>>  there is an old lead acid charger IC that was much
>>  used say a decade ago

> ... UC3906 ../

That's the one.

> A worst case of 8V for the discharged SLA, means 7V (Vsupply - Vbat) @
> C10 rate (500mA) and about 3.5W dissipated in the series pass transistor.

Yes - worst case can be very low, but 8V is rather deeeep discharge.
In normal use, if you value the life of your cells, the voltage would
almost always be much higher. If these are being run day to day then
you'd catch them well before that point - or, if not, they are
under-dimensioned (or used too long). (But, you know all that :-) ).

If you were prepared to treat them to a proper reconditioning charge
occasionally and were happy to treat them "well enough" most of the
time, and OK charger could be made using one or two LM317's per
channel.
Using two allows separation of voltage regulation and current limiting
- charge at current limit until plateau voltage is reached. By running
all units from a common bus that may be enough. Temperature
compensation can be added. A single LM317 can provide current and
vo9ltage limiting with a bit of playing - eg use an input resistor
which drops input voltage to voltage regulator so that if current gets
too high the regulator is starved of headroom. Crude but would
probably work.

A similar rather low cost scheme with no formal regulator could use a
FET per channel with a lowside current sense resistor and a battery
voltage sense.  Two op amp sections, 1 FET, a few resistors per
channel. One opamp section with care. You can have a common reference
rail which can be tweaked to meet special needs - eg today we charge
all to 14.2V - usually 13.6 etc. Somewhere around $1/channel if that
for actual electronics. This of course gives charge rate limiting and
maximum voltage setting but nomne of the fancier features.



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