Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Homework Help: Analogue Electronics: SMPS

  1. Jul 15, 2007 #1
    Hey guys ... I have inserted a picture of the question as it was easier to scan than to re-write :P It's based on Switching Mode Power Supplies.

    1. The problem statement, all variables and given/known data
    Check the picture or the upload on photobucket http://i91.photobucket.com/albums/k320/shahilj/Q3.jpg" [Broken] ... It is basically a full bridge config that is required. What got me was that the supply ranges from 220Vac down to 12Vdc BUT the o/p is 12Vdc ... I got a bit confused as to what config to use.

    2. Relevant equations

    All equations are to be derived from first principles as such.

    3. The attempt at a solution
    Heres the 3 links first .... http://i91.photobucket.com/albums/k320/shahilj/elec3ans1.jpg" [Broken]


    I kinda need someone to at least just check the work .. coz it is worth a lot of marks (25 out of a 100 mark paper) and I really need to know if my method is indeed right!

    Thanks so much!
    Last edited by a moderator: May 3, 2017
  2. jcsd
  3. Jul 15, 2007 #2
    Also, had a problem with this question as well :frown:

    Honestly, I have no idea where to even start with this one:confused:

  4. Jul 20, 2007 #3
    hey guys :( seems nobody can help ... um, I've looked on the net for help with the second question I posed ... can't find relevant help anywhere .. please can someone at least guide me in some direction coz I write on Monday and um, well ... I might fail without it!
  5. Jul 23, 2007 #4


    User Avatar

    Sorry I can't be of too much help in the time required,
    but it sounds like the configuration could sanely be
    one of several possible ones.

    The effect is certainly that of a buck-boost converter
    in that you're trying to sometimes boost a smaller
    dc voltage up to a higher output voltage.

    Yet you're also (over a much wider factor of input
    to output voltages) trying to have an efficient buck
    converter to step down higher input voltages to lower
    output voltages.

    It doesn't seem clear that they're saying you should
    accomplish the task in any particular circuit configuration,
    or with a minimum number of parts.

    They're saying to optimize the duty cycle of your switches,
    and to give equations that model the thing from first

    Yet also they ignore your control circuit and say you
    should only show the power handling circuit.

    I'd be tempted to do a cascade design where the boost
    stage is activated for only 12vdc to 24vdc input and
    it functions as a stand alone boost converter with a
    simple topology and a maximum boost of 2:1.

    When the DC in is from 24VDC to 707VDC from a
    full wave bridge, then you could have a simple buck
    converter to reduce that to 24VDC out 2-10A.

    Obviously you could get into a bit of trouble with input
    voltages within +/- a few volts of 24VDC so that you'd
    need to be mindful of your voltage drops in diodes and
    resistive losses etc. so that overall it'd still work well in
    that region.

    Sometimes that's easily done with a 2 stage
    approach where a boost converter boosts to a higher
    than necessary voltage, then a buck converter takes that
    range of higher voltages down to the desired output.

    I don't know if that's any easier for you to design and
    analyze from first principles or not as opposed to a more
    complicated / integrated topology.

    Honestly, to me, SMPS circuits are pretty much cookbook
    things and you have only so many basic topologies to
    choose from and the analyses of those are well known in
    many cases.

    But since they're apparently not concerned with 'cost'
    or control topology or non-ideal effects in the
    transformers, capacitors, inductors, you can have a lot
    of freedom to design the blocks in ways that are simple
    for you to analyze and design even if real world
    aspects of size / heat / cost / whatever might cause
    some more complicated trade-offs to be made.

    You don't even have to show HOW it's controlled, it's
    only relevant that SOME pulse train of a reasonable
    duty cycle should cause the proper input to output
    voltage transformation at the expected range of output
    loads, and it's quite arbitrary WHAT that duty cycle is
    or HOW you would actually implement the
    control / feedback circuit!
  6. Jul 23, 2007 #5


    User Avatar

    http://www.powerdesigners.com/InfoWeb/design_center/articles/DC-DC/converter.shtm [Broken]
    http://www.ipes.ethz.ch/ipes/dcdc/e_BuckBoost.html [Broken]
    http://www.national.com/nationaledge/jun03/article.html [Broken]
    Last edited by a moderator: May 3, 2017
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook