Peter Mcalpine wrote:

> my setup at this stage is...
> 3.6V/600mah battery (NiMh)

NiCd is a much better choice.

> fast charge current 10% + cct = 65ma
> trickle charge current 1% + cct = 11ma

So your circuit needs 5mA.

> Solar cell 12V 1.26W (Vmax 18V, Imax 160mA)

Be careful of these figures, 18v is no load, and
160mA is into a short circuit. Actual MPPT peak will
be about 15v @ 84mA (1.26W). A cloudy day may only
net you 1/4 that power output.

> Given that I have a 'wish' for a fixed current output, how
> about using an SMPS as a current regulator?

Better to do battery management, and make sure it's
kept as full as possible at all times.

> I know it is not going to be looking for a peak but I am
> not sure I understand how MPPT would work with this setup?

Quite well I think. Using MPPT even in a crude form
wth your setup will probably give 2x or more into
your battery. If you run the panel direct into a full
3.6v NiCd at 4.2v you'll only get about 100mA in
peak sun or 0.42W. The MPPT at 90% eff will give your
proper 1.26W x.9 or 1.13W, a 710mW benefit.

Since you are using a known panel type you can just
characterise panel performance and measure panel
no load volts for a few mS, then adjust PWM for a
few seconds on-load at the correct panel voltage for
MPP.

This requires keeping your PIC running for the
PWM, but that's only about 2mA or 8.5mW, MUCH less
on average than you will be making from the MPPT.
-Roman

--
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