From: Tim Ikeda (tikeda_at_sprintmail.com)
Date: Wed Mar 21 2001 - 20:49:32 EST
Hello Sven:
[...]
I wrote:
>> Meanwhile, I'm dissecting the old, dead battery pack and trying to
>> figure out if a NiCad battery case can be safely rebuilt with NiMH
>> batteries (the only problem I can see is if the temperature sensing
>> unit the in pack is specific for the battery type).
Sven:
>I'm afraid things aren't as simple as that. The Battery sensing unit
>is a lot smarter than that, and for all I know implemets a counter
>which, after loading the battery N times, will declare the battery
>dead, even though it might still be in perfect working condition. I
>own such a battery, which can only be charged externally and will only
>work in the TP 701 if all battery-sensing functions are disabled, but
>will then work perfectly. The only Problem with this is that
>switching the computer on has to be done on mains, as this tries to
>talk to the battery...
That's somewhat annoying. I just peeled up the three-wire chip that
came inside my old NiCad battery to read the markings. It's a Dallas
chip. The markings are: DS2434 9447C4 0190AA
Now I've looked it up: It's a battery identification and temperature
sensing chip. The chip has memory to store charge/discharge cycles
and other information (and, of course, it's now considered obsolyte).
There is something in the refs about non-volatile storage as well.
http://www.dalsemi.com/datasheets/pdfs/2434.pdf
The good news is that it should be re-programmable through the
Dallas one-wire interface. However, there are three downsides to
the problem: 1) One is that I'd need to steal the information off a
working battery pack (but no disassembly is required & I've got
a spare pack), 2) The chip comes with lockable non-volatile
memory that may be essential to battery pack identification.
(Worst case: Replace with a new chip and start from scratch),
and 3) I've got to piece together a one-wire interface (Sounds
like Basic Stamp time...).
>But please, report any success with this should you succeed.
I'll keep digging. Now I've got an incentive to learn the one-
wire device interface. And thanks, without that warning I might
never have bothered checking out the chip before rebuilding
the pack.
Regards,
Tim Ikeda (tikeda_at_sprintmail.com)
This archive was generated by hypermail 2.1.3 : Thu Jan 23 2003 - 09:56:58 EST