Hacker News new | past | comments | ask | show | jobs | submit login
The De-evolution of My Laptop Battery (ifweassume.com)
245 points by brendan_gill on Aug 16, 2013 | hide | past | web | favorite | 145 comments


Huge mistake here - he says he cycles his battery! DO NOT DO THIS TO LITHIUM ION BATTERIES! Lithium Ion will last longer by NOT allowing the cells to be fully discharged!

This is the opposite of Nickel Metal Hydride and Nickel Cadmium - these types of batteries DO require a full discharge and full recharge in order to provide a longer life.

In addition, the number of discharges, the temperature, etc all come into play with lithium ion. These thinner laptops are actually lithium ion polymer, where the materials are folded into packs...much more susceptible to the elements, like heat.

Disclaimer: Spent 3 years in the battery field. http://batteryuniversity.com/learn/article/how_to_prolong_li...

Edit: Just want to point out that charging a laptop while using it could potentially create more heat, degrading the battery. I'm not sure what is more damaging - letting the li-ion battery drop below 70%, or charge it while the laptop is being used. So many variables!

The really dumb thing is that, as far as I can tell, a battery's controller generally counts a "cycle" as "started discharging, started charging", regardless of time in between. At least, that's how it was with my MacBook a few years ago. Unplug, plug in, cycle count increases.

The crappy thing about that is warranties. Apple, for example (at least last I checked; I don't use a Mac anymore) , would replace a battery if it had under some number of charge cycles (and presumably was less than a year old), but not if it had more than that number.

That means that if you treat your battery really nicely by always keeping its charge between 40% and 60%, and it turns out to be defective, they probably won't replace it because of how many "cycles" it has gone through.

Of course, those kinds of cycles have very little to do with the longevity of a Li+ or LiPo battery anyway. A better metric would be total energy output over time, which would be far simpler to calculate anyway.

I don't believe that description of cycle count has ever been accurate for Apple hardware and certainly isn't for recent MacBook Airs:


Yeah it definitely doesnt, I have had and used dozens of macbooks and you cant ever plug in and out and see your charge cycle increase each time. I usually have my laptops 1-2 years and they rarely go above 300 charge cycles per year, and I usually dc my laptop 5-10 times a day.

If you read the documentation a cycle is considered a full charge and discharge of the battery. It doesn't have to be discharged or charged fully, but when you've charged the full capacity of the battery and discharged that same amount, that's a cycle.

Sure, but that was not reflected in System Profiler when I actually tested it. Unfortunately I can't test it again now, as my old MacBook's battery will hold up for approximately 14ms before the computer shuts off.

A battery engineer would view that as a cycle, but Apple does not.

Funny thing is, I don't imagine any li-on battery lasting 1000 cycles after full discharges! Good warranty or what!?

As they say in the renewable energy field: the majority of batteries do not die peacefully, they are murdered.

As a bit of a re-assurance, the typical laptop power circuit in a laptop using lipo is aware of this and will simply refuse to drop the battery below a certain level so likely the damage is somewhat limited. Still, if you don't need to discharge a battery all the way as far as the circuitry will let you with lipo's you'd better not.

> opposite of Nickel Metal Hydride

Fully discharging any battery isn't good for it. NiCd had the particular problem of "memory" developing though, so going through a full discharge cycle was necessary to prevent the memory forming. It wasn't good for the battery, but it was better than letting the memory develop.

I'm not aware of NiMH developing a memory, so that wouldn't be a good thing to do with NiMH either. I always tried to avoid full discharge of my NiMH batteries and always got a pretty long life out of them.

Really the full discharge cycle is an artifact of the needs of one particular chemistry (NiCd) that became something that "everybody knows" you should do long after it became counterproductive.

There are two types of discharge states - typically "fully discharge" implies within the application the battery was designed, not completely dead. For example, a car battery is fully charged at 12.6v, but discharged at 11.2 (or somewhere around there). There is still lots of capacity left, but dead for the use of cranking over an engine. NiMH typically has a very fast self discharge rate - letting these dip well below the operational capacity level will be much more detrimental than just letting your drill die after prolonged use.

NiMH has memory too, but not as bad as NiCd. We always suggested allowing NiMH to fully discharge time to time, but we never suggested keeping a phone on its base constantly.

Both these types of batteries need to be primed properly - full discharge/recharge the first few cycles. (Not sure about the new pre-charged NiMH, like Eneloop)

Memory builds faster when left in a discharged state.

Cadex machines can be used to reverse memory, to an extent.

I'm speaking from memory from 6 years ago. Hopefully I didn't mess up too much =P

There is occasionally a need to do a full or nearly full discharge on Li-Ion batteries in order to recalibrate the state-of-charge indicator if the embedded controller gets confused, or if the battery is nearing the end of its usable life and has significantly diminished capacity.

But in general, deep discharges are bad for any electrochemical cell, and should only be done to prevent or forestall something worse.

NiMH cells shouldn't be "topped off" when they're nearly full, but that doesn't mean they should be fully drained before charging.

I believe you are correct on the "cycling the battery", which I believe you want to avoid, so that you battery will be able to charge fully for longer.

"A charge cycle means using all of the battery’s power, but that doesn’t necessarily mean a single charge. For instance, you could listen to your iPod for a few hours one day, using half its power, and then recharge it fully. If you did the same thing the next day, it would count as one charge cycle, not two, so you may take several days to complete a cycle. Each time you complete a charge cycle, it diminishes battery capacity slightly, but you can put notebook, iPod, and iPhone batteries through many charge cycles before they will only hold 80% of original battery capacity." - http://www.apple.com/batteries/


There are mixed meanings in the definition of "cycle", but both make sense, just depends on the context. Thanks for the blurb!

He does state "letting the battery" cycle, so I'm assuming he is implying the battery is being allowed to be discharged near fully.

Interesting. Last I looked, the supposed optimal way to prolong a Li-Ion battery was to not let it get below 50% (ie, plug it in to charge when it gets to 50%). I think it was from a link off of Lifehacker a while back. Storing at 40-60% was recommended as well.

I do typically use a smart charger on my NiMHs, and make sure to "break-in" (full discharge -> trickle charge to full -> full discharge -> trickle charge to full) every three months (12 for the new LSDs); I've managed to keep some AAs for well over a decade doing that, albeit I have plenty of spares, so I cycle through them and most are kept at room temperature storage. I used to freeze, but read that that might not be good for them long term (not to mention you have to wait for them to warm up before you can swap, defeating the purpose of spares; also, there's the possibility of thermal shock if you don't warm them up slowly/gently in the fridge).

I too read the 50% capacity rule, but I read it in a very professional battery book, but I remember the gains were very minimal. Something like 10-15% increase in lifespan. When a battery is only lasting 2-3 years as it is, that is less than half of a year.

Going back to my link, it looks like 50% is the sweet spot in the DoD chart; multiplying the number of discharges by the DoD results in 50% DoD being the highest return on capacity.

But...also remember in an application where you can use the device while its charging changes the concept of maximizing life span. I don't believe the act of discharging will avoid shortened lifespan, but rather it avoids degradation on the battery CAUSED by discharging itself.

Batteries hurt my brain...

Wouldn't all modern power management ICs operate the laptop directly from the power supply while float-charging the battery resulting in negligent amounts of heat?

Yes to the former, but (hopefully a real battery expert can chime in here) my understanding is that lithium-ion batteries cannot be float/trickle charged safely. Their internal resistance is too non-linear and it's too easy even with a "full" battery to provide a voltage that will sink dangerous amounts of current and overheat.

All lithium ion technologies I'm aware of require active current and temperature monitoring.

Probably 90% of the time I'm using my laptops is when I'm actually sitting at my desk and they are always plugged in at that time. I recently read where someone else had stated that it was good to remove the battery (if you can) if using the laptop with external power.

Any idea if this this accurate or will prolong battery life? I've got an (almost) brand new Thinkpad with a large, expensive battery that I'd like to preserve as much as possible (like anyone else).

I don’t know if this applies to other notebooks but at least Apple used to recommend that you do not remove the battery [0]. There’s a little more discussion on Super User [1] but there seems to be no definite answer.

[0]: http://web.archive.org/web/20130615084258/http://support.app... [1]: http://superuser.com/a/12842

At least some Thinkpads used to let you limit the charge on the battery.

Capping the charge at something like 60% is a great way to preserve the battery if you're not going to need all that battery life.

Storing at 40%-60% elongates the life compared to 100% or 0%.

All of this is 2nd hand, and I have no particular expertise to validate this comment.

I don't understand. So with this setting people are artificially limiting their battery capacity to 40%, in order to preserve capacity they're not going to use? Meanwhile you'd have to ride a battery pretty hard to grind it down to 40% health, it'd take many years probably.

I'm not sure what the average lifetime of a laptop is, but I'd be surprised if it's more than 8 years. What's the upside if the battery is well preserved at the time when the device is being replaced? Let's say with normal usage the battery will perform at an average of 80% over its lifetime, that's still better than 40% - for the same money.

If your cycle count is 223 and you are nearing 80%, then your battery is defective, plain and simple. Apple will replace it. It's happened to me a couple times and Apple took care of it with no hassles.

I had my first macbook's battery replaced. I took it into the apple store for an unrelated problem, the tech noted the awful battery life and asked me how old the laptop was, said, "that's not right...," and grabbed another battery. Nice and easy.

If I remember correctly, with the Macbook Air it will have to be '"that's not right...", and grabbed another Macbook Air, and started migrating my data, etc, etc...'

Nope. Had a battery replaced in my 2012 Air recently(only because a clip separating the battery from the case broke,not because of the battery life) and it took 30 minutes to get it done at an Apple Store.

Ah, my mistake.

They recently replaced the battery on my Macbook Air. They didn't replace the whole computer, just the battery. Took about 15-20 minutes.

Ah, I thought the battery was permanently attached to the body.

The RAM is the thing with Macbook Air's that permanently soldered to the system board. Everything else is difficult to replace due to the form factor, but still replaceable.

They're just held in with torx screws, from what I could see of my old 'unibody' mac when I took it apart. Current gen may be different though.

Indeed, they're not meant to be user-serviceable, but they are very easily serviceable at the store.

Heck, they are user-servicable (if you don't mind voiding your warranty), you just need a slightly unusual screwdriver.

Really? I got a 2006 MacBook Pro in Dec 2006, and each year I had to buy a new battery from Apple as it generally got down to 30-40% capacity.

First time it happened (and was still under warrantee), I took it into Apple, and they said it was normal, and they only replaced batteries if it was defective.

I went through 5 batteries until I got a 2011 MacBook Pro.

Did they check the cycles?

Back then the batteries were only rated for 300 cycles so if you were above that they wouldn’t exchange it. Apple did exchange the battery in my MacBook Pro (2007) three or four times or so (I had Apple Care, so warranty for three years) because I had similar problems you had – but each time my battery was below 300 cycles. As far as I know that is official Apple policy. I never had to convince or smooth talk anyone, they just exchanged in, no questions asked. And I didn’t exchange the battery always in the same place, once or twice over telephone, at an Apple Authorised Service Provider, at an Apple Store.

I had previously had terrible luck with their batteries. That changed for me when they stopped making them removable (thankfully!) For the removable batteries in my 2005 or 2006 Apple said they had a lifetime of 200 cycles (mine never reached that). My current MBP is supposed to have a lifetime of 1000 cycles. I'm a few hundred in and have high 80 or low 90 % capacity. It might not meet their expectation, but quite a bit better for my needs.

2012 Macbook Air: Cycles: 138 Health: 83%

Is this replacement worthy? What am I doing wrong?

Without knowing what their exact boundaries are your numbers don't look great. Most people here seem to have Airs, but to compare with my Pro:

    730 cycles, 92% capacity after 27 months (mid-2011).

How do you prove this to Apple? I have a 1year old MBA that is definitely not getting anywhere near 3 hours on battery, let alone the 7 they advertise.

You can download Coconut Battery [1], which will show you the battery capacity and number of cycles.

1: http://www.coconut-flavour.com/coconutbattery/

The cycle count is in System Information, as is the age of the system. Their own docs say it should get 1000 cycles before you're down to 80% of the original capacity. They have diagnostic tools to test it at the Genius Bar and measure the actual full capacity of the battery.

I have a 2010 MBP 17 which reports 8772 vs. 13000 (39 months old, with 186 cycles) -- it should be around 10000. AppleCare expired 95 days ago and I didn't think to check before then. Presumably I have a decent shot at getting it replaced in variance to official policy, but I'm not sure if I actually care. (I mainly just use it as a desktop.)

I can attest to that: my Macbook white bought on October 2009 (that's 46 months folks) which I used every single day to this very day has 261 load cycles and 91% of original capacity.

That's pretty good going I would have thought? As well as charging cycles, I assume time is a factor. My Macbook Pro (bought July 2009) had reached 65% capacity by the time I replaced it, after something like 45 months and ~250 cycles. By then the computer had been begging me to replace it for about six months.

I actually thought this perfectly reasonable, with nearly 4 years being a fair lifespan for a battery, no matter how few cycles it might have been through. While it was obviously knackered, it still held enough charge to be perfectly useful, if not amazing - but perhaps I should have held it to a higher standard?

(Hopefully it's just this year's unusually hot UK summer but after a mere ten cycles its replacement is down to 93% already. So maybe there's just something up with the electricity supply to my flat.)

my blackbook (late 2007, Santa Rosa, core 2 duo 2.2GHz) still has the original battery and it still holds a decent charge (3+ hours.) My Walstreet 2's battery still holds a charge lasting circa 2+ hours. I religiously avoid running plugged in and always try to charge then run down. Seems to have worked for me.

My 2010 MacBook Pro is on its third battery, and even this one has had 97 loadcycles in six months, 90% capacity left...

Agree - I've a 2 year old mid-2011 MacBook Air, 225 cycles and 95% health.

Wow, this is an excellent article! Nice visualization of the battery life.

I am quite nervous about the battery sealed inside my own new Macbook. Is it better to attempt to use the battery as much as possible in order to get a new one before the 3 year Applecare Warranty expires? Or to maniacally keep the machine plugged in if at all possible? Or simply not to care? The price of a replacement will be high, and I wonder how many years it will even be possible to replace these batteries.

The thinness of my laptop is amazing, but I would rather have a battery that is not GLUED inside the machine.

I believe Apple charges $199 for a battery replacement, which is definitely not cheap, but might be worth it if you were still happy with the rest of the machine. I'm not sure how long it takes them to do it, though. I'd expect they'd continue to offer that service for quite a few years to come.

As other posters have mentioned, it seems to me like the glued-in batteries don't rapidly decline like the old ones did. I don't have any numbers, though, just impressions.

Last time I did a battery replacement with Apple (for a mid-2009 Macbook Pro 5,5) it cost ~$120 off-warranty.

Doing the research, seems like different machines have different battery replacement costs from Apple:


* $129 for non-retina 11", 13", and 15" Macbooks (air and pro)

* $179 for the 17" Macbook pro

* $199 for the retina Macbook pros.

Apple batteries cost no more than competitors' batteries, and that includes getting them exchanged. Sure, it's annoying you can't do it yourself, but that's about the extent of the downside.

Of course you cannot get third party batteries … but surely no one would get those for laptops amyway. (Well, some batteries are not yet glued in, so those can easily be replaced. But i don't think getting third party batteries is very wise.)

I used a third party battery with double the previous capacity for an Asus laptop without any issues

Well, not all explode, certainly.

Lithium batteries are fucking dangerous. I don’t think it’s very wise to use third party ones, at least not the big ones.

imo very conscious decision on part of Apple (and PC makers) to provide short battery half lives, rather than spending more money to deal with the size and heat/cooling constraints of laptops, especially the thin lightweight variety.

not only saves them money short term, but helps push users to new versions before they really need to jump...

Is there evidence for this? Is there a better battery tech the entire industry should be using instead? Seems unlikely.

Not to support the "evil executives" hypothesis---obvious market(ing) forces are a sufficient explanation for the dominance of LiCoO2---but I would easily opt for a laptop that offered an LiFePO4 ("iron phosphate" [0]) option. The drawback would be slightly reduced initial capacity; the advantage would be much greater cycle life. Also, less spontaneous combustion.

[0] http://en.wikipedia.org/wiki/Lithium_iron_phosphate_battery

is it possible the better batteries don't (generally) exist because the manufacturers are not motivated to invest the r&d in the first place? i certainly have had too many laptops that don't provide optimum cooling (and they could, at more expense) when we know heat is a limiter of battery life.

Can you point to data showing competitors that are doing it better? Seems like the battery life problem impacts everyone pretty equally, and the batteries being used inside Macbooks are pretty similar to those used by every other manufacturer.

agree, this is not specific to macbooks. bottom line is today's batteries only achieve (relatively) limited charge/discharge cycles and i don't believe (could certainly be wrong) there is much business motivation to improve that.

I'm sure manufacturers keep an eye on some minimum standards for battery life over product life - they do have a financial incentive to do so, both from a brand quality perspective and in terms of the cost of each visit to the Genius Bar / warranty claim. I replaced the battery on my MacBook twice under warranty, I can't imagine that laptop ended up being profitable for Apple.

There's plenty of business motivation in fields that combined exceed PCs monetarily, such radio, military, space, automotive, aeroplanes, power grid, anywhere you use energy

The best approach is to not worry. The battery is supposed to retain 80% capacity after 1000 charges, which should still be enough for decent battery life. The $200[1] cost to replace the battery is well worth it for two years of moderate to heavy use without worrying about wear.

[1] Estimate, no idea if this is accurate

Its really NBD. My 2010 MBA still holds a solid 5-6 hours, down from maybe 7.

>Wow, this is an excellent article! Nice visualization of the battery life.


Not an Apple consumer here, but my netbook battery only lasted about one year, and I attribute this rapid failure to my use of a cheap, (I live off-grid) modified sine wave, voltage inverter, but I have no valid proof of my assumption. As likely a cause for the battery failing to take a charge is the fact that I often ran the netbook until the battery charge depleted to the point of shutting the PC down.

When I bought my current, pure sine, inverter I noticed that my new battery lasted nearly twice as long, but since switching to FreeBSD last year my second battery quickly bit the dust...

I'm expecting the present battery to last somewhere between one and two years, but I'm not taking that expectation to the bank.

To confirm or disprove your idea, you could look at the output of the battery charger rather than the input to it.

In general, AC-DC converters are pretty insensitive to the harmonic content of the AC input because their role is to remove any harmonic content they do find. They do this with huge low-pass filters with corner frequencies well under 1Hz, whether the charger is either linear (unlikely) or switching (much more likely).

When AC becomes distorted, you start seeing harmonics (i.e. contributions at higher frequencies). Low pass filters are progressively more effective at higher frequencies. This is to say, I wouldn't be at all surprised if the efficiency of the charger itself suffered, but the output voltage was unaffected, meaning the laptop was unaffected.

Interesting, but does using a DC to AC inverter change anything?

I've noticed that cell phone batteries also failed prematurely when I had the modified sine inverter.

Do you mean that if my charger was operating with less efficiency due to modified sine wave input then my batteries were not likely to be affected?

It's more likely that you were simply cycling the batteries more aggressively due to the circumstances which led to you using a DC-AC inverter.

>letting the battery cycle*

Wait- I thought that was for NiCd batteries and Li-ion batteries where suppose to always be topped off? Furthermore going below 15-20% charge hurts their max charge level?

*By cycle he means running it down and charging it up?

Assuming we're talking about lithium ion chemistry, The most significant thing that will impact the longevity of the battery is heat, and keeping the battery at a high level of charge generates it. In particular, if you're always plugged in and the charging system is keeping it at 100%, this will impact the longevity in a meaningful way.

Not sure about Macbooks, but my laptop has settings to control this. I typically have it setup to only start charging when the charge drops below 40% and to stop at 85%.

Some other points:

- It's preferable not to discharge fully. Very deep discharge can damage the battery, although all modern batteries have circuits to protect against this.

- For long term storage, ~40% charge is best. That's why new phones typically come charged at around this level.

Yeah, can someone who knows how this shit works chime in? There's a lot of conjecture on the internet but few answers. Apple's advice seems to be "don't worry, be happy, don't let it be in a discharged state for a very long period of time, occasionally cycle through it".

I'm on my third battery for this 2009 Macbook unibody (granted, the first one was water damaged but still…), and 300 cycles in it's starting to give me Service Battery warnings and I must be doing something wrong.

There seems to be no exact science with Li-ion battery life. Even defining "life" is difficult, since some people say the batteries "life" ends at 80-50%.

Of all the things I've heard about Li-ion batteries, the only ones that haven't been refuted are:

- Li-ion batteries like living at around 70%. More or less will decrease their life

- Li-ion batteries will always lose 30-50% capacity in 3 years even if they're just lying on a shelf due to their chemistry

- Li-ion batteries dislike heat. More heat reduced battery life. Coldness will temporarily reduce capacity, but has no long-term effect on lifespan.

The official advice is:

"Apple does not recommend leaving your portable plugged in all the time. An ideal use would be a commuter who uses her notebook on the train, then plugs it in at the office to charge. This keeps the battery juices flowing. If on the other hand, you use a desktop computer at work, and save a notebook for infrequent travel, Apple recommends charging and discharging its battery at least once per month.


If you don't plan on using your notebook for more than six months, Apple recommends that you store the battery with a 50% charge." [1]

[1] http://www.apple.com/batteries/notebooks.html

I wonder if using your laptop plugged in at your desk 90% of the time, but carrying it around to use in meetings the other 10%, is as good as their "ideal use."

Maximizing your battery's life is not remotely close to a science, but as a rule of thumb, discharge to 50% every couple of days then recharge.

Well as long as there continue to be long, drawn-out, rambling meetings, we're good to go! Although in that case, a short battery life is a feature: "Oops, I'm at 5%! Be right back while I fetch my charger!"

It's straightforward to do experiments, control factors affecting experiment, collect data, process is well understood electrochemistry... only thing that might make it "not science" is that it's too well understood so you may claim it's just engineering instead :)

When I'm sitting at home with my laptop plugged in for a long time, I usually just unplug the battery from the computer and set it next to the desk. Is this an ok idea?

The battery is still used even if you leave your laptop plugged in -- it uses the extra juice when the system is under high load. I have used my 45-month-old MBP plugged in 99% of the time, and coconutBattery shows 97% capacity at 38 cycles.

Deep cycling only helps calibrate the battery charge indicator, but doing so will lower the battery's capacity simply because you're cycling it.

I've always kept my laptops plugged in whenever it's reasonable to do so and I get great life out of my batteries.

My 2008 T61 ThinkPad had 85-90% left when I sold it in 2010.

My 2010 15" MBP had nearly the same battery life after 2+ years as it did when it was new.

Here are the numbers from my current 2012 15" rMBP:

Age: 13 months

Cycles: 117

97% original capacity

And that's with 70+ hours a week of use.

Li-ion batteries do have a memory effect also, but it is apparently quite negligible.

Li-ion batteries should NOT be topped up for extended periods; this will shorten their lifespan considerably.

When storing a li-ion battery for an extended period, it is best if the battery has around 30% charge.

However I am not a real expert on this kind of stuff, please don't take this as gospel, if there is a qualified battery expert who can chime in here, please do :)

I've only recently started collecting data for my Early 2011 MBP, but I'm starting to see a pretty steep decline. From 72% capacity down to 58% in a little over a month feels like I might be nearing the end of life for my battery.

  date       | capacity   | loadcycles
  2013-07-08 | 4964 (72%) | 527
  2013-07-25 | 4566 (66%) | 534
  2013-08-05 | 4432 (64%) | 537
  2013-08-16 | 4016 (58%) | 542
Anyone have any advice to stretch its life longer, or is this pretty typical?

mid 2010 5020/5770 (87%) and 393. I'll have to starting track this also.

Interesting experiment, but watch out for the vertical scale on the charts showing capacity against time.

Yeah, not pegging the axes at zero makes for a much more dramatic graph -- a classic way to lie with data viz. Sometimes it's fine to do that (e.g. because without scaling it's impossible to visually understand some of the data), but the context should be made clear.

That said, the plot of the Macbook Air's battery performance vs. the Macbook Pro's is really interesting. I haven't done the tracking, but this agrees with my anecdotal sense is that my 2011 MBP (likewise a main work+personal system) has had really excellent battery longevity relative to prior Mac/other laptops I've used.

As far as I understand, there are two major factors that will shorten the battery life considerably.

One is heat. This may not be avoidable depending on the computer usage. My mom "plays" Second Life (for english lessons) and her battery barely lasted two years - expanding to the point that the trackpad on her Macbook Pro was sticking out.

Another is leaving the computer plugged in all the time, which will top the battery - and apparently lithium batteries hate that.

Some electric cars have the option to only recharge the battery up to 80%, to increase the lifespan. Would it be possible to do the same with laptop batteries? Perhaps by patching the SMC?

>Some electric cars have the option to only recharge the battery up to 80%, to increase the lifespan. Would it be possible to do the same with laptop batteries? Perhaps by patching the SMC?

My laptop (X1 carbon) does this, I assumed most did. The Lenovo power manager let's you choose when to start charging and when to stop. I start below 40% and stop at 85%.

That's nice. I wasn't aware of notebooks which let you do that.

>Another is leaving the computer plugged in all the time, which will top the battery - and apparently lithium batteries hate that.

I thought modern batteries had mechanisms in place to prevent that?

I'd think that they will actually recharge to somewhat less than 100%. But Apple still recommends unplugging the computer from time to time, so I'm assuming they get to very near full capacity.

My laptop is plugged in to a Thunderbolt Display almost all the time, so the battery gets topped off. It says that it is "charged", but I have no idea how much absolute capacity is that.

If I'm not going to unplug the notebook anytime soon (not travelling, not weekend), then I should be able to tell to leave it half charged (half, since discharging completely is even worse).

I don't know what brand of computer outerworld uses, but my MacBook doesn't start charging until below 95%. When plugged in for long durations, the battery is not running not.

One other possibility; does the "battery health" metric actually mean the same thing it did four years ago? My 2012 MBA is showing 92% capacity, and seems to last about as long as it did when new. My 2007 MBP shows ~70% capacity, and lasts about an hour; it lasted about four hours when new. I'm not convinced it's that meaningful a metric.

I have 378 cycles, 90% capacity. Original capacity, 4680mAh, current capacity 4228mAh. Same laptop as him. Mac is 18 months old. His must be flawed.

This discussion is very interesting. I'm puzzled about how to keep batteries working longer, so I looked around for some sources,





with background information about batteries. It would be great to hear comments about what these sources say. I need a reality check on how to take better care of batteries in rechargeable devices.


> The rate of degradation of lithium-ion batteries is strongly temperature-dependent; they degrade much faster if stored or used at higher temperatures.

That might be related. MacBook Airs get really hot even if you aren't doing anything too taxing.

So, this inspired me to collect similar data on my laptop. I'm collecting a few more data points, internal battery temp, individual cell voltage, etc. I'm estimating 30-60 MB/year at 1 minute intervals. My script is available on GitHub here https://github.com/mikegrb/Battery-Logger

I've set my thinkpad to 'start charging' when below 40% & 'stop charging' when it reaches 60%. I use it with AC power plugged in while the battery floats ~40-60% charge level. Once a month I power cycle it to re-calibrate the battery. It's been going good for the past 3 years. Shows 500 cycle count with 80-90 hrs/week usage & ~80% capacity.

I have exactly the same laptop, purchased from an Apple store ~14 months ago.

I have more than double his cycle count at 569. My battery is at 85% capacity.

I'd say your battery is defective, or you were doing something to it that caused it to degrade more rapidly.

It's also a little unnerving that the 2009 MBP has an apparently far superior battery, despite Apple's explicit claims to the contrary.

My laptops (two-year-old MacBookPro and three-month-old Thinkpad W530) are plugged in probably 90% of the time I'm using them (8+ hours per day).

From reading some of the other comments here, it sounds like I shouldn't use them while they're plugged in. When they're being used 8-12 hours per day, however, it's impossible to avoid.

My 2011 MacBook Pro is doing a lot better than my 2007 one. With similar use, the older MacBook went through about a battery every year and a half, replacing once they could only hold up an hour or so of charge. I'm up to 750 cycles and holding high at 98% "health".

The OPs MacBook air sounds defective, or at least, under-performing.

How do you even measure max capacity of the battery? Is it by keeping track of the current balance or is there some other direct physical way to do it?

Measuring once per minute seems like massive overkill for a property that changes on a timescale of months, but the interesting question to me is - does it do anything to the life of the battery?

Mac OS X makes the stats available directly from the battery.

Yes, but how does it arrive at the number it reports?

One assumes by the mAh.

Hmm. I've already hit 100 cycles after only 3 months of owning my Macbook Retina 13-inch...

I guess that makes me a ... power user? ;)

humm, I suppose I should not be too upset then about my Mac Book Pro's battery. It is now a little over 3 years old and I just hit the 500 battery cycles mark this week, about the same time that I noticed my trackpad is no longer working properly. I have had the computer for a while and have not used extensively for over 1 1/2 yrs (have work computer), so I am unsure what it could be, but think it may be the battery. This year we have had a hot weather in the East Coast and my laptop sit out in the changing weather (can't leave the AC on for it), so this may have impacted the battery's lifetime. In general, I can just add that for moderate average use, my Mac Book Pro's battery has lasted quite a bit, but would have liked it to last closer to the 100 cycles promised by Apple's system specs.

Replacing the cells isn't particularly hard if you are willing to take the time with a dremel -- if you're clever you can then close the battery without gluing it.

I still have a 4.1 macbook pro that I keep using specifically because it's the last one they made with a user-swappable battery.

I have a Mid 2011 MacBook Pro - so around 2 years old. According to iStat Pro, it's been through 430 cycles with 93% battery health. As far as I can see from the comments, I'm one of the lucky ones.

Although battery health and maintenance is interesting I have never bothered to nurse or baby my batteries on my equipment. If the battery gets to be too short lived its time to replace it.

Are they're people out there that replace their own MB(P|A)? Is it possible? I know there are hacks out there to take apart and fix chargers (I did it myself).

How did you fix your charger? I have always thought that we (users) shouldn't have to worry about charging, over-charging, stretching the battery (I have honestly heard people talking about this) and so on. We have the tech to make smart chargers, why aren't they here yet?

{OT: has anyone else thought why we don't have a full bay size laptop battery that can be put inside a desktop for those times the power goes out, but we don't want/need a UPS?}

One of the wires broke inside of the magsafe. So with some pliers I opened it up, then cut the broken part, strip the tip to expose the connection, reconnected it (tested it and put some electric tape around it to be sure), and assembled it back together with tiny amounts of hot glue on the edges. I wish they designed things so they can be taken apart easier. I'm not sure if this is really "fixing the charger" but some people decided to solder off the tips connection to the board, cutting the bad part, and soldering it back (I didn't want to do that much work).

I think for most users, they shouldn't have to worry about it, but I feel like there should be the ability in the hardware/software for users to have the ability to maintain/mod their own systems if they please.

I'd like to see the Air-vs-Pro graph in terms of mAh instead of percentage of factory capacity. Better yet, in terms of expected running time on a full charge.

anyone know how to do something similar on a windows laptop?

I'd guess:

- write a little code to drive this API: http://stackoverflow.com/questions/7785096/read-laptop-batte...

- and either use a task scheduler to run your code every minute, or startup a long-lived process which just does this and sleeps

- output in whatever format you like. CSV is obviously good, but given richer APIs on windows you could dump it straight into an excel sheet I guess.

RRDTool is great for storing these sorts of time-series archives, and it will generate pretty graphs as well. You could cobble together an app starting from these StackOverflow postings:

Task Scheduling: http://stackoverflow.com/a/2725908

RRD interface: https://pypi.python.org/pypi/PyRRD

Battery reading: http://stackoverflow.com/a/6156606


This one will also output in XML. The APIs should be there in Windows 7, too.

I have no experience with this kind of stuff for Windows, but I really hope somebody wiser than I can do it easily. I assume there must be an equivalent to cron in win8?

Nice article, might start tracking this too! My mid-2011 MBA is at 203 cycles and 99.5% capacity. Seems to be holding up remarkably well.

This is why I prefer replaceable batteries. Ultrabooks look nice, but ultraportables are more practical in my opinion.

My 2012 MacBook Air on a full charge only lasts 2-2.5 hours during an intense coding session. Am I the only one?

That probably depends wildly on what an "intense coding session" entails. Do you have an IDE that is keeping your CPU pegged by constantly rebuilding everything every time you enter a character? Or do you code in Vim and build every few minutes at the absolute most?

That seems abnormal, though possibly bad IDE behaviour? What is CPU usage like at this point?

Also, might be worth giving this a go: http://software.intel.com/en-us/articles/intel-power-gadget-...

Some badly behaved software (notably Netflix's Silverlight thing) will seemingly peg the CPU at maximum turbo-boost frequency for no good reason, which hits power usage hard.

Sounds reasonable if you have your brightness way up and have CPU intensive apps running like VMs.

I didn't realize battery replacement was such a thing. I have 1856 cycles on a 2008 aluminum macbook with 'normal' health and have never noticed problems. I gather from here [1] that I should be well into my fourth battery?

[1] http://support.apple.com/kb/HT1519

where are those fuel-cell powered laptops? http://techland.time.com/2011/12/23/apple-posits-fuel-cell-p...

Hilarious that I thought to post this on HN, and there it was, at the top...

2010 MBP MC374LL/A :: Cycles: 393 Health: %87

Link is dead for me, mirror?

Is this guy really a developer? I can't imagine using a single 15" monitor to accomplish anything at a productive pace.

That's funny, I can't imagine being tethered to a single spot and accomplishing anything worthwhile!

Snark aside, I love big monitors as much as the next guy, but I've seen very productive developers that never even maximized their code window, let alone used multiple screens.

Getting in the flow is way more important than having a nice setup.

Portability is important to me too. Our business continuity plan accounts for disaster scenarios. My development machine is a laptop (17") with 24" monitor and another machine with 24" monitor all using one keyboard.

Maybe it's the type of work that I'm doing is very technical but there's no way I could actually do this much work on a small screen.

I work all day on a MacBook Air 13. I spend 70% of my time in a terminal, so screen size doesn't matter much to me.

No true Scotsman would do such a thing.

I used 2+ monitors from 2000 to 2010ish. After switching to only laptops I gave up secondary displays and never went back. The only times I miss it is when I want to have docs open. I don't know why you'd need it otherwise.

Same. I made the switch when KDE started supporting the Ctrl+Alt+Arrow Keys to spin my desktop around on a six-sided cube. So much handier to flip between sides. I really miss that in Windows.

That's the thing. I always have multiple docs open - tasks lists, TDS, specifications (I deal with a lot of specifications). As of this typing I have 22 open windows among 3 monitors. Virtual machines, specs, time clock, web browsers, visual studio, skype, SSMS, iTunes, etc.

Call me crazy but I cannot imagine having to do all of that on one machine and still be productive.

Indeed. It's a good thing there were no developers between 1970 and 2000, and all programs spontaneously arranged themselves via xkcd butterflies until monitors arrived that allowed for development.

15" 4:3 monitors were standard before year 2000. You have to go back to probably the early 90s before standard monitors had less real estate than a 15" widescreen. Of course resolutions and panel quality have increased since then, which mitigates the problem of working off a laptop monitor. Personally I find im far more productive on a 24" monitor than a 14" laptop screen, but its not impossible.

I grew up developing on small monitors. It was painful.

It's actually a 13" monitor... I guess it helps you focus.

Yeah, I cannot believe he said that. The more real state the better. Why not spread out if you can afford it?

Look at your computer again and pay special note to the Command + Tab keys.

Sounds less productive than already having documentation on the screen when needed.

Single tasking is the new multitasking.

I'm not a developer, I'm a scientist (astronomer as a matter of fact).

I live on a single 13" screen, the MacBook Air I'm typing on right now. I used to have lots of screen space, driving dual monitors on my linux box at work. I replaced them with a fern.

For some people, less is more =)

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact