>
>
>  there has been much discussion
> on this list regarding cheap but discrete switch-mode buck
> regulators.  So I went looking for those discussions.
>
> That lead me to Russell's GSR regulator and Richard Prosser's design
> which lead to Roman Black's regulator.
>
> I'd probably use an MC34063 :-).
(Or an NCP3065 if I cared about current sense efficiency)
But almost any buck converter IC able to tolerate the input voltage and
sense output voltage (here =3D current sense resistor voltage) would probab=
ly
suffice.

Richard's circuit spawned Roman's circuit but is not really appropriate for
your task.
Roman's circuit is simpler than mine (which was the aim).
"My" GSR is slightly more complex but has substantially better regulation
with load and Vin.

But unless you are building thousands+ and cost is important (bonus: and
Vin is 12 - 200+ V as it was for the original GSR*) then a cheap IC
solution is easier and probably superior.



         Russell

* Received knowledge says that a 16:1+ range for Vin is not practical for a
single stage buck converter. Fortunately RK is wrong.
At 200V+ the original GSR managed about 50% efficiency, but that's a lot
less dissipation than the 6% efficiency of a linear regulator, and at say
30-100V where it would typically be used efficiency was much better. That
was in an exercise bike and you only got 200V+ when children (& idiots)
having fun put it on no load  and pedalled it at speeds far in excess of
sane or usual. In subsequent incarnations I'd dial in a smidgen of load
when voltage went to too too high. As load was increased to limit Vmax as
desired this gave the super enthusiastic a constant speed variable load
mode :-).
--=20
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.