Ubuntu 7.10 critically low battery -> dead battery

Message

spaznrq · #1 Post by **spaznrq** » Sat Mar 15, 2008 9:52 am

For those other newbs like me out there, be careful of running your Linux machine past the critically low battery state. My 8-cell just completely died after an abnormal shutdown due to low battery. I've checked my Power Management settings and it should've gone into Suspend state, but it didn't for some reason. When trying to recharge the battery now, it gives a blinking orange battery indicator. Booting up in Windows, Power Manger tells me the battery cannot be recharged and needs to be replaced.

I've had my x60 for about a year and 9 months so my battery's warranty was over by 9 months. My battery was neither part of the battery recall nor a known battery issue. This was just an expensive lesson to be learned, so I'm passing my experience to others here.

I got another 8-cell (for $116 :] ), so I'm back up and running again. This time, I'll be sure to be a little more careful.

aaa · #2 Post by **aaa** » Sat Mar 15, 2008 10:43 am

Maybe it was about to die to begin with?

My opinion on the cause is this: when a lithium cell's voltage goes too low, it might pass a "point of no return", especially if it's about to wear out. And it only takes one cell dying like this to bring the whole battery down.

So I think one of the cells was about to wear out anyways, and draining the battery all the way accelerated it's demise.

spaznrq · #3 Post by **spaznrq** » Sat Mar 15, 2008 1:11 pm

That is very possible.

My old 8-cell just showed no signs of deterioration just before it died. I was still getting a healthy 7 hours of battery life and Power Manager indicated a "good condition" battery. If it wasn't able to detect a "dying" cell, then what good is the battery condition indicator for?

Of course, we can speculate on many probable causes. The fact is, li-ion cells shouldn't be discharged completely since they like to live in the middle charged zone.

tarvoke · #4 Post by **tarvoke** » Sun Mar 16, 2008 8:38 am

very valid concern, thanks for pointing it out.

after upgrading to a new kernel, for a few days my x61 would refuse to suspend (sometimes it would not even shutdown...!) -- this sort of error could certainly allow a critically low battery to be completely drained.

similar sort of problem was seen in much earlier versions of rockbox on certain mp3 player platforms -- let the battery run down too far, you can't charge it up again.

actually, if you have a good power supply you could hook it up to, applying a bit over the normal charge voltage can often save it. just be careful.

lightweight · #5 Post by **lightweight** » Mon Mar 17, 2008 12:41 am

You should still investigate why your battery indicator/alarm failed to stay kind to the new battery. Perhaps hibernate/suspend is not configured. Worst case you could just have your system do a halt after some warning.

FYI, it's also easy to track your battery health over time. Maybe "present voltage" measuring under "design voltage" indicates a failing cell?

Code: Select all

(date; cat /proc/acpi/battery/BAT0/*) >> batterylog
cat batterylog

awong · #6 Post by **awong** » Mon Mar 17, 2008 1:29 am

I wonder if I did something similar to my battery on my r31, about 6 months old and let it drain down a few weeks ago, now my battery wont charge past a certain percentage. ended up draining by battery really fast b/c i forgot to change my bios settings for my cdrom and it was running at full speed/

spaznrq · #7 Post by **spaznrq** » Mon Mar 17, 2008 2:31 am

Thanks for the suggestion, lightweight. Here's what I get for the new battery:

Code: Select all

Mon Mar 17 00:30:58 PDT 2008
alarm:                   3744 mWh
present:                 yes
design capacity:         74880 mWh
last full capacity:      74880 mWh
battery technology:      rechargeable
design voltage:          14400 mV
design capacity warning: 3744 mWh
design capacity low:     200 mWh
capacity granularity 1:  1 mWh
capacity granularity 2:  1 mWh
model number:            93P5030
battery type:            LION
OEM info:                SONY
present:                 yes
capacity state:          ok
charging state:          charged
present rate:            0 mW
remaining capacity:      74620 mWh
present voltage:         16641 mV

And for the old battery;

Code: Select all

Mon Mar 17 00:32:24 PDT 2008
alarm:                   3744 mWh
present:                 yes
design capacity:         74880 mWh
last full capacity:      74880 mWh
battery technology:      rechargeable
design voltage:          14400 mV
design capacity warning: 3744 mWh
design capacity low:     200 mWh
capacity granularity 1:  1 mWh
capacity granularity 2:  1 mWh
model number:            93P5030
battery type:            LION
OEM info:                SONY
present:                 yes
capacity state:          ok
charging state:          charged
present rate:            0 mW
remaining capacity:      2970 mWh
present voltage:         15293 mV

It doesn't look like the bad battery's "present voltage" is lower than the "design voltage". I am doing both tests with the AC power plugged in, so the present voltage might not reflect what's happening to the battery. However, I am unable to test the old battery with it unplugged since, well, it's dead.

I don't see any other indication of a bad battery from this set of data, but this is a nice way to view battery information. Thanks.

lightweight · #8 Post by **lightweight** » Mon Mar 17, 2008 2:40 am

Yeah, remaining capacity is only 4% of your last charge. It appears as if it's just stopped charging, which goes with the failed cell theory, but maybe one of the contacts (or something) used for charging became loose or failed?

Voltage drops when unplugged for me -- it seems the AC adapter does screw with that. If you don't care about the old battery anymore and do not mind risk to the laptop, you might be able to buy more time to get a voltage sample with

Code: Select all

echo 0 |sudo tee /proc/acpi/battery/BAT0/alarm

scantech · #9 Post by **scantech** » Mon Mar 24, 2008 12:24 am

Hi - I had a similar problem with two batteries for my X60 that I tried to run battery maintenance / battery gauge reset on under Windows. They discharged OK, computer restarted and apparently it started recharging but after a while the battery indicator started flashing yellow.

Power manager reports about 90+% charged and "A battery error has occurred. The battery cannot be charged. Replace the battery.". Message Center reports "Battery error is detected.".

Since this happened almost simulatenously to two different batteries I started investigating and found a link to another Lenovo battery recall program - different from the one you linked to. The one that you linked to reports that my batteries aren't affected by any recall.

However THIS recall program detects BOTH of my batteries as being affected by a recall and provides a form where I have already ordered new replacements free of charge:

http://www-307.ibm.com/pc/support/site. ... 67765.html

ThinkPad battery will not charge or discharge and gives an "irreparable damage" or "battery cannot be charged" error message

tarvoke · #10 Post by **tarvoke** » Fri Mar 28, 2008 11:26 pm

also, maybe the controller somehow got some weird notion of charge thresholds into its head... did you try resetting the ec? -- remove the battery (and pull the ac as well) and see if that does anything.

independent · #11 Post by **independent** » Sat Apr 12, 2008 3:19 pm

There is a problem with lithium batteries comprised of several lithium cells and that is battery imbalance. If you look inside a battery you'll see wires coming off each pair of cells. That is for monitoring (or charging, not sure) individual pairs. RC enthusiasts who use these types of cell tech are very aware of this condition. So, my guess is you have an imbalanced battery. So, far so, that the battery monitoring circuit doesn't know how to fix it.

Ubuntu 7.10 critically low battery -> dead battery

Ubuntu 7.10 critically low battery -> dead battery

Who is online