Connect with us

Inductors

Discussion in 'Electronic Design' started by Myauk, Oct 25, 2006.

Scroll to continue with content
  1. Myauk

    Myauk Guest

    According to my knowledge, a practical inductor is composed of an ideal
    inductance in serires with the effective resistance of that inductor.

    Skin effect increases effective resistance with the increase in
    operating frequency.

    For receiver ckt, the selectivity is calculated by 2*pi*f L/r, which
    means the selectivity depends on frequency as well as effective
    resistance and inductance. So the design of an inductor depends on the
    required selectivity and frequency as well as consideration of
    effective resistance.

    For a SMPS inductor design, I learned that the inductance for forward
    converter is calculated by (Vin-Vo) Ton max/ I min.

    I need to understand more about this.

    How come the inductance depends on the input and output voltage
    difference, the maximum on time and the minimum current drawn.

    Are the equations I know right or wrong?

    Where can I find these practical equations for calculations as well as
    descriptions on how they are derived from the fundamental theories of
    Electricity and Electronics?
     
  2. Joel Kolstad

    Joel Kolstad Guest

    That's a decent 1st order model; there are fancier ones available if you need
    the extra accuracy.
    It also increases the effective inductance until you hit self-resonance.
    Yes, although in many cases you try to have the inductor's resistance be small
    relative to, e.g., load resistance.
    I suppose you could say that, but keep in mind that how the core behaves
    (specifically its loses, which go up superlinearly with frequency, and sizing
    it to avoid saturation) plays a large part in the overall inductor design as
    well.
    For a "standard" buck converter operating in non-continuous mode, yes.
    Well... first, you might want to mentally separate "RF inductor design" from
    "SMPS inductor design." In the former, at least for "signal" levels,
    frequency response and self-resonance tend to be the driving factors in making
    a design work. In the later, the power you need to deliver and the switching
    frequency drive current through the inductor, and this drives the wire size
    and the core size (to avoid saturation). Unless you're building switchers
    above, say, a MHz, you usually don't have to worry much about self-resonance
    and frequency response.
    Umm... err... rather than write half a page here, could you get a copy of,
    e.g., Abraham Pressman's "Switching Power Supply Design" or similar? (Check
    Amazon to find similar tomes.) It goes through the derivation, which isn't
    difficult (just algebra), but it helps to have some pictures to look at.
    Someone like Eeyore (if he took his meds today :) ) can probably quote you
    the derivation irectly. If you can't get ahold of a book, look for
    application notes on switch-mode power supplies on, e.g., Linear's web site.
    They're "correct" but I get the impression you're not aware of a lot of the
    context that they're to be used within.
    University lectures? Books? Application notes on web pages?

    If you tell us what you'd like to design, we can probably point you towards a
    reasonably specific resource.

    ---Joel Kolstad
     
  3. Tom Bruhns

    Tom Bruhns Guest

    A really useful thing to know, assuming you're not put off by a little
    derivative notation, is

    v(t) = L* di/dt

    That is, the voltage across a pure inductance is proportional to the
    inductance and also to the rate of change of current. So if you put,
    say, 5 volts across an ideal inductor that's 20 microhenries, then
    current will increase at the rate of 1/4 amp per microsecond. If you
    want the current to build to 3 amps, leave the voltage applied for 12
    microseconds: 12usec*0.25A/usec = 3A. To more accurately account for
    what happens, you need to include effects like the resistance (note
    that current through the resistance lowers the voltage applied to the
    ideal inductance), the stray capacitance among the turns, saturation of
    the core material that effectively makes the inductance a function of
    current (and therefore of time), and some other things like that. But
    to a first order, with inductors that are properly applied, the v = L *
    di/dt will go a long ways in aiding basic understanding. Can you, for
    example, now see where that SMPS formula you posted comes from?

    Of course there's lots more to know, but see where that takes you.
    (Transformers are coupled inductors, and the above can be extended to
    cover them, also to a first approximation, pretty easily.)

    Cheers,
    Tom
     
  4. Myauk

    Myauk Guest

    I generally know that where this equation comes from.
    What I do not understand is how come Ton is maximum and the I is
    minimum for forward converter inductor design.
    But now by your practical calucation example, I would be able to figure
    out it.
    Thanks
    Regards
     
  5. Myauk

    Myauk Guest

    I would like to work on the flyback and resonant power supplies.
     
  6. Fred Bartoli

    Fred Bartoli Guest

    Joel Kolstad a écrit :
    Yes and no. It's a bit misleading.
    The apparent inductance raise with frequency below resonance because of
    shunt capacitance.
    But it'll also raise *without* this parasitics, between 2 asymptotic
    values (one DC value and one HF value) thanks to skin effect that
    changes the current distribution and thus the total stored energy. (look
    for internal inductance).
     
  7. Joerg

    Joerg Guest

    As Joel wrote, start by studying either a good book or app notes. When I
    learned this stuff in the 80s/90s I found that Unitrode's app notes were
    excellent. Still are, except now you find them on the web site of TI
    because they acquired Unitrode.

    With SMPS you need to start thinking about inductors as energy storage
    devices. IOW what goes in must come out in discontinuous conduction mode
    (DCM), or partially come out in continuous conduction mode (CCM) but
    there you must pay attention that the balance remains at healthy levels.
    Especially in the latter case great care needs to be taken to avoid
    inductor current "ratcheting" which will result in profound destruction
    if unchecked. When the voltage on a cap increases the cap will
    eventually break down. When the current of an inductor increases it will
    simply go into saturation and the device feeding it will typically be
    destroyed, usually in a rather spectacular manner.

    Also, study diodes (especially their storage time effects), gate control
    of FETs, ringdown etc.
     
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

-