Article: 9312 of comp.sys.laptops
Path: uwm.edu!spool.mu.edu!hri.com!noc.near.net!news.bbn.com!bbn.com!sher
From: [email protected] (Lawrence D. Sher)
Newsgroups: comp.sys.laptops
Subject: Re: T1000SE battery life
Message-ID: <[email protected]>
Date: 24 Feb 93 13:18:02 GMT
References: <[email protected]>
Organization: Bolt Beranek and Newman Inc., Cambridge MA
Lines: 66
NNTP-Posting-Host: bbn.com

In article <[email protected]> [email protected] (Tom Proett) writes:
>Hi,
>
>I have a T1000SE which has a problem with the battery.  It will only
>work when it is plugged in now.  If I try to unplug it, it will
>shut down right away.  I took the battery out and hooked it to a
>flashlight bulb to drain it.  It lighted the bulb brightly for
>just under 6 hours.  I charged it again and it still would not work.
>
>Is there a setting on the computer itself which tells it to not shut
>down?  It seems that the battery is at least fairly good if it can
>light a light for more than 5 hours.
>
>Thanks.
>
>--
>[email protected]
>NASA Ames Research Center

This seems to be a common problem.  I don't know any definitive answers,
but since I faced the same problem, I had occasion to look into it.

1. Buy replacement cells, open the pack, and put in the new cells.  A
good source for replacement cells (in New England): Gates Energy
Products, Inc. 1 Prestige Drive, Meriden, CT 06450-7105 Tel: (203) 238
6912, FAX (203) 238 6887 Voice Mail: (904) 462 8725.  Attention: David
Childs, sales engineer, Northern Sales Region.  He was very helpful for
me.  The cells to get: GLF-1700A Ultramax NiCad.  (The cell size,
officially, is 4/3A.)  Cell cost is around $5.50-6.00 as I recall.

Opening the pack requires a careful dissection along the ultrasonically
welded perimeter seam, visible with the pack upside down.  You can put
it back together again with glue of various flavors.  Be sure to put
back the temperature sensor that is nestled in between two of the cells.

2. Exhume the existing cell(s).  Once the pack was open, I found one
cell that was weak (discharged quickly) and one that was zero volts,
zero ohms.  (Even so, of course, the remaining cells would easily run
most flashlight bulbs.) Taking a cue from various other posters ("put a
good cell in parallel with the bad one", "discharge a capacitor into the
bad cell") and feeling I had little to lose, I decided that if a little
was good, more would be better.  So I connected a 12v automobile battery
charger across each bad cell (one at a time).  I set the current level
to "trickle", connected positive-to-positive, negative-to-negative, and
turned on the power for about 3 seconds.  Scarey!  Longer connections
obvious raise the ante, like inadvertent explosions.  An ammeter (on the
charger) showed currents exceeding 10 amperes, tending to decay over a
few seconds to less than 10 amperes, but I wasn't about to get too
academic about this.  A voltmeter showed 0 volts before the treatment
and 1.2+ after.  I repeated this treatment about 3 times, with a
respectful interval in between, like a minute.  Voila, the cell now
looked normal, with an open circuit voltage of about 1.3.  I put
everything back together, let the Toshiba charger care for the pack
overnight, and the next morning, it all seemed to run fine.

Note:  I first tried this treatment across several cells in series, only
one of which was bad.  It did not help!

I can hardly guarantee the safety or long-term consequences of this
procedure.  But, clearly, there is some rationale to the talk that
a sufficient goosing of these cells can exhume them.
            .----------------------------------------------------------------
           /Internet email: [email protected]
Larry Sher <        US Mail: BBN, MS 6/5A, 10 Moulton St., Cambridge, MA 02138
           \     Telephone: (617) 873 3426    FAX: (617) 873 3776
            `----------------------------------------------------------------