> I'm having a look at the LTC3862 SMPS controller. I've at
> least 4 applications in mind that could use a high-current battery-
> powered converter and this looks a chip of interest
>
> http://www.linear.com/pc/productDetail.jsp?navId=3DH0,C1,C1003,C1042,C103=
6,P85098
>
> http://cds.linear.com/docs/Datasheet/3862fb.pdf

> If I set the output voltage to 15V - 20V, I'm assuming the available
> amperage will rise accordingly, if the source can supply it

Yes.

> Note Page 37, A 4.5V to 5.5V Input, 12V/15A Output 4-Phase
> Boost Converter Application Circuit, that they suggest two whole
> circuits in parallel
>
> Would this be the preferred and least smokey method ?

As opposed to what?

Two phases give you less output ripple and ease the filtering
requirements. The lower "peakiness" probably improves efficiency due
to nicer diode conduction. Also lower I^R losses in inductors and FET
and/or double the R allowable.  So, two phases has dew disadvantages
once power gets up. Two of inductor, Dout , FET, Rsense. More board
area but not double. The claimed efficiencies are exceptionally good.

Four phases with a 90 degree pja\\hase offset gives better again
results. Double the above again.

The output power mentioned is not vast (180 Watts) but at 4.5V in and
say 90% efficiency (one can dream) that's about 45 amps.

With a single FET and say  0.2V drop (about 5% of Vin) that's an Rdson
at working temperature of about 4 milliohm.
That usually means a spec of 3 milliohm or ;ess.
Using two FETS takes that to 6 milliohm and 4 FETs to 12 milliohm.
Even that is "no slouch."

If your Vin is higher the task gets easier.
You need to provide actual requirement to allow good what-if ing.

What Vin.
What pout.
What Vout ...?

 Russell
--=20
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.