Philip Pemberton wrote:
> Has anyone built an "intelligent" NiMH charger out of a PIC?

Yes.  I built one for some small 3-cell 400mAH NiMH packs.  It's been a
while, but I think it was controlled by a 18F252.  I used to have it on the
web site somewhere, but I'm not sure it's there anymore.

>  - Onboard voltage-vs-time data log (I need an excuse to use one of
>    those FRAM chips that's been cluttering my junkbox for ages)

I just had it send data out a serial port.  If I wanted log information I
would run it while connected to a PC.

>  - PC link to download data log, view charger status, upgrade firmware
>    and tweak algorithm timing

RS-232 is easy.  It may be easier to provide the ICSP connection and install
new firmware with a programmer than to mess with a bootloader.  You will
need the ICSP connection for debugging anyway unless you have a ICE.  Or you
could even socket the PIC and program it externally if you don't want to
give up the pins.  Even if you update the firmware 100 times, it's still
very unlikely that this will take more time than integrating and testing a
bootloader.

>  - 16x2 LCD display

Unless you really really need it to display what it's doing stand alone, it
could just send regular status out the serial port and a PC program displays
it for you.  I dealt with the status of each of the 5 ports of my charger
with a single LED per port.  You can do a lot with various flashing
patterns.

> As far as current regulation goes, I was going to use an opamp with a
> push-pull driver stage to do both charging and discharging. A DAC will
> be used to produce a +/- 2.048V signal that'll be fed to the opamp to
> select the charge/discharge voltage.

Charging should be current, not voltage, controlled.  I did mine with two
separate circuits.  The discharge was a resistor to ground via a NPN
transistor.  You don't need anything too fancy to discharge the cells.  The
charge circuit was a voltage controlled current source.  PWM signals from
the PIC were filtered to produce the control voltages.  This was all
software PWM, but nothing needs to be done very quickly so it could be
heavily filtered to allow good resolution.

> That said, it seems Galaxy Power have vanished,

They are now Bel Power, a division of Bel Fuse.


******************************************************************
Embed Inc, Littleton Massachusetts, (978) 742-9014.  #1 PIC
consultant in 2004 program year.  http://www.embedinc.com/products
-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist