Posted by ASAAR on 11/05/06 22:12

On Mon, 06 Nov 2006 01:17:21 -0800, SMS wrote:

>> After seeing that, I simply soldered 2 pieces of wire from the (+) and
>> (-) of the dead battery to the (+) and (-) of the main battery
>> terminals. The player works fine now.
>
> That's probably not a good idea. The lithium button cell is not designed
> to be charged, or to have voltage applied directly to it. On computers,
> there are two diodes that prevent the power supply +5V from trying to
> charge the CMOS/RTC coin cell battery (required by UL/CSA/TUV) because
> the Lithium battery can explode, and it's toxic.
>
> I've gone up against UL on this issue, where we had the protection
> circuitry inside the SuperIO where the CMOS and RTC resided . . .

You assume too much. Namely that it is was a lithium battery that
was shorted. I've programmed SuperIO chips back in the days of
IBM's PC and AT as well as OKI clock/calendar/CMOS chips and various
chips before them, and they generally used large lithium batteries,
fairly large lithium button cells or 9v alkalines. Today things are
different. Many cameras use small internal backup batteries.
They're usually soldered to circuit boards, but they're usually NOT
lithium, but small NiCd button cell batteries instead, intended to
be trickle charged by the device's main replaceable battery. By
closely examining the button cell or its immediately surrounding
circuitry it should be possible to determine if the one in the mp3
player is a lithium or NiCd battery. If the latter, it might be
possible to replace it with a small supercapacitor if a similar
battery can't be found.


> What's strange is that the button cell would go dead. It's purely for
> back-up when the regular battery is completely discharged (or removed
> for replacement). The battery life of the button cell should be at least
> 10 years.

We can probably assume that Sandisk didn't buy its SDMX1-512's
batteries from eBay, but any component can fail, even mil-spec
parts. The battery may or may not be the defective component. It
could be that the battery is being discharged by some dirt or other
contaminant near the battery, which is acting as a fairly high
resistance, but low enough to discharge the battery after enough
months go by. Or a short circuit in another component near the
button cell could have the same effect. Unless the button cell can
be removed from the circuit, it won't be very easy to tell if it is
the defective part or if something else is the cause of the problem.

[Back to original message]