Anyone used NCP3065 driver?- Mtbr.com
Results 1 to 6 of 6
  1. #1
    mtbr member
    Reputation: hootsmon's Avatar
    Join Date
    Feb 2008
    Posts
    762

    Anyone used NCP3065 driver?

    Planning to make a cheap 'n cheerful driver-board for my next project, namely an XM-L triple for the handlebars.

    My intention is to use this NCP3065 Constant Current Buck Boost Inverting Switching Regulator , in a SEPIC configuration, based upon OnSemi's reference design.

    • I searched this forum (and CPF too), however didn't turn up a lot of info.
    • Given there's a lot of brainy folks on this forum, I'm hoping for some advice
    • The intended application is 3x XM-L in series @ approximately 2500mA.
    • Haven't settled on a battery configuration (yet), but I will probably cobble-up something based on 18650 cells salvaged from a defunct laptop battery-pack. Also I have some 3S LiPo R/C battery-packs that might be OK.
    • For a prototype, I'll probably mount a NCP3065PG (ie. DIP packaging) on breadboard.
    • Will setup the NCP3065 to control an auxilliary FET, boosting output to around 3000mA or so.
    • Assuming proto works as hoped, then I may consider doing a small batch of PCBs.
    • Heck, if doing a PCB then why not mount the 3x XM-L's directly on the board?


    Thoughts?
    Brisbane, AU

  2. #2
    mtbr member
    Reputation:
    Join Date
    Jan 2010
    Posts
    291
    I'd suggest you look for a better switcher IC. That ON device doesn't appear appropriate for a 3A driver. 0.235V voltage sense means you have 0.7W loss just in the sense resistor. Slow 250KHz switching means big inductor/capacitors.

    Just lots of things about that chip that would make me look elsewhere.

    I recommend you check out Linear Tech's website for some "good" switcher IC's.

    Also, for a 3A boost or sepic driver you need to go straight to a PCB with good layout, not something like a DIP package and a breadboard....

    Efficiency of a Sepic isn't wonderful and at 3A any losses will lead to high power dissipation. I'd recommend you look at a buck driver and use 4 series li-xxx cells. Efficiency would be up in the low/mid 90% and life will be a lot easier.

    cheers,
    george.

  3. #3
    mtbr member
    Reputation: hootsmon's Avatar
    Join Date
    Feb 2008
    Posts
    762
    Quote Originally Posted by georges80
    I'd suggest you look for a better switcher IC. That ON device doesn't appear appropriate for a 3A driver. 0.235V voltage sense means you have 0.7W loss just in the sense resistor. Slow 250KHz switching means big inductor/capacitors.

    Just lots of things about that chip that would make me look elsewhere.

    I recommend you check out Linear Tech's website for some "good" switcher IC's.

    Also, for a 3A boost or sepic driver you need to go straight to a PCB with good layout, not something like a DIP package and a breadboard....

    Efficiency of a Sepic isn't wonderful and at 3A any losses will lead to high power dissipation. I'd recommend you look at a buck driver and use 4 series li-xxx cells. Efficiency would be up in the low/mid 90% and life will be a lot easier.

    cheers,
    george.
    Ah, thanks Georges! Yes, everything you've said makes sense.

    You know, after a bit of thought I might just go the H6flex route instead, as that way I'd have a working light with much less effort . Didn't say this before, but on my last project I blew-up a beautiful MaxFlex (mostly due to user error), which led me to want to try and build something idiot-proof next time .

    Anyway, thanks for the excellent advice, as always.
    Last edited by hootsmon; 12-22-2010 at 02:53 AM.
    Brisbane, AU

  4. #4
    mtbr member
    Reputation:
    Join Date
    Feb 2009
    Posts
    169
    +1 on everything that George had to say. I looked at the OK semi products and decided to pass on them for the reasons he stated. High frequency helps on the inductor size, but it can make gate drive more problematic. You don't want to give up the gains to heat in the switching transistor. It's amazing how complicated it can be given the rather small number of components.

    Quote Originally Posted by hootsmon
    Heck, if doing a PCB then why not mount the 3x XM-L's directly on the board?
    Do you know who makes small prototype metal clad PCB runs? I know where to get thousands made, but not 10. Well not for a reasonable cost that is.

  5. #5
    mtbr member
    Reputation:
    Join Date
    Jan 2010
    Posts
    291
    Quote Originally Posted by vroom9
    +1 on everything that George had to say. I looked at the OK semi products and decided to pass on them for the reasons he stated. High frequency helps on the inductor size, but it can make gate drive more problematic. You don't want to give up the gains to heat in the switching transistor. It's amazing how complicated it can be given the rather small number of components.

    Do you know who makes small prototype metal clad PCB runs? I know where to get thousands made, but not 10. Well not for a reasonable cost that is.
    There's some very nice modern FETs with extremely low gate capacitance and charge these days. Makes the gate drive issue nearly a non-problem - though it DOES require PCB and short trace runs and good layout practice. Anyhow, as you say, it can be quite complicated given the small number of components...

    Proto MCPCB... never found a place and probably not likely to. Just the tooling and special run aspect makes it something that will be expensive. The only reason the standard FR4 proto stuff is cheap is that they can run it as a normal lot without any special processing.

    cheers,
    george.

  6. #6
    mtbr member
    Reputation: hootsmon's Avatar
    Join Date
    Feb 2008
    Posts
    762
    Quote Originally Posted by vroom9
    +1 on everything that George had to say. I looked at the OK semi products and decided to pass on them for the reasons he stated. High frequency helps on the inductor size, but it can make gate drive more problematic. You don't want to give up the gains to heat in the switching transistor. It's amazing how complicated it can be given the rather small number of components.
    Good info, thanks

    Quote Originally Posted by vroom9
    Do you know who makes small prototype metal clad PCB runs? I know where to get thousands made, but not 10. Well not for a reasonable cost that is.
    I'm told PCBCART does FR-4 PCBs at a reasonable price.
    However I haven't a clue about MCPCB prototyping (sorry), cos' I'm an idjut.
    Brisbane, AU

Members who have read this thread: 0

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  

THE SITE

ABOUT MTBR

VISIT US AT

© Copyright 2020 VerticalScope Inc. All rights reserved.