> I'm thinking of just pasting a dallas 1-wire temperature sensor on my
> battery pack, and another on the circuit board, which will be remote.  That
> will get me a measure of the battery vs ambient temperature.

One of these days I'll get some of those, they sound real cool and I'm
tired of thermistors :-)

> How well does the voltage into a charging NiMH need to be regulated?  I am
> planning on using an unregulated 12V supply only regulating the current ,
> not the voltage.    The rated amp-hour capacity of my battery stack will be
> about 5-6 amp-hours.  I understand that is also the maximum quick-charge
> current.

That's fine, although it does mean you'll get less efficient charging(but
it's how almost everyone does it except laptops that already have a
regulated power supply). Figure out what your peak current is and make
sure the battery and your connectors are rated for it.

> Then I'll have a pass transistor that can turn off the power either briefly
> ( to measure actual battery voltage) or for good.  Won't need trickle
> charging, this device will be charged, then taken to the field, then brought
> back for charging.

You can also measure the peak-to-peak voltage and get an idea of what's
going on as the unregulated supply goes on and off 120 times per second.

> I could also charge off some solar cells.  These will have a max. capacity
> of only an amp at best.  I'm not sure how this lower-current charger would
> work, though.  Would the temperature method still work at 1/6th the
> battery's mAH rating?  Or do I also need to use the voltage method?  Seems
> like if you pump in some current, and deltaV is less than zero, zero, or
> just a little above zero, then you can't add any more juice.  If you could
> hit the flat spot just before it peaked, by stopping at a certain minimum
> slope of voltage, that might save the battery from early death??

At 1/6 you still get temperature rise although it will not get more than a
few degrees over ambient.

One thing I noticed about voltage, during the first 15 minutes of charging
you get a peak effect. That was another reason the peak effect didn't work
for me, it seemed like I'd spend a whole lot of effort trying to
characterize a battery, only to find it's really a moving target. I think
I have something perfect and then it stops charging after 10 minutes, or
charges until thermal overload.

One important thing I learned was that RA4 on a pic, being an open-drain
output, has no diode to V+, so you can use it as an output to drive a PNP
transistor or P-channel MOSFET on a high side up to 12 volts. This works
great for about a week or so, then RA4 doesn't work any more. I have a
feeling it's because the input circuitry is still connected and nasty
things happen over time. So don't do it :-)

Cheerful regards,

Bob Blick

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads