• State Resolved
  • Locked Locked
  • Replies 9 replies
  • Answers 2 answers
  • Subscribers 63 subscribers
  • Views 748 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:

UCC28780: UCC28780

UCC28780: UCC28780

robin gangopadhya18
Mastermind 6296 points
Part Number: UCC28780
Other Parts Discussed in Thread: UCD3138, UCC28701

Pl see attached screenshot of the output rectifier connection of the UCC28780EVM

Drain of the MOSFET is connected to the xmfr secondary with "dot" side of the secondary. With MOSFET, one can perceive how the start up occurs. Hopefully, UCC28780 sends otu SYNC ON signals on start up.

But how will this work if we are using a GaN such as EPC2412?

We have 10 such modules in our system. So we better be right in connecting GaN here.

Pl see slide 2 of the attached pptx file.

Apologize for a bit messy look of the slides. It is meant to show the drafting person how LMG3410s are hooked up as opposed to the UCC28780EVM schematic & also how EPC2214 replaces the MOSFET.

How will it affect startup?

thnx

robin

LMG3410_HOOK_UPS.pptx

  • 0
    TI__Mastermind 27095 points
    Hi Robin,

    I will review the slides and get back to you.

    Best Regards,
    Ben Lough
  • 0 in reply to Benjamin Lough
    TI__Mastermind 27095 points
    Hi Robin,

    I would caution against using LMG3410 for the high side clamp switch. The bootstrap circuit will not be sufficient to keep the high side LMG3410 powered during the period of no switching when the converter is in LPM or ABM modes. We have seen successful designs using UCC28780 + LMG3410 (low side switch) and a silicon MOSFET as the high side clamp switch and I would suggest using this approach in place of the high side GaN module.

    Best Regards,
    Ben Lough
  • 0 in reply to Benjamin Lough
    Mastermind 6296 points
    Ben: thnx much for the alert.
    I will get a Si-MOSFET back in the high side.
    Appreciate
    robin
  • 0 in reply to Benjamin Lough
    Mastermind 6296 points
    Ben
    After my earlier post on getting back to Si-MOSFET, I was checking a few posts we have had.
    So before I jump out of the hot water, let me recall salient features for this design. #1 It is fed by a regulated 380V bus. So there is no holdup or missing AC event here. #2 Our load is constant, mostly. It is possible to have no load- in which case the whole ps maybe disabled. Besides, consistent with other commands, we have a static load to keep dc dc converters in regulation mode all the time. The load is a totality of high power rf amps. They are always doing something. It is conceivable there will be LDOs downstream.
    So it is a fairly garden variety load- but we need to run at highest efficiency because of high-density packaging of many such modules.

    Hope that exonerates the bootstrap floating supply from being put into question.
    There was a response that explained why EMVM has 47uF.
    That also included extreme line/load situation. And we do not have those here so we can choose something more amenable to our needs.
    Would that kind of lessens your concerns for using the GaN pkg on the high side?
    thnx for your time & would be looking forward to your review.
    robin
  • 0 in reply to robin gangopadhya18
    TI__Mastermind 27095 points
    Hi Robin,

    Any chance you are able to add a small unregulated, isolated supply to provide biasing for the high side switch? Perhaps something like DCP021212U? This would get rid of the biasing concern. Other designs have used this approach in order to use LMG3410 as a high side FET such as PMP20289.

    If your loading is enough to stay in AAM mode, then this would help alleviate the concern as well provided the boot capacitance is sufficiently large.

    Best Regards,
    Ben Lough
  • 0 in reply to Benjamin Lough
    Mastermind 6296 points

    Ben:

    Yes. We can . I will fix it. And it is a good idea.

    And I would think it cannot be derived from the flyback "xmfr". 

    We have several "tiny"( complete 5v flyback from 380V using UCC28701 type- smallest ever isolated primary controlled smps of a few W) all over doing various floating biases. & this 380V also regulated by a super compact 3ph pfc( controlled by UCD3138)

    thnx much

    robin

  • 0 in reply to robin gangopadhya18
    TI__Mastermind 27095 points
    Hi Robin,

    Great! Yes, the high side bias supply would not be able to derived from the flyback transformer. You can follow PMP20289 for example of using 2x LMG3410 as a half bridge configuration. The difference would be connecting the high side drain to your clamp capacitance instead of Vin.

    Best Regards,
    Ben Lough
  • 0 in reply to Benjamin Lough
    Mastermind 6296 points
    will do. I have checked out the PMP20289- thnx for pointing this design out. Interesting!...because ours also is "flat pcb" xmfr but with very different structure. Methinks, better in many ways.
    I will share that with you asap.
    Can you then send me your ti.com email?
    I cannot post it yet in the forum.
  • 0 in reply to robin gangopadhya18
    TI__Mastermind 27095 points
    Hi Robin,

    My email is ben.lough@ti.com

    Best Regards,
    Ben Lough