Subject: Re: Core gapping techniques and general SMPS magnetics stuff
Date: 16 Dec 2002 12:28:08 -0600
References: <6hNK9.982$qA3.email@example.com> <firstname.lastname@example.org>
X-Newsreader: Forte Agent 1.92/32.572
On Mon, 16 Dec 2002 09:29:12 -0800, "Christopher R. Carlen"
>> I didn't see anything about your input voltage, but assuming that
>> it's relatively low, like 12 volts, one reason to use a transformer
>> rather than a single inductor is that for large step up ratios the
>> current pulses during the off time of your switch get very large at
>> the small pulse widths when you use a single inductor. With a
>> transformer you can get some of the step up from the turns ratio and a
>> more favorable duty cycle; the output current pulse magnitude is
>> proportionately smaller. But, at low power levels, this is quite
>> doable even with a single inductor.
>But those current pulses are going through the diode and into the output
>capacitor, not through the switch. So I think it bears upon ripple in
>the output, but not upon the stresses faced by the MOSFET, right?
That's right; it may require a bigger rated output diode. Also,
the input MOSFET will have to be rated for the full output voltage,
rather than the lesser voltage seen if you get some of your step up
with turns ratio. The higher voltage rated MOSFET will have a higher
Rds(on), and therefore, higher I2R losses.
>The other aspect is that with a transformer, there is a leakage spike
>across the MOSFET, forcing me to either use a rather over-rated voltage
>MOSFET anyway, or to snub, which I really don't want to do. The
>inductor avoids this entirely.
As I just mentioned, the inductor will still require an over-rated
(voltage rating) MOSFET anyway. Modern FET's are quite able to absorb
the leakage spike without damage; it's just a matter of the loss of
efficiency. If the leakage inductance is made low with good
transformer design, and the power level is low, just allow the MOSFET
to absorb the spike energy in repetitive avalanche mode. You need not
over rate the voltage of the MOSFET for the spike. Just pick a MOSFET
voltage rating sufficient to stand off the output voltage pulse as
reflected to the primary (and reduced by the turns ratio), and let the
higher voltage of the spikes avalanche the MOSFET. The rating for
this is usually given in modern MOSFET data sheets. You could compare
the higher I2R losses with the single inductor (because of the need
for a MOSFET able to stand off the maximum output voltage), with the
losses caused by absorbing the repetitive leakage spike energy, and
pick the best one on that basis.