In message 
          The Natural Philosopher  wrote:

> On 14/01/2021 14:17, David Higton wrote:
> > In message 
> >            The Natural Philosopher  wrote:
> >
> > > No, that's not true. To step up you need to use a buck converter with a
> > > choke big enough to store energy to push the voltage up. To step down
> > > you only need a choke to smooth HF ripple
> >
> > Not true.  The step-down inductor needs to store energy too.
>
> It isn't a step down inductor.

Just a short form of words - I was referring to an inductor in a step
down regulator.

> you just use variable PWM  on the raw voltage. All you need to do is use  a
> cap  and some form of current limiter - a resistors works but  dissipates
> power

There are regulators that use capacitors, sure (commonly step up, e.g.
for EIA-232 line drivers), but they inherently require significant
resistance in the switch elements to limit the current, so they're
lossy.  For a low power application the losses don't matter, but in
the case discussed here, I don't think you'd want that level of loss.
It works out the same as a linear regulator, according to my maths.

Using an inductor in a conventional step-down switching regulator is
much more efficient, and of course requires the inductor to store
energy.

(Apologies for the delay in replying - the news server I use changed
its implementation of NNTP and I have had to modify my news transport
client's code to suit.)

David

--- SoupGate-Win32 v1.05