> look at this alternative...

> http://homepages.which.net/~paul.hills/Batteries/BatteriesBody.html
>
> --> 11.2.2 --> UC3906

Does the job. But, as I said, cost is important and specialist ICs 
like this tend to be several times more than the whole parts budget 
alone.

What I'm most liable to do at present is to have a simple resistor 
charger from a slightly higher supply (whose voltage I can set with a 
little latitude) switched with a high side FET with the battery 
voltage monitored by the processor. I could easily PWM this for linear 
charge control but shouldn't need to. Resistor is dimensioned to 
provide an acceptable range of charge currents across the normal range 
of battery voltages and allow battery to be both boost charged and 
floated if desired.

The saving grace is that the maximum designed charge is low enough 
(for other reasons) that the series resistor will survive long enough 
under battery terminal short circuit for the processor to detect this 
and turn off the charge circuit.

Battery short circuit is unlikely due to construction method and 
environment but can also be made tolerable with a small series 
resistor as a fuse under extreme (and probably non existent) 
conditions. This latter resistor is NOT intended as a fuse per se.

Principal parts are PFET, small npn, 3 small R, 1 x 5W R (maybe 1W) 
and a diode. And a processor drive pin. Plus, of course a processor 
ADC port - but I was measuring this rail already. (Actually a 
downstream point which can be higher when another supply cuts in but 
it serves the same purpose). The really enthused could probably work 
out how to measure and drive with the same pin but I don't need to.


        Russell McMahon

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