AVR-Chat

>
>I am trying to develop a modular portable power system based on
>a unit of 10 or 11 NiMH F cells to replace the common 10 NiCad or
>6 lead acid.  Use of 11 NiMH cells would give about 12 V at the low end,
>which would be an improvement, and would only be too high (>15V) when
>charging. 

Check your specs, I've seen as high as 1.6V as "ok" under 1C charge, for an old cell.

> However, charging may be done out of the load system, and
>in any case, most 12V inverters, lights, etc now can handle 15-16 volts 
>because the recommended lead-acid equalizing charge in now up to 15.5V 
>(although it still doesn't help extend battery life).
>I'll be using the lightweight 12V modules in series and parallel
>configurations
>to power electric scooters, in-the-field 120VAC power packs, etc, and as house
>back-up.  They can be charged anywhere where there is a temporary excess
>of power.  And of course, the smarts for their interconnectivity,
>manintenance, etc
>will come from AVR!

I can't hand you the code, but I can help you with the design of a charger. 
BTDT.  We used the tiny-26's PWM output to control the current, but if you know the charge rates you want, a buck switcher altered to be a constant current output, would be a lot easier to control.  You'll want to be able to turn it off rapidly.

You'll also want the ability to put a LARGE (>15C) discharge on the cells, for just a moment. Don't rely totally on firmware to limit the pulse duration. I favor a large electrolytic that is discharged by a parallel resistor, and charging the cap is what does the discharge pulse.  A kick-ass mosfet on the ground side controls it.