X301 battery experiment

[JoeHallenbeck]

I read a lot of threads and opinions from users about batteries and their life expectancy. To sum all attributes whether to buy genuine or after-market battery I came to this conclusion:

New lenovo x301 6 cell battery costs about £150 (USD 225) postage and VAT included.
After-market (e-bay) x301 6 cell battery costs about £24 (USD 36) postage and VAT included.

Life expectancy of genuine battery lets say average 18 months
Life expectancy of after-market battery lets say average of 10 months (and I'm being really critical about this value)

After 18 months I loose value of roughly 6 after-market batteries
6 after-market batteries life expectancy TOTAL 60 months!

What if after-market battery will die after 3 months or next week?
There is still much more options to choose from, producer or seller vise, so I can buy another freely.
There is only ONE genuine producer seller, literally limiting any options at all. (outlet sales excluded because of shelf degradation)

Would you abuse your original genuine battery for test? NO
Would you abuse your after-market brand new battery to check which is better in case it fails HELL YEAH!

My current original lenovo genuine X301 battery is near its life end. Cycle count 169, type SANYO (worst of all) charge capacity roughly half size 21.98Wh of original 43.20Wh, since I bought the laptop on ebay I don't know past usage of this battery. First usage date is 2009 11 meaning it was good for more than 3 years (still less than average of 60 months - 5 years of after-market ones). It gives me about 1.5 hours of usage on power source optimized power plan, maximum brightness, radio stream, writing this post, looking for CPU upgrade for my SL500 and so on...

The thing is that YES it is still alive, YES it is not failed one BUT one and half hours is nothing, you are not able to do much. And will be eventually less and less.

Yesterday I did little experiment. I canceled all warning actions of battery after it will reach its 5% level. I think many of you know that you can run battery on zero or one percent if it is in relatively good shape. (DO NOT TRY THIS AT HOME!) My one gave me another ONE HOUR and 10 minutes of run when I decided (NO I DIDN'T LET IT DIE!) to plug in charger when voltage decreased from roughly 12-ish voltage when charged to designed charge of 10.08V. Power manager went crazy using CPU for while (about 20 minutes or so) until it reached it's "imaginary" 5% of charge and gauge started showing that percentage of charge is increasing.

Now writing this post I'm repeating same experiment again. The thing is that, it would be perfect if it is relatively safe to use this battery like this but change or edit power manager's 5% deadline and update somehow or edit it to add missing one hour and 10 minutes.

I was thinking about power manager gauge reset but how? I read that this makes things worse. This battery gives me "real" more than 2.5 hours instead of 1.5. For now it seems that after low battery alarm I have to just watch clock or launch timed hibernation manually to prevent data loss.

Any thoughts would be appreciated.

[Cigarguy]

Sounds to me like your battery is in desperate need of recalibration. Pretty common with Lith-Ion. I don't use Power Manager to recalibrate. To recalibrate, I'll charge it up, then let it drain completely. When it reaches 5% Windows will hibernation. I'll turn it back on and let battery drain some more. When it shuts down again, I'll turn it on hit F1 to go into BIOS then let it drain some more. Finally when it shut down again I'll plug in the power cord and leave it charging for at least 3-4 hours.

[600X]

What you are doing is basically resetting the battery. (but the totally wrong way) Let it reach 0% and completely empty the battery so that it shuts down. Repeat this a few times. A modern OS will not carry away any damage from this.

My X301 has an original 6c as well, it's 4 years old and still holds 39Wh out of 43Wh with a cycle count of 320. It's a Sanyo. I have done multiple recalibrations after receiving my unit. At that time it had 230 cycles and 32Wh left.

I'd say these batteries do last approximately 5 years, if you treat them right.

[JoeHallenbeck]

Ok I'll try it again and let it run until it will switch it off by it self. Do I have to charge it back to 100% turned off or with computer running?

Will the zero turn off damage in any way my SSD? Did you manged to increase your wattage back to higher amounts by this? I have to try it on battery which is suddenly lowering its charge level during discharge but I guess it is not gonna help it or?

Thanks

[Cigarguy]

Here's my experience. While draining the battery for purposes of resetting it, it's best to let it idle down till Windows 7 hibernate (@5% capacity) the machine. Turn machine back on, let it drain to zero. When the battery is truly exhausted it'll turn itself off. After that I'll plug it back in and let it sit for about 4-5 hours. When recharging, the key thing is to have it continuously charge until 100%. Doesn't matter if you are using the machine or have it off like I do.

As is typical with Lith-Ion batteries, the charging cycle is not linear. Initial charge till about 90% is pretty fast then it'll charge slower beyond that. When Windows or Power Manager reports "no activity" or "not charging", then you know you are there. I use HWMonitor to monitor mWh. I love HWMonitor and use it all the time.

As far as what you'll end up with after this will depend on the state of the battery cells. The resetting/calibration is calibration of the firmware electronics in the battery. Good cells will get you more run time. Bad cells are bad cells, nothing much you can do about it. Time, heat and usage are not good for these batteries.

Using HWMonitor to check battery capacity, I find that I need to cycle a new battery about 3-4 times to get to design capacity. As for old or bad battery, there's not much that can really be done.

Doing above is hard on a battery. I only do it every 30-50 cycles or when I suspect the capacity is off and needs calibration.

[600X]

Ok I'll try it again and let it run until it will switch it off by it self. Do I have to charge it back to 100% turned off or with computer running?

Will the zero turn off damage in any way my SSD? Did you manged to increase your wattage back to higher amounts by this? I have to try it on battery which is suddenly lowering its charge level during discharge but I guess it is not gonna help it or?

Thanks

In theory, sudden power loss to an SSD could kill it instantly. However, I have not had any problems so far. The best I have been able to do was get a 20Wh battery back to 40Wh. Like I stated above, I got my X301 battery from 32Wh to 39Wh.

After the computer shuts itself off you will need to plug it in and leave it for a few hours until the battery is fully charged. Best to just leave it overnight.

[JoeHallenbeck]

I decided to protect my SDD against sudden power failure by letting the battery go to the first 5% level and then power down machine, remove it from bay and start it without SSD to last screen when it is asking for installation disk or SDD to boot from. Then I will let it run until death.
I hope it will not under-voltage the battery.

[JoeHallenbeck]

... so far so good...

Manged to get back energy storage from 21Wh to 37Wh but still looking for maximum in third experiment. Just to note that windows will hibernate the OS whatever the settings in the power manager to save data, even when you set it to "no action" after reaching battery reserve level. Do the OS "somehow" know that there is hidden (true) battery level?

Just in case for sure I always pulled SSD out and left it in boot part after bios to die.

[Cigarguy]

Windows have a remaining capacity calculator. Probably just reading the info from the firmware on the battery.

I have about 12 laptops with SSD. I never thought of pulling the SSD while doing this. So far I haven't experience any problems.

You can also hit F1 to go into BIOS and wait for the battery to drain down.

[JoeHallenbeck]

Check this very interesting article about SSD behavior during power failure from Ohio State University. There is 14 pages pdf detailing tests and results included. It is relatively recent article but nevertheless I always backup data on classic spin disk and that is why I'm about to purchase bay adapter for my backup drive to replace CD combo at least once a month for simple copy and paste backup.

http://www.zdnet.com/how-ssd-power-faul ... 000011979/

Same suspicion applies on running machine in BIOS first page until battery will die. I don't know how power failure can affect running BIOS. Frankly I don't even want to know. Just for sure I'm calibrating battery by safest possible method.

Looking forward for tomorrow fourth run down on this battery to "bring her back to life"...

[Pete B]

Trying to understand how battery calibration works and if I read this thread correctly
it doesn't matter if Power Manager, or the user runs the battery all the way down,
either will cause the battery firmware to get a new measure of the real battery
capacity - is that right?
And everything pertaining to battery capacity is done in the battery firmware without
interaction with the OS or Power Manager?

I understand the concepts here just never was sure about what is done in the battery
or externally from it.

Also, my questions are regarding T61 batteries, safe to assume they all calibrate in
the same way?

Thanks

[brchan]

I have fully discharged different Thinkpad and Panasonic Toughbook batteries (as well as my phone) by accident, and one of the following happens:

1. No change in battery capacity (usually if only fully discharged a couple of times)
2. Killing one of the cells, causing capacity to be ~%15 worse, depending on # of cells (usually fully discharging > 3 times or more)
3. Bricking the entire battery (more random, but usually after > 3 full discharges)

I have never seen a capacity increase after fully discharging lithium ion batteries. IMO, don't fully discharge lithium batteries unless they are already nearly dead. In this case, there isn't much to lose.

[Pete B]

Interesting, do you know of a correct way to "recalibrate" a battery, if there is one?

[brchan]

Interesting, do you know of a correct way to "recalibrate" a battery?

Other than the full discharge and then recharge, no.

I am not sure how this method actually helps, since lithium batteries do not accumulate "cell memory" like NiCad batteries. IIRC they also have protection circuits that do not allow the batteries to be recharged if the voltage becomes lower than 0 (fully discharging can cause polarities to reverse).

[Pete B]

I don't think it has anything to do with changing the actual battery capacity,
rather it provides conditions that the controller needs to get a much better
estimate of the capacity.

[Pete B]

I tried this with my Dell D830 and now it is VERY slow to shut down and
early on in the boot process, then it seems to go back to normal. Nothing
else changed, I suppose it is possible that a Win update snuck in but I don't
think so.

Ran CPU-Z and the processor does run at full speed once booted.

Any suggestions before I start reading and digging into log files?