Re: Toshiba T1910CS power problem

[Elizabeth Barham <lizzy@soggytrousers.net>]

Yves writes:

> On Wed, Sep 03, 2003 at 05:35:51PM -0500, Elizabeth Barham
> wrote:
> > The theory here is that because the old CMOS battery was
> > not plugged in for sometime and then plugged in, the
> > computer sensed it had some electricity and so booted-up.
> > But after a little while plugged in, the computer drained
> > the CMOS' battery and so it behaved as before.
> 
> While having nothing to add to other people's suggestions, I don't
> think this theory is a valid one; CMOS battery is really only used
> to power the real time clock (to keep the time on the computer when
> no other power source is available). It's not normally used for
> anything else, let alone power a full computer. In particular,
> starting up a hard drive requires currents that a CMOS battery would
> be unable to deliver.

There are two internal batteries:

	*) CMOS/NiMH battery which saves state (it can go into a
	   resume mode when the operator closes the unit [folds the
	   screen up] and when she re-opens it. During this
	   intermission, the state of the machine is as it was
	   before).

	*) RTC/Li Battery. This is the tricky one to reach and I
	   recently obtained a new one.

The symptoms of my particular machine have been rather peculiar. As it
is, when I hit "power on" the machine turns out for a very brief
period of time and then goes down. If I keep hitting power-on for
about ten to fifteen minutes it will start. This seems to me that
something is charging up a little bit each time it tries to start and
after so many times there is enough power somewhere to allow the
system to come up.

When it first powers-up, it most likely does a simply hardware
check. One of these checks may be a determination if the CMOS battery
has much power in it. If it does not then it turns off and causes the
green DC-IN light to flash.

So, the evidence suggesting it may be the CMOS/NiMH green battery are:

	1) It appears as if something is charging up between quick
	   power-on/power-off cycles (the 10-15 minute "keep pressing
	   power" thing). Perhaps its a capacitor? If so its soldered
	   to the board so I'd probably trash the machine. OTOH, a
	   rechargeable battery does this sort of thing as its main
	   job plus its easily replaceable.

	2) After disconnecting the CMOS/NiMH battery for an elongated
	   time and noticing no change in behavior, upon reconnecting
	   it the machine behaved differently and seemed to power up
	   as it should but I could not verify this as the screen was
	   not hooked up.

I didn't mean to imply that the CMOS/NiMH battery powers the whole
thing but rather because the battery has so little storage ability
and/or has so little power that the machine refuses (as in a bad
self-test) to power-on fully.

So, sometime in the future I may purchase a new CMOS battery and see
how well it works.

Thanks for your contribution,
Elizabeth