• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 42 subscribers
  • Views 466 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.

Original question:

PMP9656: up to 280w (23.3A)

PMP9656: UP to 300W operation

Will Wang47
Expert 6800 points
Part Number: PMP9656
Other Parts Discussed in Thread: UCC2897A, UCC24612

Hi team,

Could you help check if PMP9656 could goes up to 300W? 

If not, what customer should modify to achieve 300W? 

And customer also want to know, could they increase the frequency to 400kHz~ 500kHz to let the solution get smaller?

Hope you could share some comments.

Thanks and Best Regards,

Will

  • 0
    TI__Intellectual 2875 points

    Hi Will,

    For 300W, the output inductor should be changed.  The new inductor should be rated for 25Arms and around 30Asat.  I also recommend decreasing the current sense resistor (R5) to around 1.62ohms.   Of course, everything will be hotter at 300W compared to the performance documented in the test report at the original design of 250W.

    Increasing the switching frequency to 400kHz-500kHz is  a major change.  Nearly all of the power stage components would need to be redesigned.  The design could shrink, but the efficiency will be lower, resulting in higher temperatures.

    thanks,

    Brian

  • 0 in reply to Brian King
    TI__Expert 6800 points

    Hi Brian,

    Thanks for your feedback. 

    After checked with customer, they agree 400kHz ~ 500kHz is too high. And they are consider increase to 300kHz, how do you think this? Do you think this need to change all the power stage components including the transformer and MOSFET? Let me know if you get any concern. 

    Thanks and Best Regards,

    Will

  • 0 in reply to Will Wang47
    TI__Intellectual 2875 points

    Hi Will,

    For increasing the frequency from 200kHz to 300kHz, I advise reducing the magnetizing inductance of the transformer from 50uH to around 33uH, so that the peak magnetizing current remains the same.  

    The efficiency will be lower with the higher switching frequency.

    thanks,
    Brian

  • 0 in reply to Brian King
    Prodigy 30 points

    Hi Brian,

    I'm a Celestica Power Engineer. 300W output power is my request.

    Is there any solutions to improve the efficiency under 300k working frequency? We don't want to see the efficiency drop when we increase working frequency.

    Here are some items:

    a) I will redesign the transformer;

    b) Find low Rds(on) power switch;

    c) Optimize timing of SR.

    Most important of all, do you have a solution to fix Io reverse at light load during power off for this converter without too much cost increasement? 

    Thanks a lot.

  • 0 in reply to Brant LUO
    TI__Intellectual 2875 points

    Hi Brant,

    Optimizing the efficiency at a different switching frequency will require some tuning of the three items you mentioned (transformer design, FET selection, and timing of the SR FETs.)  Lower Rdson is not necessarily better for higher frequency, because switching characteristics typically get worse with lower Rdson.

    For preventing reverse current at light loads, one option is to turn off the SR drivers when the load is reduced.  The converter will still provide power, but current will conduct in the body diodes of the SR FETs.  Another option is to use the UCC24612 driver for the free-wheeling SR FETs (Q4 and Q7), and use an isolated driver controlled by the UCC2897A "OUT" signal to drive the other two SR FETs (Q1 and Q2.)   The UCC24612 provides diode emulation, and will prevent reverse current.  

    Both techniques mentioned  above allow the converter to run in DCM, so the primary bias circuit of D6, D7, L5, and C21 will need to be changed.  This is because this bias circuit provides a steady BIAS voltage with forced CCM, but the BIAS voltage will drop at light loads if DCM occurs.

    thanks,

    Brian