With a couple of D-cells, I've confirmed that 3.0-3.2 volts is enough to
sustain the CMOS settings across a shutdown. That setup is unwieldy and
not up for donation, so I'd like to use some heat shrink tubing to put a
couple of AAs together and solder on some leads. (There's already a
pair of leads coming off the board with a pin header jumper on the end.)
Will I explode something if I try to solder leads onto an AA battery?
What can I do to make the solder stick?
Is there a better way to do what I'm attempting without having to spend
another dime? (I already have the crap I'm proposing to use.)
To preempt a possible answer to that last question: Yes, I know I should
drive by a construction dumpster late at night and "solve" the problem
with one heave, but I already have my soldering iron hot.
On Sat, 16 Aug 2008 04:20:21 GMT, Grinder
<grinder@no.spam.maam.com> wrote:
>Another question from the front lines of PC turd polishing:
>
>I've cut yet another crusty barrel-type 3.6V battery out of a '486
>board. It was too crudoded to get specifics on, but it's the ones that
>look like:
>
>http://www.dsmiller.com/Images/B-430PC2.GIF
>
>With a couple of D-cells, I've confirmed that 3.0-3.2 volts is enough to
>sustain the CMOS settings across a shutdown. That setup is unwieldy and
>not up for donation, so I'd like to use some heat shrink tubing to put a
>couple of AAs together and solder on some leads. (There's already a
>pair of leads coming off the board with a pin header jumper on the end.)
>
The voltage threshold depends on the specific RTC used.
Generally, 3 to 4 cells is a safer choice than 2.
>Will I explode something if I try to solder leads onto an AA battery?
In theory yes, though in practice it depends on the
specifics. Rechargable batteries have an insulator and
value which are potentially subject to failure if exposed to
too much heat. Normally they are spot-welded to try to, and
usually elimiante excessive heat. It is possible to solder
them and not have excessive heat, in fact I have done so,
but the far better, more conservative answer is to use a
pack either pre-welded so you don't have to, or a battery
holder with contacts for std, AAA or AA cells so you don't
have to risk it.
>
>What can I do to make the solder stick?
If you reallyt want to attempt this, do the following:
Put the cells in a frige so they are cold.
Take sand paper and rough-up the area you will solder.
Put a liberal amount of flux on the area.
pre-heat the soldering iron to a higher than normal temp
Have solder ready, heat the battery tab and tin it with the
soldler, not trying to do the whole soldering job at once.
Pre-tin the lead, wire, etc, you want to attach.
Put flux on the tinned battery terminal again, then heat the
wire and hold it against the tinned battery terminal with
the iron, until the solder flows then hold it still as with
any soldering job.
>
>Is there a better way to do what I'm attempting without having to spend
>another dime? (I already have the crap I'm proposing to use.)
Spending a dime has everything to do with what parts you
have on-hand.
>
>To preempt a possible answer to that last question: Yes, I know I should
>drive by a construction dumpster late at night and "solve" the problem
>with one heave, but I already have my soldering iron hot.
You should spend the $1 and get a battery holder for 3 or 4
cells. Solder it's leads to the mobo compatible
connector. Soldering directly to cells can work, but if
the system is worth the bother, you probably want to do it
right instead of risking suffering potential downtime from
something as simple as a battery failure.
Grinder wrote:
> Another question from the front lines of PC turd polishing:
>
> I've cut yet another crusty barrel-type 3.6V battery out of a '486
> board. It was too crudoded to get specifics on, but it's the ones that
> look like:
>
> http://www.dsmiller.com/Images/B-430PC2.GIF
>
> With a couple of D-cells, I've confirmed that 3.0-3.2 volts is enough to
> sustain the CMOS settings across a shutdown. That setup is unwieldy and
> not up for donation, so I'd like to use some heat shrink tubing to put a
> couple of AAs together and solder on some leads. (There's already a
> pair of leads coming off the board with a pin header jumper on the end.)
>
> Will I explode something if I try to solder leads onto an AA battery?
>
> What can I do to make the solder stick?
>
> Is there a better way to do what I'm attempting without having to spend
> another dime? (I already have the crap I'm proposing to use.)
>
> To preempt a possible answer to that last question: Yes, I know I should
> drive by a construction dumpster late at night and "solve" the problem
> with one heave, but I already have my soldering iron hot.
Is that a NiCAD ? That implies a constant charging current, when the computer
is running. What will happen to an alkaline under those conditions (application
of constant charging current) ? I've be a bit cautious about the application.
I was going to suggest adding a diode to the battery output, so the current
can only flow one way (so the battery never gets charged), but then that will
drop your terminal voltage to the CMOS circuit. (Maybe you could go to 4.5V
with three cells, and use two 0.7V drop forward biased diodes ?)
I've soldered to batteries. I have four C cells sitting in the kitchen, that
are connected together with stranded wire and solder. But the cells are
cheapies, and have steel contact plates on each end, making soldering
pretty easy. I've had other batteries, where all I managed, was a bunch of
cold solder joints. YMMV. No explosions yet.
Paul wrote:
> Grinder wrote:
>> Another question from the front lines of PC turd polishing:
>>
>> I've cut yet another crusty barrel-type 3.6V battery out of a '486
>> board. It was too crudoded to get specifics on, but it's the ones
>> that look like:
>>
>> http://www.dsmiller.com/Images/B-430PC2.GIF
>>
>> With a couple of D-cells, I've confirmed that 3.0-3.2 volts is enough
>> to sustain the CMOS settings across a shutdown. That setup is
>> unwieldy and not up for donation, so I'd like to use some heat shrink
>> tubing to put a couple of AAs together and solder on some leads.
>> (There's already a pair of leads coming off the board with a pin
>> header jumper on the end.)
>>
>> Will I explode something if I try to solder leads onto an AA battery?
>>
>> What can I do to make the solder stick?
>>
>> Is there a better way to do what I'm attempting without having to
>> spend another dime? (I already have the crap I'm proposing to use.)
>>
>> To preempt a possible answer to that last question: Yes, I know I
>> should drive by a construction dumpster late at night and "solve" the
>> problem with one heave, but I already have my soldering iron hot.
>
> Is that a NiCAD ?
It appears the the picture I linked is a NiCD, but I just picked the
picture because it showed the proper shape.
> That implies a constant charging current, when the computer
> is running. What will happen to an alkaline under those
> conditions (application of constant charging current) ?
> I've be a bit cautious about the application. I was going
> to suggest adding a diode to the battery output, so the
> current can only flow one way (so the battery never gets
> charged), but then that will drop your terminal voltage
> to the CMOS circuit. (Maybe you could go to 4.5V with three
> cells, and use two 0.7V drop forward biased diodes ?)
[snip]
The original cell was very crusty. The plastic wrapping it was almost
completely destroyed, and I did not dissect it to look for identifying
marks. The board, where the battery was soldered, is marked "DS12887."
That, apparently, is a RTC with an inbuilt Lithium power
source--clearly not what was actually on the board.
I can't find a specific reference to what battery was on the board, but
other boards from the same manufacturer in the same period used external
Lithium batteries.
Grinder wrote:
> Another question from the front lines of PC turd polishing:
>
> I've cut yet another crusty barrel-type 3.6V battery out of a '486
> board. It was too crudoded to get specifics on, but it's the ones that
> look like:
>
> http://www.dsmiller.com/Images/B-430PC2.GIF
>
> With a couple of D-cells, I've confirmed that 3.0-3.2 volts is enough to
> sustain the CMOS settings across a shutdown. That setup is unwieldy and
> not up for donation, so I'd like to use some heat shrink tubing to put a
> couple of AAs together and solder on some leads. (There's already a
> pair of leads coming off the board with a pin header jumper on the end.)
>
> Will I explode something if I try to solder leads onto an AA battery?
>
> What can I do to make the solder stick?
>
> Is there a better way to do what I'm attempting without having to spend
> another dime? (I already have the crap I'm proposing to use.)
>
> To preempt a possible answer to that last question: Yes, I know I should
> drive by a construction dumpster late at night and "solve" the problem
> with one heave, but I already have my soldering iron hot.
Probably won't explode yourself but the results may not be very
satisfactory. The materials used in dry cell manufacture just don't solder
very well -- that is the reason that makers of battery packs spot weld the
jumpers. You'd be way ahead if you just went out and bought an
old-fashioned battery holder instead.
Grinder wrote:
> Paul wrote:
>> Grinder wrote:
>>> Another question from the front lines of PC turd polishing:
>>>
>>> I've cut yet another crusty barrel-type 3.6V battery out of a '486
>>> board. It was too crudoded to get specifics on, but it's the ones
>>> that look like:
>>>
>>> http://www.dsmiller.com/Images/B-430PC2.GIF
>>>
>>> With a couple of D-cells, I've confirmed that 3.0-3.2 volts is enough
>>> to sustain the CMOS settings across a shutdown. That setup is
>>> unwieldy and not up for donation, so I'd like to use some heat shrink
>>> tubing to put a couple of AAs together and solder on some leads.
>>> (There's already a pair of leads coming off the board with a pin
>>> header jumper on the end.)
>>>
>>> Will I explode something if I try to solder leads onto an AA battery?
>>>
>>> What can I do to make the solder stick?
>>>
>>> Is there a better way to do what I'm attempting without having to
>>> spend another dime? (I already have the crap I'm proposing to use.)
>>>
>>> To preempt a possible answer to that last question: Yes, I know I
>>> should drive by a construction dumpster late at night and "solve" the
>>> problem with one heave, but I already have my soldering iron hot.
>>
>> Is that a NiCAD ?
>
> It appears the the picture I linked is a NiCD, but I just picked the
> picture because it showed the proper shape.
>
>> That implies a constant charging current, when the computer
>> is running. What will happen to an alkaline under those conditions
>> (application of constant charging current) ?
>> I've be a bit cautious about the application. I was going to suggest
>> adding a diode to the battery output, so the
>> current can only flow one way (so the battery never gets charged), but
>> then that will drop your terminal voltage to the CMOS circuit. (Maybe
>> you could go to 4.5V with three cells, and use two 0.7V drop forward
>> biased diodes ?)
> [snip]
>
> The original cell was very crusty. The plastic wrapping it was almost
> completely destroyed, and I did not dissect it to look for identifying
> marks. The board, where the battery was soldered, is marked "DS12887."
> That, apparently, is a RTC with an inbuilt Lithium power
> source--clearly not what was actually on the board.
>
> I can't find a specific reference to what battery was on the board, but
> other boards from the same manufacturer in the same period used external
> Lithium batteries.
What you have there is a Dallas Semiconductor clock with built-in battery.
A real antique -- I have a 486 MB in the basement which uses two DS parts.
On Sat, 16 Aug 2008 08:14:08 GMT, Grinder
<grinder@no.spam.maam.com> wrote:
>Paul wrote:
>> Grinder wrote:
>>> Another question from the front lines of PC turd polishing:
>>>
>>> I've cut yet another crusty barrel-type 3.6V battery out of a '486
>>> board. It was too crudoded to get specifics on, but it's the ones
>>> that look like:
>>>
>>> http://www.dsmiller.com/Images/B-430PC2.GIF
I checked a picture of a 486 board I used to have, it had a
Varta 3.6V, 60mAH NiCad. Next to the battery was a jumper
that presumably determined whether the power came from that
battery or an external battery pack which was connected by a
4 position, 3 pin header nearby. IIRC, the typical battery
pack for that was a 4 cell so it seemed able to tolerate at
least up to 6V as 4 x 1.5V.
>
>It appears the the picture I linked is a NiCD, but I just picked the
>picture because it showed the proper shape.
Yes, in that era it would've been NiCad.
>
>> That implies a constant charging current, when the computer
>> is running. What will happen to an alkaline under those
>> conditions (application of constant charging current) ?
I had assumed he meant a rechargable battery, an alkaline
should not be connected to it, would be damaged. There may
be a pin header elsewhere on the board to which an alkaline
could be connected and wouldn't be subjected to charging, or
as mentioned a diode could isolate it.
>> I've be a bit cautious about the application. I was going
>> to suggest adding a diode to the battery output, so the
>> current can only flow one way (so the battery never gets
>> charged), but then that will drop your terminal voltage
>> to the CMOS circuit. (Maybe you could go to 4.5V with three
>> cells, and use two 0.7V drop forward biased diodes ?)
>[snip]
There's no need to get the voltage down to a minimal level
IIRC, though the RTC chip spec sheet should detail the
voltage levels. The voltage does have to be below a certain
threshold for it to go into low power mode IIRC, which could
be fairly important so the battery pack capacity is not
exhausted inbetween times the system is turned on to
recharge the cells.
>
>The original cell was very crusty. The plastic wrapping it was almost
>completely destroyed, and I did not dissect it to look for identifying
>marks. The board, where the battery was soldered, is marked "DS12887."
> That, apparently, is a RTC with an inbuilt Lithium power
>source--clearly not what was actually on the board.
>
>I can't find a specific reference to what battery was on the board, but
>other boards from the same manufacturer in the same period used external
>Lithium batteries.
Lithium cells were generally seen in coin cell holders like
they are today. Soldered on packs were practically if not
completely all NiCad. NiMH would work too, minor
differences in their charging profiles wouldn't be relevant
at the low charge rate a motherboard uses.
On Sat, 16 Aug 2008 04:20:21 +0000, Grinder wrote:
> Will I explode something if I try to solder leads onto an AA battery?
nope, that is what I did, but I suggest liberal use of fine sandpaper for
metal on all ends before you start and a good capacity soldering iron.
Getting the solder to stickto the ends of the battery is hard first time.
> Is there a better way to do what I'm attempting without having to spend
> another dime? (I already have the crap I'm proposing to use.)
Go for it.
BTDT.
worst problem was sticking the batteries on the side of the power supply
and keeping them there.
If you can do it, make the leads 12-24" and just tape the bateries in
place on the bottom of the case(tower?).
Will I explode myself? 
Linear Mode
