Resetting my laptop's battery?

Well, yes and no.

If the battery cells are truly bad, then no software will reset any bad
cells.

BUT .... many batteries have an intelligent battery controller chip in
them, quite often a BQ2092 or a BQ2040. These chips monitor the battery
and control charging, their purpose is to prevent excessive temperature
and overcharging, as lithium batteries can become dangerously explosive
when improperly managed.

Unfortunately, sometimes under a variety of circumstances, they can come
to believe that the battery is bad (won't take a full charge) when it
really isn't. However, their control over the internals of the battery
and the individual cells is so complete that if they come to believe
that the battery is bad (even when it's not), then they won't let you
fully charge the battery.

There are programs to reset these chips in these situations, but the
process is really messy in most cases. Usually, you have to open up the
battery, make a special cable to connect the battery controller PCB to
the parallel port, and also open the circuit of the cells, to remove
power from the battery controller (forcing it to reset itself).

There have been detailed articles published on the net on how to do
this, do a google search on BQ2092 and BQ2040. There are other battery
controllers as well, and some laptops have the battery controller in the
laptop rather than in the battery. Finally, as noted, if the cells
really are bad, nothing is going to reset them, the description I've
given describes a fairly special (although not all that uncommon) case.


Lawrence Glasser wrote:

> "David B." wrote:
>
>>There is no magic program to reset a bad battery, when it's bad, it's bad, period.
>
>
> Hallelujah! Somebody with a brain!
>
> Larry

Well, yes and no.

If the battery cells are truly bad, then no software will reset any bad
cells.

BUT .... many batteries have an intelligent battery controller chip in
them, quite often a BQ2092 or a BQ2040. These chips monitor the battery
and control charging, their purpose is to prevent excessive temperature
and overcharging, as lithium batteries can become dangerously explosive
when improperly managed.

Unfortunately, sometimes under a variety of circumstances, they can come
to believe that the battery is bad (won't take a full charge) when it
really isn't. However, their control over the internals of the battery
and the individual cells is so complete that if they come to believe
that the battery is bad (even when it's not), then they won't let you
fully charge the battery.

There are programs to reset these chips in these situations, but the
process is really messy in most cases. Usually, you have to open up the
battery, make a special cable to connect the battery controller PCB to
the parallel port, and also open the circuit of the cells, to remove
power from the battery controller (forcing it to reset itself).

There have been detailed articles published on the net on how to do
this, do a google search on BQ2092 and BQ2040. There are other battery
controllers as well, and some laptops have the battery controller in the
laptop rather than in the battery. Finally, as noted, if the cells
really are bad, nothing is going to reset them, the description I've
given describes a fairly special (although not all that uncommon) case.


Lawrence Glasser wrote:

> "David B." wrote:
>
>>There is no magic program to reset a bad battery, when it's bad, it's bad, period.
>
>
> Hallelujah! Somebody with a brain!
>
> Larry

Barry Watzman wrote:
>
> Well, yes and no.
>
> If the battery cells are truly bad, then no software will reset any bad
> cells.
>
> BUT .... many batteries have an intelligent battery controller chip in
> them, quite often a BQ2092 or a BQ2040. These chips monitor the battery
> and control charging, their purpose is to prevent excessive temperature
> and overcharging, as lithium batteries can become dangerously explosive
> when improperly managed.
>
> <snip>

Thank you for the education... Greatly appreciated!

But, because of the age of Dell's CPi series, one would be lead to
believe that the battery (Not "batteries," so this was probably the
original owner's only one.) is just plain old.

Is there any documentation as to what age, in the battery's life,
these chips usually fail? Can it happen at any time? Does it happen,
in reality, at any time, or does it usually show up, early, as a
defective battery?

Again, thanks!

Larry

Lawrence Glasser wrote:

>
> But, because of the age of Dell's CPi series, one would be lead to
> believe that the battery (Not "batteries," so this was probably the
> original owner's only one.) is just plain old.
snip
> Larry

This is my serious pet peeve with web helpers...sweeping conclusions
based on NO actual information for the specific case under discussion.
While comments on AVERAGE situations can be useful, there's no
substitute for suggestons pertinent to the UNIQUE situation at hand.
mike

--
Bunch of stuff For Sale and Wanted at the link below.
laptops and parts Test Equipment
4in/400Wout ham linear amp.
Honda CB-125S
400cc Dirt Bike 2003 miles $550
Police Scanner, Color LCD overhead projector
Tek 2465 $800, ham radio, 30pS pulser
Tektronix Concept Books, spot welding head...
http://www.geocities.com/SiliconValley/Monitor/4710/

Normally, the battery controller chips don't fail. And the situation
that I described is not really a "failure", but rather case of the
battery controller chip somehow getting false information or becoming
"out of sync" with the actual status of the battery cells. It doesn't
happen often, but it does happen "often enough".

Properly treated, Lithium batteries can last the better part of a
decade, I have a number of Toshiba 2487 laptop batteries from 1997 that
are still going strong (and that battery model is still in production
and use in current model Toshibas, although in it's latest incarnation,
they have increased the rated capacity to 4500mah and on that model it's
called a 3107 rather than a 2487, but it's really essentially the same
battery.



Lawrence Glasser wrote:

> Barry Watzman wrote:
>
>>Well, yes and no.
>>
>>If the battery cells are truly bad, then no software will reset any bad
>>cells.
>>
>>BUT .... many batteries have an intelligent battery controller chip in
>>them, quite often a BQ2092 or a BQ2040. These chips monitor the battery
>>and control charging, their purpose is to prevent excessive temperature
>>and overcharging, as lithium batteries can become dangerously explosive
>>when improperly managed.
>>
>><snip>
>
>
> Thank you for the education... Greatly appreciated!
>
> But, because of the age of Dell's CPi series, one would be lead to
> believe that the battery (Not "batteries," so this was probably the
> original owner's only one.) is just plain old.
>
> Is there any documentation as to what age, in the battery's life,
> these chips usually fail? Can it happen at any time? Does it happen,
> in reality, at any time, or does it usually show up, early, as a
> defective battery?
>
> Again, thanks!
>
> Larry

Barry Watzman wrote:
>
> Normally, the battery controller chips don't fail. And the situation
> that I described is not really a "failure", but rather case of the
> battery controller chip somehow getting false information or becoming
> "out of sync" with the actual status of the battery cells. It doesn't
> happen often, but it does happen "often enough".
>
> Properly treated, Lithium batteries can last the better part of a
> decade, I have a number of Toshiba 2487 laptop batteries from 1997 that
> are still going strong (and that battery model is still in production
> and use in current model Toshibas, although in it's latest incarnation,
> they have increased the rated capacity to 4500mah and on that model it's
> called a 3107 rather than a 2487, but it's really essentially the same
> battery.

My understanding of lithium batteries is that they have a "limited"
number of charging cycles available, before degradation. E.g., the
number might be "200." The battery will accept 200 charges, whether
or not it was depleted partially or completely before each charge.
After that, it's all downhill.

Long shelf life, no memory effect, just limited charges.

Larry

That's basically correct, but the limit is not as "hard" as your post
might be read to imply. By that I mean that the number of cycles that
you varies widely from battery to battery, and that the loss of capacity
with additional charge-discharge cycles is very gradual and is not a
sharp "cliff" where the battery is good after 347 charge cycles and dead
as a doorknob after the 348th cycle. The number of cycles depends on
the exact history of the particular battery, the variable include age,
temperature over it's entire life and both the number of
charge/discharge cycles and also the nature of the cycles --
low-current, high current, and so on. The typical range on the number
of charge cycles that you get is probably a higher than your example
figure of 200, but it's in the range of hundreds. The one point where
I'd differ is that the effect of a full discharge-charge cycle is not
the same as that of a partial cycle.

For maximum life, keep the temperature down and don't "deep discharge"
the battery, discharging it below about 25%-30% should be avoided if
possible. Overcharging (continuing to charge after the battery is fully
charged) is also a killer (in part because it produces unnecessary
internal heating).



Lawrence Glasser wrote:

> Barry Watzman wrote:
>
>>Normally, the battery controller chips don't fail. And the situation
>>that I described is not really a "failure", but rather case of the
>>battery controller chip somehow getting false information or becoming
>>"out of sync" with the actual status of the battery cells. It doesn't
>>happen often, but it does happen "often enough".
>>
>>Properly treated, Lithium batteries can last the better part of a
>>decade, I have a number of Toshiba 2487 laptop batteries from 1997 that
>>are still going strong (and that battery model is still in production
>>and use in current model Toshibas, although in it's latest incarnation,
>>they have increased the rated capacity to 4500mah and on that model it's
>>called a 3107 rather than a 2487, but it's really essentially the same
>>battery.
>
>
> My understanding of lithium batteries is that they have a "limited"
> number of charging cycles available, before degradation. E.g., the
> number might be "200." The battery will accept 200 charges, whether
> or not it was depleted partially or completely before each charge.
> After that, it's all downhill.
>
> Long shelf life, no memory effect, just limited charges.
>
> Larry

Barry Watzman wrote:
>
> That's basically correct, but the limit is not as "hard" as your post
> might be read to imply. By that I mean that the number of cycles that
> you varies widely from battery to battery, and that the loss of capacity
> with additional charge-discharge cycles is very gradual and is not a
> sharp "cliff" where the battery is good after 347 charge cycles and dead
> as a doorknob after the 348th cycle. The number of cycles depends on
> the exact history of the particular battery, the variable include age,
> temperature over it's entire life and both the number of
> charge/discharge cycles and also the nature of the cycles --
> low-current, high current, and so on. The typical range on the number
> of charge cycles that you get is probably a higher than your example
> figure of 200, but it's in the range of hundreds. The one point where
> I'd differ is that the effect of a full discharge-charge cycle is not
> the same as that of a partial cycle.
>
> For maximum life, keep the temperature down and don't "deep discharge"
> the battery, discharging it below about 25%-30% should be avoided if
> possible. Overcharging (continuing to charge after the battery is fully
> charged) is also a killer (in part because it produces unnecessary
> internal heating).

Once again, thanks for taking the time for the education!

Larry

>> There is no magic program to reset a bad battery, when it's bad, it's
>> bad, period.
>
> Sometimes GOOD batteries get out of sync with the battery gauge chip
> and ACT like they're bad.
>

The battery controller (AKA the gas-gauge) hardly ever goes nuts all by itself.
You'd have to accidentally short the terminals or something.
More likely, there's something wrong with the cells, so that the pack sags
under load - or the capacity is just gone due to age.
Normally, resetting the battery makes sense only after replacing the cells.
If there's a defective cell in the pack, it's not likely to self-heal.

In this particular case (old battery), my guess is that the cells are just
worn. If replacing the cells is not an option in your area, you need to
get a whole new battery pack. If you manage to get the cells replaced,
a reset of the pack's chip is due - caveat, not all gas gauge chips can
be reset. With some, it may take quite some hacking, and with some
of the newer chips it's just plain impossible (on-chip flash + write-protect lock).


The gas gauge learns battery capacity upon full charge cycles (judged by
voltage tresholds). It's more willing to re-learn down than up - e.g., it may
be bounded to re-learn up by max. 20 per cent in a single charge cycle.
It's usually also hard-wired to refuse to re-learn upward, once the last
learned capacity reaches a particular "battery dead" lower bound, say
25% of the design capacity.

Thus, if you overload (or even short) the battery pack, you may make
the gas gauge think that the battery cells are completely worn.
That's where a reset may help.
Also, if the cells really are worn and you manage to replace them, you have
yet to reset the chip to remove the "battery dead" alarm status flag.

I've noticed that, as the battery cells are ageing (collecting charge cycles),
they start to tend to sag under load.
Their equivalent internal series resistance is gradually growing. Thus, the
battery is able to deliver its full capacity, but only over a longer period
of discharge. The point is that the notebook drains the battery in two
hours or so, which is quite fast. Thus, due to the current drain combined
with the gradually growing series resistance, the battery controller's
over-discharge alarm may fire even though the battery is not really
empty. And the gas gauge re-learns to a lower capacity.

The only solution is to replace the cells or buy a whole new battery.
A reset may help temporarily, until on another occasion (disk activity
peak or something) the cells sag once again.
This problem may be getting worse as the cell manufacturers are
trying to stuff ever more capacity into the space available - higher
capacity often means higher series resistance and lower recommended
current handling. Some manufacturers and battery models are known
to perform better in this respect than others.

Imagine that you get a brand new battery upgrade for your notebook,
with nominal capacity 50% higher than the original - but the magic only
lasts for a few months, then you get sudden shutdowns at "50% gas left
in the tank..."
Okay, seems like I've wandered off topic :-)

Frank Rysanek