X60 sudden battery death mystery

[twigdip]

Hi - my battery just died from one day to the next. It is frozen at 30% but the moment I remove the mains cable the laptop dies...so there is no electric connection between the battery and the laptop. Any idea why? It was working fine yesterday and the battery holds 2-3 hours charge no problem. I wonder if this is a software problem? But then why would the power be 30%?
Also - the battery isn't charging: it remains stuck at 30% despite the icon in taskbar saying plugged in, charging.

Many thanks for any help.

[geka3250]

I think it can be burned fuse - in battery controller or battery laptop circuit.
You can measure battery voltage at two edge terminals. Also you can check laptop power circuit with other working battery

[Cigarguy]

How long have you had this battery? This is typically how these batteries die. I've had a few that showed 60%, 70% or 80% but is dead. It will not charge and LED flash orange when trying to recharge. Nothing much you can do to repair it. Just replace.

[twigdip]

Thanks Cigarguy and Geka3250-
Reckon you're right, there's a problem with the battery - fused or sudden death for some reason. I've just ordered a replacement so hopefully I'll be able to prolong the life of this super X60 beast I've had for 9ish years now!

Thanks again

[Qing Dao]

I had the exact same thing happen to me in August with my T500. The battery was stuck at a certain % charge, and it wouldn't go up or down. When the power adapter was unplugged, the laptop died. A new battery fixed the problem.

[nakopla_plechovka]

hello

If it is an Lenovo original battery, this is the "sudden battery death" syndrome. There is some tec-blahblah behind it, but in general, the battery encuntered an internal error and refuses to cooperate to avoid any damage to the laptop and/or user. With My X61t battery was the same. I crack-open the case, discharged all the modules (to 0,1V), assembled back, jump start them using a 5V source and someway it works

Brgds,
M.

[pgoelz]

As another data point, the battery that came with a used X61S I just bought did the same thing. As received, it had zero charge remaining. It charged normally, although a little fast. It reached 99%, stayed there for a while and then suddenly the battery LED started flashing orange and battery manager declared it had "failed due to normal wear" and it would not charge.

It will power the laptop, although the charge level drops faster than normal. It will not charge.

I opened the battery and measured each cell. All four were at about 3.9V +/- 0.05V (with ~66% remaining on the battery monitor), so the problem was not an unbalanced cell (my first suspicion).

Power Manager shows 172 cycles, a full charge capacity of 14.27WH and a design capacity of something like 29WH. I'm wondering if it declares a failed battery when the full charge capacity drops to 50% of design capacity?

Since it will not charge, I can't play with it further although I may try to jump start it and if that gets it going again I'll try a battery gauge reset. But I suspect the battery is toast. I could replace the metal cased LiIon cells with LiPoly but I'm not sure it is worth it with non-OEM batteries available for not much more than the cells themselves.

Paul

[dr_st]

Power Manager shows 172 cycles, a full charge capacity of 14.27WH and a design capacity of something like 29WH. I'm wondering if it declares a failed battery when the full charge capacity drops to 50% of design capacity?

Not as far as I know. I have some old batteries with <50% capacity, and they work.

I believe that it is what was said by nakopla_plechovka. Something in the controller detects an error and locks the battery down. I seem to have heard of this behavior more often with X6x batteries than any other Thinkpad, but it is just a feeling.

As far as I know, it has not been yet determined what causes this failure, whether it's genuine or just an attempt to get you to buy a new battery, and whether there is a workaround that consistently works.

[pgoelz]

If the controller does not look at cell balance (mine are perfectly balanced) or cell capacity (mine are showing 50% of new capacity) then I can't see what else it would use to permanently kill the pack. In my case, it happened as I was looking at the laptop and after the charge level had reached 99%.

If anything, I would have expected it at the beginning of the charge cycle because the pack was showing 0% on the battery gauge so the cells were at an unknown voltage, possibly lower than 3V/cell. But it appeared to charge normally until it errored at the end of the charge.

The only other thing I can think of is that the cells WERE out of balance initially, the pack balanced then during charge, but the amount of balance effort required crossed a threshold at the end of the charge. Since I have the pack open, I may play with it (I fly electric model airplanes and am very familiar with lithium cells) and see if I can observe any issues during a controlled discharge. I can easily add a balance connector to it and charge it on my RC charger. That will allow me to measure cell balance and charge delivered as well as infer internal resistance.

Paul

[pgoelz]

UPDATE:

As an experiment I attached a balance cable to the cells and charged them with an external four cell RC charger. They were initially at about 55% capacity as reported by Power Manager and the external charger agreed with that figure when the charge began. The cells were all within about 10mV of each other at the beginning of the charge and the charger automatically balances to within 1mV during charge. The pack (initially at a 55% state of charge) took about 600mAH to reach full charge, which is roughly in line with a Power Manager reported capacity of about 50% of its 2AH design capacity.

I then discharged it to about 40% through an external 300mA load, bypassing the onboard coulomb counter electronics. The cells remained within 11mV of each other.

At this point the only issue I can see with the pack is that it has only about 50% of its original 2AH capacity remaining. I did not observe anything that should have triggered a failure mode.

Just for grins, I put the pack back together (still at about a 40% state of charge) and plugged it into the laptop. It reported zero state of charge, with a flashing orange LED. For more grins, I initiated a battery gauge reset. The LED remained orange flashing but the pack began charging. After a couple minutes the state of charge began rising from 0%, confirming it was indeed charging. To my surprise, when the state of charge passed 20%, the flashing orange LED turned green! It is still flashing, but it is not the continuous flash of a failed battery.... it is the brief off flash of a normally charging battery.

At present, it is on its way to 100%, at which point the battery gauge reset routine should switch to discharge to complete the reset. I'll update this post when all is complete. But for now, it looks like a battery gauge reset can reset a spurious failed battery indication?

Oh, and FWIW, this battery shows it was manufactured in May of 2006 and first used in March of 2007. So if I can't actually revive it, it has had a decent life

Paul

[dr_st]

Great detective engineering work and very interesting.

So I understand that the message "failed due to normal wear" is gone and the battery appears to be functional.

Do you have a theory as to which of the things you did is the likely cause of the fix?

[pgoelz]

Great detective engineering work and very interesting.

So I understand that the message "failed due to normal wear" is gone and the battery appears to be functional.

Yes, it appears as if it has reverted to normal operation. It showed 58% and charging when I walked away for about 15 minutes. At that point, it was actually at 100% based on voltage, but the coulomb counter was out of synch because I had externally charged and discharged it, bypassing the coulomb counter. The charge current as reported by Power Manager could be seen to be cycling on and off and the voltage showed 16.72V or thereabouts. When I returned, it had apparently decided it was fully charged and switched to discharge mode. The battery gauge reset is still in progress as I write this so I cannot be 100% sure yet. It is currently at the end of the discharge cycle, showing 5% remaining with a continuous flashing orange battery LED. Since Power Manager is still reporting the battery is OK, I am for now assuming the orange flashing LED indicates the battery is almost fully discharged. We'll see shortly.

Do you have a theory as to which of the things you did is the likely cause of the fix?

I have several, but have not done further detective work to figure out which is the most likely.....

Theories:

1. The internal battery monitor has an algorithm that declares a failed battery after X number of cycles and/or X number of years and/or X amount of remaining capacity.

2. The internal battery monitor is capable of cell balancing during charge and declares a failed battery if one or more cells are more than X mV out of balance at full charge. This condition would be only reached if a cell was badly degraded vs. its neighbors.

3. The internal battery monitor IS NOT capable of cell balancing during charge and only monitors individual cells during series charging. It declares a failed battery if one or more cells are more than X mV out of balance. This condition could be reached if a cell was only mildly degraded vs. its neighbors.

As I was writing the above, the battery gauge reset discharge cycle completed with between 0% and 5% showing on the battery gauge (I wasn't looking at it when it switched). At that point, it began charging. The battery LED was flashing orange at the end of the discharge and remained that way until it reached 10%. At 10%, it switched to steady orange with a brief "off" blink. At 20%, it turned green with the same brief "off" blink. So far, I believe this is all normal behavior.

Power Manager is showing that it is charging at 1.36A and will take 56 minutes to charge from its current level of 43% to 100%. That would be roughly the full 2AH capacity, so I am waiting with bated breath to see what happens next.

More news as it happens.....

Paul

[pgoelz]

Battery gauge reset completed successfully. It still shows 14.85WH, down from the original 28.8WH. But it charged all the way to 100% and no longer shows a failed battery.

I am not sure what to conclude from this. I think I tried a battery gauge reset after the battery reported failed and the reset simply aborted but I am not 100% certain. But this time, after charging (and accurately balancing) the pack with an external charger, the gauge reset completed and power manager optimistically reports the battery is in "good" condition.

Not sure how much more energy I want to spend, but it might be interesting to discharge to around 50%, then intentionally unbalance the pack by about 100mV and see what happens on the next charge cycle. Any charger that can balance cells should be able to handle a 100mV unbalance. But 100mV would likely be sufficient to generate an error if the internal battery monitor can't actually balance the cells.

In any event, the battery is very old and tired and probably not worth saving.

Paul

[rkawakami]

Just a wild bit of conjecture... the battery pack charge controller might also take into account the temperature of the pack during charging. If the thermistor(s) indicates that the cell(s) exceed some predefined limit, that could also set a flag which tells the laptop that the pack is "dead".

[pgoelz]

Definitely possible. But in this case, the cells are fine and do not get hot. The thermistor is also OK and was correctly attached to a cell when I opened the pack.

My best guess at this point is that the charge circuitry either has no balance capability or perhaps its balance capability is limited. As the cells age at slightly different rates, at some point the resulting imbalance exceeds the balance tolerance of the charge circuit and it declares an error. This should be easy to test. Discharge one cell 100mV below the others, then charge the pack and observe if the imbalance decreases during charge.

The only thing that doesn't make sense here is that the pack was well balanced when I opened it after it was declared "failed". But it is possible it sat for an extended period (it was in a used X61S and arrived at zero charge) and the tired cells drifted too far out of balance. During charge, the pack attempted to balance itself but could not quite get there.

It is currently still working correctly and has been through one discharge and recharge cycle without incident.

Paul