will work too, depending on the free-running frequency. From U-111, if the free-running freq is set to say 85% of the sync frequency, a sync pulse amplitude of 15% of the 1.9-V pk-pk oscillator ramp is required.

    You should ensure that the added sync pulse still keeps the RtCt pin max level below its abs max rating.


    As for the sync pulse width, a larger pulse width will result in extended dead-time - the dead-time will be the sync pulse width plus the time taken to discharge the RtCt cap - which will be slightly shorter than the free-running dead-time since the RtCt cap is being discharged from a slightly lower level as a result of the sync pulse.

    It makes sense to keep the sync pulse width short, unless longer dead-time is required or desirable to limit Dmax.

    The minimum sync pulse width is difficult to quantify, but per U-111 should be at least equal to the lost-dead-time through the addition of the sync pulse. Per U-111, a sync pulse width of 50 ns minimum works well.

    I hope this helps to answer your questions.

    Thanks,
    Bernard
  • Bernard,
    Thanks for your help!
    I will let you know if he customer has any further questions.
    Thanks,
    Richard Elmquist
  • Bernard,

    The customer has come back with additional questions. I am not sure that you will be able to answer them,. but here they are:

    For this application of the UCC28C43D we are investigating potential noise pulses levels retriggering the sync.

    We would like to know how much design margin we have to the 1.6Vpp level and the 50 nSec width.

    Are these the absolute numbers or would you recommend some % of design margin to ensure not re-triggering the sync due to noise?

     What is the process tolerance for this voltage level and pulse width in the PWM chip?

    Would it be likely to change from lot to lot of parts?

    We are just trying to quantify the absolute requirements for voltage and pulse width that will trip the pin3  CS and re-clock the pin 4 RT/CT .

    Through the use of filtering we are alleviating the noise but we need to know how much margin we can live with based on the absolutes for those pins.

    Can you comment on any of these questions?

    Thanks for your help with this!

    Richard Elmquist

  • Bernard,
    Have you had a chance to look over the further questions posed by the customer?
    I understand if we do not have a definitive answer to these questions.
    Just let me know so that I can respond back to the customer.
    Thanks for your help with this!
    Richard Elmquist
  • Richard,

    Sorry I should have updated you yesterday.

    I have passed this thread along to one of my colleagues inside the product line to get more information, since it will likely need inputs from the IC designers.

    You should get a more detailed response from them.


    Thanks,
    Bernard
  • Bernard,
    Thanks for your help!
    Have a wonderful holiday!
    Richard Elmquist
  • Bernard,
    Can you send me the email for the person that will be handling this?
    I am going to be out of the office starting Wednesday and I was haping to get an answer before I left for the holidays.
    Thanks for your help with this!
    Richard Elmquist
  • Richard

    I have asked the product line team to contact you directly.

    Thanks,
    Bernard