On Jun 5, 2013, at 4:30 PM, Jesse Lackey <jsl-ml@celestialaudio.com> wrote:

> Hi all, I'm looking at dc/dcs for running 400mA - 1Amp highpower LEDs=20
> with some special requirements, and in thinking about various schemes, I=
=20
> wonder how (or if) something this simple would work.
>=20
> Setting aside the minor additional complexity of making a=20
> constant-current vs. constant-voltage dc/dc, what if there were a=20
> circuit like:
>=20
> +12V-->PFET-->Cap-->Output(say 5V target).
> The output would have a voltage divider (i.e. feedback) going to a=20
> comparator, which has a reference voltage for its other input.  The=20
> comparator would turn on the PFET if the feedback voltage were below the=
=20
> reference voltage, and turn off the PFET if not.
>=20
> Schematic...
> <http://www.celestialaudio.com/pfet-comparator-cap-dcdc.png>
>=20
> Now I haven't even tried a simulation (which I rarely do, I don't do=20
> much analog that isn't largely 'cookbook' with tweaks) much less built=20
> it, so I have no 'real data' on how well or if it works.
>=20
> But it sure seems like it would operate more or less correctly.  As C1's=
=20
> value gets larger, it takes more time to charge (thru PFET) and=20
> discharge (thru load), and as the comparator hysteresis gets smaller,=20
> the PFET switching on-time gets shorter, so it switches more often (less=
=20
> efficient) but you have less output ripple.
>=20
> If the comparator were an opamp, driving the PFET gate directly, you'd=20
> have a DIY linear regulator.  By using a comparator with hysteresis, you=
=20
> get a dc/dc.  I think.
>=20
> For my application, I will know the load (a string of LEDs being=20
> high-speed PWM dimmed), the approximate needed output voltage to light=20
> them at the designed current, and will be doing a little trickery to=20
> handle all that.  I can tolerate a moderate (humm... maybe 10%? TBD)=20
> output voltage ripple.
>=20
> This seems somehow too simple to work reasonably, but I'm not sure why.=20
>  C1 would have to handle the ripple current and the inrush current (as=20
> would the PFET - something that would need a bit of additional design=20
> work, can be handled a few ways), R3 has to be low enough value to turn=20
> off the PFET quickly when the NFET turns off, but high enough value to=20
> not be too wasteful when the PFET is on.
>=20
> I like that I can set the output ripple by choosing C1's value and the=20
> comparator hysteresis.
>=20
> Thoughts?
> What's going to fail first if run 24/7 for years?
> Similar, polished, known-working designs anywhere?
>=20
> Thanks all!
> J
> --=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
--=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
.