• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 762 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.

Original question:

UCC24612: Is there any recommendation for CCM Flyback around 300KHz

UCC24612: The application for high frequency CCM Flyback

chentsu chung
Mastermind 6180 points
Part Number: UCC24612

Hi,

I know that UCC24612 can not support high frequency CCM Flyback, but this result from what parametr in datasheet?

Thank you.

C.T.

  • 0
    TI__Mastermind 40520 points
    Hello Chentsu,

    Thank you for your interest in the UCC24612 SR controller. There are two variants with different timings: UCC24612-1 and UCC24612-2.
    The -1 part is optimized for higher frequency power stages (up to ~1MHz), typically using GaN MOSFETs.
    The -2 part is optimized for lower frequency power stages (up to ~800kHz) typically using Si MOSFETs.

    These maximum frequency limitations are actually minimum period limitations. The minimum switching period at which each of the controllers can operate is summation of tdVGON, tON(min), tdVGOFF, and tOFF_ABSMIN. These are the gate turn-on propagation delay, the minimum SR conduction on-time, the gate turn-off propagation delay, and the minimum SR off-time.

    If you add up the datasheet values for these 4 parameters for each part variant, the result is the shortest switching period that the part is able to support. Some margin is added to that and the inverse gives you the maximum frequencies listed.
    It is not proper to simply add up the values in the MAX column of the specification table, because the timings are uncorrelated and they will not all be max at the same time. However, one can certainly use this worst-case calculation to provide a significant indication of margin to the actual switching frequency used.

    Does this answer your question?

    Regard,
    Ulrich
  • 0 in reply to Ulrich Goerke
    Mastermind 6180 points

    Hi Ulrich,

    Thanks for your explanation in detail. According to other discussion thread, TI doesn't recommend use UCC24612 in CCM Flyback that frequency > 200KHz, but good for DCM flyback in higher frequency.

    I think the frequency limitation 800kHz(1MKz) is enough for Flyback application, but why has the 200KHz frequency limitation in CCM? and result from what parameters in datasheet?

    Thank you.

    Best regards.

    C.T. 

  • 0 in reply to chentsu chung
    TI__Mastermind 40520 points

    Hi C.T.,

    I spoke with the Systems Engineer on this part and he said it is a complicated answer.

    For DCM the frequency limitations are "black and white", based on the 4 timing parameters I mentioned earlier.
    For CCM the limitation is more of a "grey area", not based on IC parameters, but on system loss considerations.

    The SR controller determines the turn-off point when the SR-FET Vds becomes less negative than 9mV below GND.  When operating in CCM, the SR controller cannot predict when the primary side switch turns on, but the secondary current reverses and Vds voltage becomes positive and triggers turn-off of the SR.  But this is after the fact of primary turn-on, so there will be some peak current of cross conduction depending on the turn-off delay and the leakage inductance.  This is similar to, but not exactly the same as, reverse recovery of a diode.   This cross-conduction current is a loss each switching cycle.

    So at lower switching frequencies, the cross-conduction switching loss is relatively small, and at higher frequencies, it becomes quiet significant.  In his experience, 200kHz seems to be an upper limit of tolerable loss vs. SR performance for CCM applications using this controller. It will continue to work at higher frequencies, but the turn-off cross-conduction losses begin to degrade the efficiency gain of the SR over a diode.  It is not readily quantifiable, because it depends on several factors and parasitic parameters that vary from one design to another.  For example, an SR in a high voltage output (such as 24V for example) doesn't reduce loss much over a diode, so CCM switching losses can impact the SR at lower frequencies.  In contrast, an SR in a 3.3V output is a huge improvement over even a schottky diode, so the frequency may be higher before switching losses severely erode the SR efficiency.    

    Even if the CCM switching losses are somehow mitigated at very high frequencies, there may come situations where the primary controller on-time may be shorter than the SR controller minimum off-time which then delays SR turn-on and increases loss.   

    Thus it is a grey area for "upper limit" on CCM frequency where the empirical observation is that, above 200 kHz in most typical CCM applications, SR switching loss may be a problem. 

    If this "answers" your question, please click on the green "Resolved" button.

    Regards,
    Ulrich

  • 0 in reply to Ulrich Goerke
    Mastermind 6180 points
    Hi Ulrich,

    Understand, appreciate for your detail answer and great support.

    Best regards,
    C.T.