Looking for Help on Tesla Battery
I have read a number of stories about how Tesla batteries become bricked if they are completely discharged. What I have not seen is an explanation of the physics or chemistry of why this is true. Can anyone explain it or give me a pointer to an explanation? Certainly if this happened to, say, iPod batteries we would have had torches and pitchforks outside of Cupertino long ago.
Brian D.:
I don't have the explanation myself, but a good one was posted on Hacker News (news.ycombinator.com). Apparently it's something of a false-scare.
http://idc-insights-community.com/posts/c7c83d1a9d
February 23, 2012, 9:15 amBrian D.:
And here's the link on YC news with comments:
February 23, 2012, 9:16 amhttp://news.ycombinator.org/item?id=3624568
aeronathan:
It's my understanding that it would take multiple system failures to truly destroy the battery and given the battery chemistry, if that happens there's a fair chance the car will end up in flames.
I was involved in an unmanned vehicle competition in college and some of our competitors used the same battery chemistry to power their onboard surveillance systems. They had just such a failure in their batteries. Unfortunately it happened while the batteries were unattended in their hotel room. I imagine it's not fun trying to explain to the university why you burned down a Holiday Inn....
February 23, 2012, 9:40 amlukas:
It would happen to those batteries too. If you leave your iPod unused for a while (especially at high temperatures) its battery will self-discharge and eventually go into deep discharge, at which point it is not possible to charge the battery safely. Protection circuits will kick in and your battery will for all intents and purposes have turned into a brick. But then an iPod replacement battery isn't $40,000 and Apple's warranty will usually cover the replacement.
Tesla is trying to stick the users with the costs of replacing the battery pack.
February 23, 2012, 10:31 amadmin:
I guess I am still confused. Either it is real irreversible damage to the battery, of it is simply a protection circuit shutdown without permanent battery damage. If its the latter, I understand why on a $100 ipod no one is going to bother to try to reset it, but for $40,000 it strikes me there should be some factory reset option rather than just saying that they need to buy all new batteries.
February 23, 2012, 11:19 amAndrew Meyer:
If you want to investigate this, you might want to contact Marc Fleury. He owns one (http://bit.ly/zcZiq3) and (http://bit.ly/zn9NmY) and I believe by training he's a physicist. He's certainly one of the most interesting entrepreneurs to come out of the tech boom and a fascinating guy.
It's pretty easy to reach him through his blog and it's the sort of thing that would interest him. Especially if you talk about your "weightless" experience.
Andy
February 23, 2012, 11:48 amNormD:
The IDC and Ycombinator "rebuttals" are not really rebuttals. They are anti internal combustion engine screeds with an added helping of anyone who does not follow their owners manual to the letter is a moron.
I keep an iPod Nano around as a unused backup. It constantly shows 100% charge until it suddenly shows 0% charge and needs to be plugged in and reset.
Just a guess, but if this happened to a car sized battery perhaps the "reset" would draw too much current for the charging system?
February 23, 2012, 12:12 pmMatt:
What I have read is that this is a possible issue with all lithium ion batteries.
The batteries internal structure can be permanantly dammaged in a way that drastically increases the probability of pyrotechnic failure if it is ever recharged after full discharge.
However most small electronics and (according to Nisan) the Nisan Leaf have protective circuts that both prevent total discharge and prevent recharge if total discharge occurs.
NormD's iPod Nano is probably reporting 0% charge when in reality it still has a 5% charge.
The Tessla's protective circuits on the other hand are set up to balance across the individual cells to prevent any cell from going into thermal failure. Even after everything else is shut off, the protective circuit itself continues to consume power and drain the battery making total discharge much more likely.
February 23, 2012, 12:39 pmDCSpotter:
TTAC has this good summary of the events as well as the battery composition: http://www.thetruthaboutcars.com/2012/02/the-tesla-roadster-bricking-story-details-deconstructed/
February 23, 2012, 12:44 pmlukas:
Warren, it's a protection circuit shutdown that kicks in because of permanent battery damage. You could charge the battery if you managed to circumvent the protection circuits and it would seem to work, but at some point it might blow up on you.
If NormD hadn't recharged his iPod when it showed 0%, the same thing would eventually happen to his battery and it would just refuse to charge.
February 23, 2012, 1:07 pmSean:
Found a little bit of chemistry over at Wikipedia. The normal operation of the battery at the negative electrolde is to reduce the lithium ion against a 6 carbon atoms in graphite to an Li-C6 complex. The Lithium metal is oxidized back to lithium in when it goes back the other way. At the positive electrode, you generate lithium ions when a Lithium Cobalt oxide releases some of the lithium ions and when reversed, reforms Lithium ions recombine with the Cobalt oxide. When the battery is stored for extended periods and completely discharges, excess lithium ions and the lithium cobalt oxide can react to make two separate oxides, lithium oxide and cobalt oxide. Once those separate oxides form the battery cannot be recharged.
February 23, 2012, 3:00 pmSean:
By the way, here is the Wikipedia page,
February 23, 2012, 3:02 pmhttp://en.wikipedia.org/wiki/Lithium-ion_battery
and here is another good link about battery durability
http://batteryuniversity.com/learn/article/how_to_prolong_lithium_based_batteries
Evil Red Scandi:
Not to be a jerk, but who the heck doesn't RTFM on an expensive purchase? If I spend five or six figures on a car or whatever, taking a few minutes to read the manual is almost always time well-spent. Especially for something that's based on relatively new technology.
February 23, 2012, 5:15 pmRick C:
Evil Red Scandi, but the Tesla owners manual doesn't say "don't run the battery to 0%, or you'll destroy it," it just says "take care not to run it to 0%."
February 23, 2012, 6:05 pmeddie:
Most types of batteries develop serious problems when allowed to discharge to very low charge levels.
Almost no battery is actually at "zero" when you think it is, i.e. when your flashlight stops working or your laptop shuts down. In most such cases, the falling battery charge has also lowered the battery's output voltage, causing the output voltage to drop below some threshold that your device requires to operate. Alkaline batteries output about 1.6 volts when full and fall off linearly to about 0.8v when empty; NiMH batteries output about 1.2v when full, stay at that level until they're almost empty, then drop off suddenly to 1.1v. Most electronic equipment is designed to expect 1.2v per battery and will shut down when they get less than that, so your "dead" battery may still have some charge left in it, just not enough to power your equipment.
A battery that's actually completely discharged all the way to zero charge undergoes a lot of awful things internally and is basically destroyed, regardless of what type of battery it is. For alkalines, this doesn't matter, because you're just going to throw them away when they're "dead". But for rechargeables, this matters a lot. A rechargeable that's "dead" but not at zero charge can be recharged; a rechargeable that's hit zero charge may well be useless (it may only be a little damaged, or it may be a lot damaged; even if damaged it may be recoverable to some degree; it depends on a lot of factors, plus luck).
With lithium ion batteries there's an extra wrinkle. If you get them too hot they catch fire and explode. If you overcharge them, they get too hot and catch fire and explode. And - very troubling, this bit - the type of damage that they suffer from being discharged down to zero charge is the type that will make them catch fire and explode if you try to charge them again.
So pretty much every lithium ion battery out there is manufactured with various electronic and physical safeguards that prevent the batteries from being charged too fast, being charged too much, or being recharged after being discharged down to zero charge. Because exploding batteries make for lousy PR.
Also, electronic devices that use lithium ion batteries (like laptops) will have various controls - either in the device or in the battery pack - to prevent the battery from being discharged down to zero charge. It'll shut down while there's still enough charge left in the battery to prevent the kind of internal damage that would make it unsafe to recharge.
However... all batteries also self-discharge, even alkalines. So even if your laptop shuts down and stops draining your lithium ion battery while it still has a safe (but not sufficient) charge, if you just let it sit there long enough it will eventually drain itself all the way to zero. And then it's toast. The internal protection mechanisms will kick in, and you'll be unable to charge it without bypassing those mechanisms and risking a fireball.
I'm not an expert; this explanation is necessarily somewhat simplified and undoubtedly wrong. I also haven't explained the chemistry or physics, I've only explained "how it works" at a very general level. If you'd like to explore the question in more depth, here are a couple of pointers:
http://www.buchmann.ca/Chap5-page3.asp for a bit about over-discharging solid lead acid batteries and lithium ion batteries. Click on the "previous page" link for a bit about over-discharging NiMH and NiCad batteries. Read the entire site to learn an awful lot about all kinds of batteries.
http://batteryuniversity.com/ - This site appears to have even more technically-oriented information than most battery-specific sites, although probably still not enough to land you a job designing new ones.
Enjoy!
February 23, 2012, 7:12 pmeddie:
As to why Tesla batteries could brick themselves but iPod batteries don't:
My guess, based on what I've read, is that Tesla's battery management system allows the batteries to come closer to being flat-out drained than the iPod's does. If 100% is a full battery and 0% is the point at which the internal safety cut-off kicks in thus bricking the battery, then (using made-up numbers here) the Tesla will let you drive the car until the battery is at 5%, whereas the iPod will turn itself off when the battery gets down to 20%.
In both cases, the batteries won't be bricks until their self-discharge takes them down to the safety cut-off threshold. But the Tesla lets you start much closer to that threshold than the iPod does, so it takes a lot less time to reach it than the iPod does.
Why does the Tesla cut it so close? In order to get as much range as possible. Apple doesn't care that much about squeezing out an extra hour of playtime, but Tesla cares deeply about having an electric car that people can use for long(ish) trips.
Again, this is just a guess.
February 23, 2012, 7:30 pmSean:
Eddie,
In addition to the things you mentioned there is also the environment. Most electronic devices never see huge changes in temperatures, hot or cold. They may on occasion but not regularly. A battery in a car might see temperatures close to or below zero winter or has high as 140F in a hot asphalt parking lot in summer. A car's battery will be exposed to extremes like these on a daily basis and they take a big toll on the longevity of the battery. I am impressed with how well modern batteries in electronics devices hold up but their normal environment is a whole lot gentler than a car's.
February 23, 2012, 7:54 pmIGotBupkis, Three Time Winner of the Silver Sow Award:
>>> Tesla cares deeply about having an electric car that people can use for long(ish) trips.
No, the OWNERS care, and Tesla is aware that they can't provide that without risking screwing up the car.
Nice choice, Tesla! Your namesake is spitting at you from the grave, and missing due to all the high velocity spinning...
February 23, 2012, 10:43 pmIGotBupkis, Three Time Winner of the Silver Sow Award:
Man, they followed the Idiotic Sports Car Design manual to a 'T', didn't they?
In addition to this, an even casual look at the car itself shows that its visibility in the 3pm-5pm and 7pm-8pm areas to be rather totally sucky. This is, surprise surprise, very common with overpriced sports cars that focus on looks over function during design.
...And here we find that the car has another stupid design flaw, in the behavior of the batteries. No, I frankly don't care if it's inherent in the tech being used, that only says the tech ain't "casual roadworthy" yet.
February 23, 2012, 10:59 pmJS:
I can speak with some authority as to what Apple does in its batteries. None of the contributors here is close in his speculation about what Apple does. I can only speculate what Tesla does, but I can assure you that Apple employs state-of-the-art protection circuitry in all of its batteries. Underwriter Labs, as well as the late Steve Jobs, would accept no less. For their small personal devices (phone and pod), you have to leave a fully discharged device (i.e., you used it until the battery "ran out") alone for a period of many months, perhaps a year, before the battery runs the risk of self destruction. If there were a better way to do it, Apple would have done it already.
If the stories are true, I find the Tesla time period for self discharge to be atrociously bad, completely unacceptable. It certainly sounds like they are not employing the usual safety chips in each pack, although I don't understand how UL would ever approve of them without. A standard battery protection chip, buried in each of the pack's thousands of individual cells, would certainly prevent such short self destruct times.
For an example of a commonly used lithium battery protection chip, you can see one here: http://tinyurl.com/7pfmdtd
Each individual battery includes a thermistor to sense the battery's temperature. Its sole use is during a charging cycle. The device's charger will prevent charging if the thermistor senses an out-of-range temperature, usually below 0 degrees C or above 50 degrees C. Charging beyond such temperature extremes would spell a rapid death to the battery cell, which is the purpose of the thermistor. Try charging an iPod that has sat in your car overnight during a snowy evening, and you will get a warning that the pod must be warmed up before charging can commence. That is temperature sensing for you.
@Sean: note that Apple specifies its operating and storage temperatures as 0 to +35 degrees C, and -10 to +45 degrees C, respectively. Yes, this is a far less stressful environment than what a car sees, although it's highly unlikely that these cars in question are being subjected to either extreme during their "self discharge" times. Does your garage reach 85 degrees C (185 degrees F)?
WIkipedia says that Tesla pumps coolant through their batteries when the engine is running and/or when the batteries are charged between 90-100% of full charge. So much for low-power "trickle charging" them, eh?
Electric car batteries can only be discharged down to about half of their rated capacities. If they went any further, they would sacrifice total number of charge/discharge cycles before the battery began showing signs of inability to hold a charge. If you limit the discharge to about half of the battery's capacity, you get about 1,000 cycles. If you fully discharge the battery, you may only get 300 or so such cycles. Speculation that Tesla routinely runs its batteries down to 0% would translate into a bad and costly user experience.
@Eddie: Apple does not toss away an hour of battery life. Battery life, "play time" if you will, is sacred.
@NormanD: you've got a broken internal wire on your battery. Nothing more mysterious than that.
February 24, 2012, 1:08 amKneil:
According to the jalopnik story below, Tesla leaves a lot of stuff on when it's off, whereas an iPhone seems turn itself truly off when its charge gets low enough.
The article tells of Tesla using a car's GSM connection to locate a car whoose battery was in danger of being bricked. It is implied that the car sent the alert to Tesla. I think the car should have texted and emailed its owner and then shut down all it's standby power draws.
http://m.jalopnik.com/5887265/tesla-motors-devastating-design-problem
February 24, 2012, 7:20 amRyan:
I don't know how the chemistry works, but I run some high performance RC cars that use Lithium-Polymer batteries that you have to be very careful with. You charge them in a fireproof sack in case they spontaneously combust and the charger has to keep the cell voltage balanced while charging to avoid the apparently guaranteed fire that results from a voltage differential. If you run the batteries down too far, you get the brick that you are referring to. But all that aside, the performance is blistering. Makes the nitro internal combustion engine RC cars look very weak. Unfortunately, you can't carry 3 spare batteries for the Tesla in your back pocket.
February 24, 2012, 8:21 ameddie:
@JS: Thanks for the additional information, and the correction.
Regarding cycle depth vs. cycle life - In following up on this story, I've read, um, let's call it "informed speculation" (i.e. random guys on the Internet that sound like they know what they're talking about but in all likelihood are just pulling stuff out of thin air) that Tesla does in fact discharge more deeply than other electric vehicles such as the Leaf, and consequently reduces the total lifespan of the batteries.
Wikipedia says that Tesla says the batteries have an expected lifetime of 7 years / 100k miles, and retain 70% capacity after 5 years / 50k miles. Wikipedia says that Nissan says the Leaf batteries have a "lifespan of 5–10 years under normal use", are warrantied to 8 years / 100k miles, and retain 70% to 80% capacity after 10 years. So it looks like maybe Tesla is a little more aggressive with their batteries, but not by all that much.
February 24, 2012, 9:42 amDoug:
Here is a link to a lecture by Jay Whitacre of Carnegie Mellon about lithium ion batteries.
http://www.ri.cmu.edu/video_view.html?video_id=60&menu_id=387
February 24, 2012, 11:23 amDruid:
Apologies for not reading all the above...
$100 fix simple engineering fix - when the sensors read 10% reserve light up the 'low gas' light, when at 5% shut the system down. Call it a 'governor' for a battery powered car - so it doesn't self destruct.
February 24, 2012, 7:23 pmCorky Boyd:
I use Li Poly cells (similar chemistry to LiIon) for model airplanes. If the voltage drops below approximately 3.0 volts per cell there can be an irreversable chemical reaction. I have nursed cells back by using a very low charge rate and they have held their capacity. The real damage is done at the high end. If you overcharge (somewhere over 4.2 volts per cell) the batteries are toast.
All lithium cells will lose capacity over time depending on the number of cycles. High heat is also killer. The closer you come to the high and low endpoints the shorter the life of the battery. The Chevy Volt has a 16 kWh capacity battery but uses only 8.8 to stay away from the endpoints. The hope they can live by their 100,000 mile warranty. Replacing those batteries could bankrupt the company.
I suspect the Tesla owners who had the problems used the full battery capacity or did significant fast charging. No matter what the battery manufacturers claim, Li Ion (cobalt) chemistry doesn't hold up long if they are repeatedly charged over 1x capacity.
February 24, 2012, 11:09 pmjay:
Part of the problem is Tesla's approach (the management there is a bit whacko, anyhow). They won't even sell you a replacement on time payment, nor will they offer any kind of insurange package. This in itself will kill any resale value for those vehicles.
What's also weird is that one person who used an extension cord was told that it failed for that reason. There are two problems I see here... first since it is claimed that only a small charge is enough to keep this from happening, why could not the system maintain a trickle charge with an extension cord. Also, why did the system not warn him that the charging rate was catastrophically low?
Realistically short cycle hybrids (which recycle braking energy) provide the most bang for the buck. The plug and drive electrics and hybrids are really trouble waiting to happen, huge cost with only incremental gain.
February 27, 2012, 11:39 amjay:
BTW Apple HAS taken some heat for its non user replaceable battery designs. Those products (as all rechargeable products) do brick occasionally and good sense would be to make the repair as troube free as possible
February 27, 2012, 11:41 am