At 10:06 AM 8/16/01 -0700, you wrote:
>I have a power supply problem which I am hoping to get some good suggestions
>on..... (I have lots of my own bad ideas)
>
>I know very close to nothing about switching power supplies, but I suspect
>that there is a buck regulator which will do this in a neat package.  I have
>looked in the obvious places, but I cannot find a device which will take the
>high input voltages in a small (low pin count) package.  I prefer to not
>have an inductor in this supply, current demand is low, I wish to minimize
>stray magnetic fields due to the Hall effect sensor, and there is so little
>space.

80V is difficult (ie. expensive). ISTR some LTC ones that will take 80V..
ones for line voltage operation tend to shut down due to the UVP assumption
of non-logic-level MOSFET drive near the 10V minimum you need to operate at,
and often don't include the switch. And you would still need the inductor.

What about a simple discrete linear regulator followed by an IC regulator?
Size the transistor so it will take the ~1.5W required for the 2 minute test
(should be ok when potted). The linear regulator consisting of a TO-263
or SOT-223 transistor, a zener and a resistor, plus a series diode 1N4005
for reverse voltage protection, plus a 2nd linear regulator such as an LM2931
or a LM78L05. You can set the first regulator to output, say, 6V or 9V,
so the second regulator sees very little dissipation. Normal operation,
this is less than 0.4W max from the transistor (at 26V), not unreasonable.
Maybe plus an additional capacitor or two, and a TVS if you feel very unlucky.

Alternatively, with a few more parts you could shut down the input regulator
under over-voltage conditions, but I don't think it's necessary with the
specs you've given (and assuming that the 80V/2minute test is an ABS-max
type of one-time test and is performed at 25'C).

Other than the 2-minute test, a shunt regulator might be nice
at these power levels. A TL431 plus a tiny 1A MOSFET would easily handle the
surges, the power dissipation would go into the series resistor (say 220R
wirewound). But the 30W dissipation during the 2 minute test will be rather
destructive, methinks.

Best regards,
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Spehro Pefhany --"it's the network..."            "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com
Contributions invited->The AVR-gcc FAQ is at: http://www.bluecollarlinux.com
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics