Connect with us

Feedback for Flyback converter

Discussion in 'Electronic Design' started by The Al Bundy, Sep 27, 2004.

Scroll to continue with content
  1. The Al Bundy

    The Al Bundy Guest


    I've a little problem with making a feedback from an isolated secondary to
    the primary PWM controller.

    I plan to use a LM431 as opamp to drive an optocoupler. However the maximum
    A-K voltage of this device is 37V.
    The output voltage at the secondary will be 60V. Since there will be only 1
    secondary output I can not take a lower output to drive this optocoupler and
    to keep the Vak lower then 37V.

    I saw some supply with a 48V output that had a zener diode in series with
    the optocoupler. I wonder how this can go ok if the LM431 is not conducting
    and the full 48 (or atleast >37V) is across the A-K of the device. Ok, there
    was a series resistor also of course, but in any way the LM431 will be
    (shortly) used out of spec.

    So my question... is there any good way to make a feedback from one 60V
    output? This output must be well regulated and therefore the feedback must
    be done from this output.

  2. Joerg

    Joerg Guest

    Hi Al,

    How about a couple of resistors as a voltage divider on the 60V side? If the 1% tolerance is too much you can obtain resistor arrays with much better relative accuracy.

    Best would be to post a schematic on your web space or on a.b.s.e. so we can all have a look at what you got so far.

    Regards, Joerg
  3. The Al Bundy

    The Al Bundy Guest

    the 1% tolerance is too much you can obtain resistor arrays with much better
    relative accuracy.
    can all have a look at what you got so far.
    Well the problem wouldn't be there if the secondary didn't had to be
    isolated from the primary. This means I have to cross that barrier with an
    optocoupler. And the only way (afaik) is to do this with a LM431 to 'drive'
    the optocoupler.
    I will try to make some schematic readable and will post it on a.b.s.e. Will
    notify when it's there (can take a day or 2 with my agenda... :)

  4. Joerg

    Joerg Guest

    Hi Al,
    You could also drive it with a comparator, or even a simple emitter
    follower but that all depends on the desired precision.
    Ok, plenty of folks out here that will have a look at it.

    Regards, Joerg
  5. legg

    legg Guest

    The biased series zener or a zener limiter on the regulation node are
    both capable of producing acceptable performance from the TL431. You
    should also consider protecting the TL431 input from transients in the
    voltage sensing string, if any feedforward is used. Observe the input
    bias expectations of the TL431, and power expectations in the led
    emitter current string.


  6. Hi there. The last several pages of this datasheet:

    Talk about loop compensation of an isolated control loop using the TL431
    (Texas Instruments brand of LM431). Figure 13 shows a typical schematic. I
    haven't studied it closely enough to be certain but I think Q5 can probably
    be replaced with a simple resistor of appropriate value for your
    application. D14 would then protect the LM431 from bad stuff.
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day