>
>
> I don't have a temperature sensor yet. It's on the drawing board though,
> just
> as soon as I get my mitts on a few LM35 temperature sensors.


I used 10k thermistors, worked nicely.


> A 10min lockout? What, you ignore the dV for the first ten minutes of the
> charge cycle?
> I guess the negative dV is the delta from the peak voltage, i.e. Vpeak -
> Vnow.


Yup,  Packs can lower their impedance during the first part of the charge
cycle.

>
> Got that (or something close to it anyway). If Vcell > 1.7V then the
> charge
> algorithm fails with an "EOVERVOLT" error (maximum safe voltage exceeded).


Check your cell specs, I remember 1.5-1.6 depending on who I was reading.


> It really all boils down to the fact that I'd rather have two PICs working
> together than try and implement the UI code and the charging / PWM / etc.
> code
> on one chip. PICs aren't exactly expensive (16F88s seem to run about =A31=
.50
> each), so using two of them isn't really a big deal.


Hmm.. I had that, plus 500 kHz PWM, plus clocked serial communications with
the host, and still had plenty of time to spare.  Very little really happens
during charging, except the 10mS current control loop.



> I'm using Kodak 1600mAh AAs and Energizer 2500mAh AAs - I've got three
> sets of
> four of the Kodaks (plus a set of three), and one set of four of the
> Energizers. Note that these are loose AAs - I wanted them to use in my
> Gameboy
> (don't deny and old man his few pleasures in life) and my digital camera
> (which consumes Duracells at a stupid rate).


It's fun to compare cell curves. Very interesting.  The good cells give a
thermal SPIKE at the end, the cheap ones more of a ramp with a hill at the
end.
You can also cycle them into memory effect without too much work, though
it's pretty underwhelming when you see it in person.
-- =

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