Battery Loss Rate

[ranjit]

Hi All,

I have lost approximately 3% of my 6-cell battery (Sanyo) in exactly two months of average use. The status of my battery right now is:

Full Charge Capacity: 46.10 Wh
Remaining Capacity: 46.10 Wh
Design Capacity: 47.52Wh
Cycle Count: 44

Please let me know if the loss rate is normal.

Cheers and best regards,
Ranjit

[Flightvector]

Unfortunately this drainage that you cite is most likely to be normal; Lithium Ion batteries are by far the most efficient sources of portable electricity, producing a higher energy per unit mass than anything else in its price range. These types of batteries are also minimally affected by partial charging cycles. However, these benefits come at a penalty, and this lies in the constant degradation in charge potential. Irregardless of usage, there is a basic function of loss in maximal charge potential over time due to the chemical composition of the battery. A loss of 3% in 2 months is definitely likely since these types of batteries will usually lose over 10% in a year's time (closer to 20% of charge potential loss when considering a typical pattern of light usage and practical storage conditions).

My battery is at 40.19 W hr after 7 months of usage and 50 battery cycles. Admittedly, I don't take care of my battery as well as I should, but I usually try to store my battery at 40% whenever it is in a prolonged period of disuse.

[z.entropic]

If I may slightly augment your otherwise helpful and correct post, the term is not 'battery potential', but 'battery energy'. It is true that the definition of energy is the potential/capability to do useful work, but...

For those interested, the term 'potential' is used in electrochemistry only when defining a given electrode's potential vs. the potential of some standard reference electrode, for example, metallic lithium, i.e., vs. a Li/Li+ couple. Potential as such is undefined as it is always a relative quantity. The best known reference potential is that of a hydrogen electrode (hydrogen gas bubbling in a 1 molar aqueous acid around a piece of platinum foil or wire;its potential is assumed as 0.000 volts), but there are more practical and easy to use reference electrrodes.

What we read at the battery terminals is the difference of two electrode potentials vs. the same reference; this difference of two potentials is called 'battery voltage'.

To add a bit more info, the capacity fade rate a typical Li-ion battery usually decreases with time, and is a strong function of temperature. For example, you may lose 3% during the first month of use, but only 2 or 2.5% the next month and so on. Unfortunately, the internal impedance (resistance) of a battery always increases with use which will cause the average discharge voltage at the same current load to drop, thus giving back less and less energy under the same discharge current conditions.

z.entropic

[Tsuioku]

I often read that Deep Cycling degrades the battery faster however, regular reconditioning (draining your battery to <3% ) is recommended to maximize the lifespan.

My question is, is it still better to have light cycling and recondition regularly or to recharge the battery only when it's <3% (so it's more like reconditioning every time)

[Leon]

My advice is NEVER to Deep Cycle, except as a last resort if your remaining capacity is VERY low, and you are close to "giving up" on it. YMMV.

Please add location and other pertinent information to your profile. Not required, but appreciated.

[Tsuioku]

My advice is NEVER to Deep Cycle, except as a last resort if your remaining capacity is VERY low, and you are close to "giving up" on it. YMMV.

Please add location and other pertinent information to your profile. Not required, but appreciated.

Happy now?
You say NEVER Deep Cycle... what about Reconditioning?

[Leon]

Happy then, happy now!

Reconditioning = Automated Deep Cycling.... my advice remains the same.

[dfumento]

Click on the battery (e.g. 100%) in the task bar and select "Improve Battery Health" and then select HELP.

Here is part of what it says:

Reconditioning The Battery
Reconditioning your battery can increase the full charge capacity of your battery. It is recommended that you let the battery run to less than 3% at least once a month.

Occasionally, it may be necessary to "deep cycle" your battery one to three times to give it optimum performance. This will require you to let your battery run completely down, and then to recharge it fully. It is recommended that you charge your ThinkPad overnight for a full charge from a deep recharge condition without leaving your ThinkPad powered on. A brand new battery should be fully charged overnight before its first use.

To do this, click Improve Battery Health button, then click Recondition Battery Now button

[z.entropic]

Deep cycling of Li-ion battery packs may be harmful only if (1) the cells are out of balance in terms of capacity (due to an uneven capacity fade) AND (2) they are charged and discharged in series without an intercell shunt/monitoring circuit--but well-designed laptops and packs always monitor individual cell pairs/triplets which should prevent an overdischarge from occurring.

Commercial Li-ion cell can easily take >500 cycles down to 2.75-2.8 V per cell, which is well below the voltage where your laptop would shut down on its own. Only below this cell voltage the dissolution of copper may occur. In addition, there's absolutely no memory efect in standard Li-ion cells. So, in contrast to Ni-Cd and Ni-MH cells, in practice it really doesn't matter if Li-ion cells in laptop packs are deep cycled or not; a high pack temperature is a much bigger problem, so it may make sense to keep the pack out at 40-70% state of charge if the laptop is mostly run using its external power supply.

To be absolutely clear, I am not saying that the cycle life of Li-ion cells in general is not affected by the depth of discharge--there are examples where they were cycled for 6,000-25,000 times or more (for LEOS applications where LEOS = low-earth orbit satellite), but these were shallow cycles (between 30 and 70% state of charge) under very tightly controlled conditions. What I am saying that other factors, such as calendar life and pack temperature, will swamp any effect of the depth of discharge on the battery life.

z.entropic

[Leon]

z.entropic, welcome to our Forum. Your knowledge and detailed posts re: batterys are much appreciated.

Many members of this Forum (including myself) have noticed a significant loss of Full Charge Capacity after a deep discharge. Do you have an explanation of this apparent inconsistency with your information above?

[skuehne]

As a related question: I use my T42 about 90 percent on AC, so I thought it would be a good idea to lower the charging threshold from the default 96 to 75 % to avoid frequently 'topping off the battery'. Is that a good decision?

Also, I can only set the charging threshold on the T42. The battery maximizer for my X20 doesn't offer these settings. Is there a way I can control the threshold, maybe via registry settings?

By the way, I also think that there should be a battery FAQ for this forum, or at least a sticky thread. Taking care of batteries and tips on usage and charging seems to be one of the more common discussions here.

Sven

[JaneL]

>By the way, I also think that there should be a battery FAQ for this forum, or at least a sticky thread. Taking care of batteries and tips on usage and charging seems to be one of the more common discussions here.
>

Feel free to collect the information and write one! This is a community effort, after all.

[krosenstein]

Occasionally, it may be necessary to "deep cycle" your battery one to three times to give it optimum performance.

Notwithstanding what IBM says, deep cycling a lithium battery is about the worst thing you can do if you want to optimize performance.

See: http://www.batteryuniversity.com/parttwo-34.htm

This site has lots of good info about most battery types.

[WJB]

I bought my notebook from a retail e-seller and NOT a licenced IBM dealer. The computer was brand new (warranty expires at 12-29-07) but the Battery Maximizer says the battery's take off took place in 04-2004. Today it's total capacity fell from 47.52 down to 41.94. I did reconditioning (as suggested in the Batyery Maximizer help) and apparently it only stole snother fraction of Wattage from my battery's overall capacity (41.60 now). There were a total of 11 battery cycles ran until now. Apparently this is way over the average. Is there something I could do about it?

regards!

[Leon]

Not much right now... if the Battery Maximiser condition indicator goes to "Red" before the expiration of the first year of your warranty, you can try to get it replaced through the standard warranty channels in your country..

Best of luck with your new Thinkpad. Welcome to our Forum!

[Matt]

After 161 cycles since June, my battery's full-charge capacity is now 26.87 Wh, down from the design capacity of 47.52 Wh. I find it hard to believe that a battery losing half its capacity in less than a year is typical. At this rate, am I really looking at having to replace my battery every year???

[Leon]

The only "hard and fast" rule from IBM is if the Condition Indicator goes Red and stays there before 1 year. Anything else is a judgment call on their part. Call at 11.5 months and try.

Please add location and other relevant information to your profile. Not required, but appreciated.

[skuehne]

Feel free to collect the information and write one! This is a community effort, after all.

I've been thinking about starting a battery FAQ for a while. As soon as my current deadlines are over ...

Sven

[kev009]

You actually have pretty decent battery wear. For 44 cycles, 3% is quite good.

[z.entropic]

Many members of this Forum (including myself) have noticed a significant loss of Full Charge Capacity after a deep discharge. Do you have an explanation of this apparent inconsistency with your information above?

I don't believe the numbers you see are very accurate, nor do those observations make any sense from the technical standpoint. Unless some cells in the pack were seriously out of balance (one cell or cell block having capacity several, maybe as much as 5-10, per cent lower than the rest), or the pack was very deeply discharged, i.e., below 2.5-2.8 V per cell/cell block AND the control circuit was poorly designed and/or malfunctioning, there is no reason for contemporary cell to lose capacity or, God forbid, die due to deep cycling (by 'deep cycling' I understand getting the pack to a mulitply of 2.8-3.0 V per cell). In fact, recent Li-ion cells can survive much deeper discharges (to as low as 0.5 V) without a significant loss of capacity; however, no cell can survive voltage reversal.

On the other hand, there is plenty of poorly designed and made cells and packs, especially from China, but IBM's packs are decent and from (mostly?) reputable and qualified manufacturers (Sanyo etc.). If you buy a replacement on e-bay or from another non-IBM site, you can never tell what you'll get.

Finally, I've been in the Li-ion R&D for more than a decade and I think I know what can cause real problems and what are imagined ones. There's a lot of voodoo science and UFOs in the battery field, and I'm very skeptical when I hear of such observations that cannot be reproduced in the lab under well-controlled conditions...

z.entropic

[Leon]

Thanks! Again, we appreciate and welcome your current and continued expert help and support in this sometimes "voodoo" area of Laptop Science. FYI, the numbers (Full Charge Capacity Loss after Deep Discharge) that many members are getting is from IBM's own Battery Maximiser "Battery information" tool.

[Leon]

z.entropic, to expand on my comment above.... Are you familiar with the information and method of calculation utilized by the Thinkpad Battery Information? Do you have reason to believe that these calculations are inaccurate?

[rivalarrival]

To quote from http://www.batteryuniversity.com/parttwo-34.htm

Although lithium-ion is memory-free in terms of performance deterioration, batteries with fuel gauges exhibit what engineers refer to as "digital memory". Here is the reason: Short discharges with subsequent recharges do not provide the periodic calibration needed to synchronize the fuel gauge with the battery's state-of-charge. A deliberate full discharge and recharge every 30 charges corrects this problem. Letting the battery run down to the cut-off point in the equipment will do this. If ignored, the fuel gauge will become increasingly less accurate. (End Quote)

It would seem, IMHO, that the problem with the battery meter dropping from 20 - 40% capacity to 5% instantly (as seems to be a common occurance with the thinkpad) can be attributed to this digital memory problem.

When I first got my (used) T22 (with 2 batteries), the gauge would take 45 minutes to drop to ~20%, then drop to 5% , then to 0% where it would hang on for another 20 minutes or so. I noticed a similar effect on the second battery. My friend tells me that his thinkpads do the same thing at times.

[rivalarrival]

I've seen a few sources suggesting freezing (0 degrees C) the battery pack before attempting a charge. I'm trying this, and I'm getting some interesting behavior.

First, I noticed that the maximiser software was reporting the temperature at 255C and refused to charge. After about a minute, the battery warmed to 3C and began to charge. I'm assuming the thermister had a conniption fit from being so cold...

Second, and the thing I can't explain, the Maximizer gauge and the windows 2000 power meter are reading vastly different charge rates. Right now, I'm getting 74% on the maximiser, and the power meter is reading 58. Temperature is at 17C and rising. The maximiser continues to climb steadily, while the power meter stays the same. I've noticed minor discrepancies between them that usually resolve in a minute or so under normal conditions...

Third, it boosted the stated capacity on the battery pack by about 10 percent. I haven't had a chance yet to verify the new capacity, so it's probably artificial.

Any thoughts? I'm particularly interested in the power discrepancies...

Thanks

[Leon]

rivalarrival
I concur with the information you posted, but the member experiences that I am referring to are AFTER a deep discharge (which, by that info, would recalibrate the sensor). They are finding a significant loss of Charge Capacity after these (supposedly calibrating) Deep Discharges. So, either the sensors are incorrectly reporting, or the Deep Discharges ARE taking away capacity.

z.entropic, anxiously awaiting your opinion on this, as well as your comment on rivalarrival's freezing experiment.

[Matt]

I did a "battery recondition" yesterday, and today was surprised to find that:

1) the battery's full-charge capacity jumped to 30.32 Wh (up from the 26 or so yesterday), and

2) the battery health indicator is now green again, having been yellow.

[z.entropic]

[quote="Leon"]z.entropic, to expand on my comment above.... Are you familiar with the information and method of calculation utilized by the Thinkpad Battery Information? Do you have reason to believe that these calculations are inaccurate?[/quote]

No, I am not (although I use it the utility from time to time). I have no reason to suspect that its results are grossly in error, although it's not trivial to measure the amount of charge (times the instantaneous voltage to get the integrated energy) accurately using a tiny circuit. Most of the methods use a voltage drop measurement across a calibrated shunt resistor which often is inaccurately callibrated; it may also waste up to 5 % of the battery charge just to keep track of all the coulombs flowing back and forth...

My 1-year-old 6-cell Chinese-made battery pack has a design energy of 39 Wh, 33 of which are still available after a hundred deep cycles (when the laptop actually shuts itself down). Knowing the behavior of individual Li-ion cells (18650s) from different manufacturers, I would not expect anything better than that. If you see a big drop in energy after a single deep discharge (when the laptop shuts down), there must be, I believe, a problem with at least one cell or a cell block and the battery should be returned.

BTW, is there a utility out there that would run as a process and record periodically the various battery parameters visible in the battery MaxiMizer Info screen to a file? That would be one interesting record to study! I have MobMeter installed, but as far as I know, it doesn't write any files.

z.entropic

[z.entropic]

[quote="rivalarrival"]To quote from [url]http://www.batteryuniversity.com/parttwo-34.htm[/url]

Although lithium-ion is memory-free in terms of performance deterioration, batteries with fuel gauges exhibit what engineers refer to as "digital memory". .[/quote]

I compltely agree with that statement.

z.entropic

[z.entropic]

[quote="rivalarrival"]I've seen a few sources suggesting freezing (0 degrees C) the battery pack before attempting a charge. [/quote]

Attempting to charge a regular Li-ion pack at (very) low temperatures might kill the battery in a dangereous way and should never be done with laptop packs--here's why, with some nasty details of the mechanism for those interested:

When a Li-ion cell is charged, Li cations (Li+) move from the positive electrode material (typically, LiCoO2, lithium cobaltite) through the liquid electrolyte (1 molar solution of LiPF6, lithium hexafluorophosphate, in a mixture of carbonate ester solvents) to the negative electrode (typically, an artificial or processed natural graphite or, sometimes, a coke-like carbon (Sony researchers once reported testing battery carbons made from coffee grinds--the Kona grinds gave the best performance! ).

The electrical resistance of the liquid electrolyte solution decreases several times at 32oF (vs. room temperature conductivity) and the mobility of Li atoms inside the carbon at low temperatures (the so-called lithium diffusion coefficient in graphite) is also quite a bit lower than those at room and higher temperatures.

Now, when a standard charging voltage is applied to a very cold battery, the higher resistance of the battery components causes a large voltage drop between the cell electrodes and may cause the potential of the negative (carbon) electrode to fall below lithium plating potential (-2.97 V vs. the hydrogen standard electrode). Metallic lithium doesn't plate from organic solutions as nice, mirror layer--it plates as a nasty, gray, high-surface area mess called 'mossy lithium', an extremely reactive, pyrophoric (self-igniting), highly resistive stuff. Its presence leads to shorts (dendrites)through the separator and, due to self-heating caused by its reaction with the electrolyte, may result in battery venting and/or explosion, especially if the cell is subsequently overheated or otherwise abused.

Not to mention the fact that lithium once plated is gone from the charge/discharge cycle, thus a permanent capacity loss and impedance increase.

This is a nightmare scenario which is more likely to happen when the fast charge is attempted at -10 or -20C,not necessarily at 0oC, but why tempt the devil?

Storage at low temperatures (not too low, though, maybe 0-5oC), on the other hand, is highly recommended as typical chemical reactions, especially the rate of all the bad passivation and self-discharge reactions, slow down by 2-3 times for each 10 degree C drop in temperature (the well-known Arrhenius relationship).

If you freeze the electrolyte, which may happen at temperatures lower than -20 - -25oC, the battery will most likely be half-dead after thawing for the same reasons that the frost kills summer plants--the crystals of the electrolyte destroy the electrode microstructure and connecivity of the electrically connected particles.

Are you all ?

z.entropic

[Leon]

ok, one last (for now ) follow-up... accepting what you said about deep discharges not being bad under normal circumstances, all things created equal, is it even marginally better to do shallow vs deep discharges, or entirely equal over time?

Also, while the Dr. isn't charging , do you recommend long term storage in the fridge or freezer, and at what charge level?