Slashdot Mirror
Why Tesla Cars Aren't Bricked By Failing Batteries
itwbennett writes "Don't believe recent claims made by a blogger that non-functioning batteries in the Tesla Roadster cause the electric cars to be bricked, says IDC analyst Sam Jaffe. 'Here's the primary fact that the blogger in question doesn't understand: the Tesla battery pack is not a battery,' says Jaffe. 'It's a collection of more than 8,000 individual batteries. Each of those cells is independently managed. So there's only two ways for the entire battery pack to fail. The first is if all 8,000 cells individually fail (highly unlikely except in the case of something catastrophic like a fire). The second failure mechanism is if the battery management system tells the pack to shut down because it has detected a dangerous situation, such as an extremely low depth of discharge. If that's the case, all that needs to be done is to tow the vehicle to a charger, recharge the batteries and then reboot the battery management system. This is the most likely explanation for the five 'bricks' that the blogger claims to have heard about.'"
There seems to be a fundamental misunderstanding of the terms 'battery' and 'cell'. A battery is the collection of cells. So a Tesla could be bricked by a failed battery but it is tolerant to a failure of individual cells. This is not surprising.
Good thing slashdot is here to help us debunk everything I'd never have heard about from random dipshit bloggers.
It's a shortcoming of LiIon technology that if the cell becomes over-discharged, the cell may fail short circuit, and a subsequent recharge may cause an "exciting" failure (think flames). That's why all LiIon packs have a protection circuit that permanently disables the pack if it's discharged to the danger zone. Given the massive size of an automotive battery pack, it's easy to believe they have some very conservative safety devices in them. And it's also easy to believe that the cost of individually testing/replacing cells and "rebooting" the protection circuitry in a pack that has tripped its safety limits is prohibitive.
Well technically towing an electric vehicle, missing a clutch, would make it a generator, which could possibly damage the battery. However there is a youtube video showing a Nissan Leaf being towed and the battery being recharged. Don't try this at home!
Another option with the Tesla could be to lift the back wheels and tow it with the front wheels on the ground, unless there is some regenerative braking system which still acts as a generator. And yes, you want to lock the wheels if you do that.
the IDC analyst isn't "just a blogger", it's a guy who's trying to write a rebuttal to sound cool, but too bad he didn't actually go and try to look for the actual cases - instead he's just going "can't happen because of pr materials a, b and c". if he starts with that a battery of batteries can't have anything wrong with it by design.. comes off almost as a fanboi who didn't even read the news piece about the blog posting(which states that you can't even tow them, which sounds a bit strange but not _that_ strange if it won't fire up any elecs. of course you could tow them still lifting it on a truck or whatever)
world was created 5 seconds before this post as it is.
AHA!!!! SEE? They admit it!!!
I'm a leaf on the wind. Watch how I soar.
Just make sure you tow it backwards...
No sig today...
When you hear Lithium Ion Battery, you need to understand there are many different types of cell.
A battery consists of an Anode, Cathode and Electrolyte.
In LiIon based batteries, the electrolyte is a Lithium Salt, and the Anode is generally Carbon.
In LiPolymer batteries the electrolyte is held in a polymer of Lithium Cobalt or Lithium Maganese (this is the most common format of battery in consumer electronics)
In a recent project a for a hand held RF device, we chose LiFePO4. Mainly because it is so robust. Although it does not have the same capacity as LiPoly, you can grossly overcharge it and even drive a nail through it and it wont catch on fire. It also has much longer life over LiPoly.
LiPoly are very sensitive to overcharge, overdischarge, and mechanical damage, thus have a circuit to disconnect the battery when over discharged, thus the 'bricking' effect.
Tesla orginally used 18650 LiIon batteries with I believe had a LiCoO2 cathode, although I now think they are changing to pupose built cells. They would have a more sophisticated battery management that would prevent 'bricking'...... well at least one would hope...
46137
http://www.autoobserver.com/2011/02/towing-raises-new-issues-when-evs-are-involved.html
However... the Tesla vehicles already take advantage of "becoming a generator" as that is part of the "regenerative braking system" used in the vehicle. That ability to "generate electricity" not only doesn't damage the battery, but it helps to recharge the system as well and is an intended behavior... at least if you are going downhill with a tailwind.
I've seen several electric vehicles that have a gasoline-powered "pusher" trailer that provides "emergency power" for long haul trips instead of looking for an outlet for the car. It isn't even that new of an idea for that matter.
Regardless, because of the simplicity of the drive train and that the engine is not an internal combustion engine, calling a dead Roadster "a brick" is going over the top even if you can't disengage the engine from the transmission. Yes, there is a transmission in a Roadster, and there was even going to be a "clutch", but that feature was removed due to the torque issues and other problems from the supplier that was originally going to provide the transmission (something that nearly killed the Roadster when it went into production).
The Roadster is a rear wheel drive vehicle, so I don't think the front wheels are connected to anything other than the steering mechanism. In that regard, it is more like a conventional automobile too. In other words, towing the car is just like towing any other vehicle when you don't have the keys to unlock the transmission from the drive train.
It seems Jaffe only speculated the five bricked cars only needed servicing. Jalopnik did the research and also got an admission from Tesla. http://jalopnik.com/5887265/tesla-motors-devastating-design-problem
Actually, the "rebuttal" article seems to say that yes, the battery will become unrechargeable if you leave your Tesla car unplugged for too long....which is what the article it is "debunking" says as well.
The first article says, "This is a problem and Tesla should do something about it." The second article spends three paragraphs explaining that the original article gave a simplified explanation of how the Tesla works and is wrong about the possibility it discusses and how something the original article never said can't happen. Then it spends a paragraph explaining how the problem the original article actually talks about can indeed happen and how that is the nature of rechargeable batteries and there is nothing Tesla can do about it. Finally it spends another paragraph talking about how the original article is one of many that is spreading misinformation about electric vehicles.
The truth is that all men having power ought to be mistrusted. James Madison
Regenerative braking requires some pretty sophisticated power electronics, controls, and software. The Tesla's motor is an AC induction motor. (The AC induction motor was invented by Nikola Tesla.) An AC induction motor has copper coils for both the rotor and the stator. This is different from a DC motor (brushed or brushless) where (usually) the rotor has permanent magnets on it.
Backdriving an induction motor will result in no power generation unless the stator is energized. Even then, the associated power electronics have to commutate which phase of the stator is energized in sync with the spinning rotor. In other words, you need at least some external (i.e., battery) power in order to regenerate - this is true of all induction generators. Without the stator being energized, you're just spinning one set of copper coils past another set (this is different from a DC motor, where the rotor has permanent magnets, which will induce current in the copper coils).
So the Tesla cannot be "jumpstarted" by towing it or rolling down a hill if the battery has discharged so deeply that it has disabled itself.
that depends entirely on the architecture of the motor, motor drive, and battery management circuits. The Tesla roadster, for instance, uses an AC induction motor, which has no permanent magnets in it. Unless the stator is energized and properly commutated, backdriving the wheels will not generate any power. Even in the case of a DC motor, backdriving the wheels will generate power, but if the motor drive is disabled, that power won't backfeed onto the power bus. Even then, if the battery has discharged so deeply that it has disconnected itself internally, it won't accept power unless it first communicates with a compatible charger.
Yes, the claim seems to be actually true: http://jalopnik.com/5887265/tesla-motors-devastating-design-problem
When not plugged in or when plugged in with an unsuitable extension cord, the battery can run completely dry within a few days (one claim say that the battery can go from full to 50% within a week, another claim on that website says that according to Tesla, one car went from 4% to dead in a week).
I guess most people who buy an electric car will say "well duh, I know that when I have my car not plugged in, eventually the battery will be empty. I'll just have to recharge then". But I guess also most people will not know that the battery will be DEAD, as in "you have to get a new one, this one's DESTROYED" when it goes to zero charge.
Having a battery which can be destroyed in a matter of days if the car is not plugged in is a pretty big issue. An issue people really need to be made more aware of. Go on holiday and leave your Tesla at home? You better ask a friend to check every couple days if the car is still charging. Or what if you park your Tesla at an airport, plan to leave it plugged in there for a week, but on the first day you're gone, some kid unplugs the car "for the lulz"?
If you can afford a Tesla, you can afford to RTFM and get a trickle charger.
Probably (although I am sure that many people who have enough money to buy one won't read/understand all the technical stuff and will want the car to "just work") - but that still leaves us with the point that the battery pack can go kaputt within a couple days if the car is not charged (if the car was already at low charge) - which is something which needs to be communicated to customers far more clearly than just a sentence here and there that it is not good to let the battery go completely flat. Because I am sure that for almost every person who is not very familiar with battery technology, the EXPECTED consequence of a flat battery would be "recharge it again and you're good to go". If there is the possibility of making a $40k mistake, I'd expect the car to go full "star trek red alert" on me when I park it somewhere at less than 10% charge, and to start sending "help! I am dying!" SMS when the battery goes below 5%.
Just what we need--more "analysts" fighting bad bloggers' bad information with more bad information.
Let's start with the "more than 8000 individual batteries". These are 18650 cells (a standard form factor, a bit larger than an AA cell), and a Roaster has only 6831 of them. They are not "individually managed". Rather, they are grouped into a 69-parallel module, with 99 modules in series. (69x99==6831)
It is asinine and a distraction that Tesla (and everyone else) constantly obsesses about the 6831 cells. For all practical purposes it is a 99-cell Li battery, but rather than using monolithic cell modules, Tesla (like ACP before them) builds modules from smaller component cells, because they yield better cost ($/Wh) and specific energy (Wh/kg), with more-favorable cooling and safety characteristics.
Other than a built-in per-cell PTC device (which Tesla is likely no longer using), any "management" is done at the module level, and the battery is treated as a 99-cell series pack. The PTC is a passive cell protection device, designed to save a cell from a failed-short condition, but they also cause as many problems as they prevent.
Secondly, the "solution" is not nearly as simple as "shutting the pack down" when it reaches "an extremely low depth of discharge".
The Li cells themselves do not discharge themselves quickly when idle--perhaps 5 or 10 percent per year. However, small parasitic (e.g. maintenance) loads will slowly deplete the cells' energy. Herein lies the "grain of truth" that is probably at the center of this greatly dramatized "journalism".
Li traction batteries typically have on the order of 100 cells (or more for 600V systems), and each cell must be monitored to keep its voltage and temperature within a safe range. Typically the monitors are powered from the cell modules directly, and the competing design constraints are many: Small packaging, low cost, low power, electrical isolation, and so on. It is possible, but not trivial (nor cheap) to make a cell monitor draw zero current when its host module is at low voltage.
The original rant (er, blog) claimed that the parasitics would deplete a battery in 11 weeks, which is bordering on implausible, and if true, it would represent a staggeringly high rate of self discharge. Per Chelsea Sexton (who knows what she's talking about), there has not been "a single 'brick' story that didn't involve some extraordinary circumstances".
Lastly is the notion that the traction pack is necessarily destroyed by a deep discharge event. While it is true that deep discharge (and particularly cell reversal) will cause some permanent damage, the damage is in the form of higher impedance, and this is far from rendering the module useless. The battery can be brought back via trickle charging and a per-module impedance test will reveal if any are too far gone.
I've had this happen myself with several Toshiba laptop batteries that I left in a unplugged laptop for several weeks, and a friend of mine had to pay Apple for a new battery when his Macbook-Air was unplugged for a month (while he moved).
In the case of my Toshiba batteries, I was able to open up the two battery packs and construct one working pack from the remaining good cells. The bad cells were unchargeable. (I did this because Toshiba wanted $150 for a new battery pack and I was poor at the time.)
All of these cases involved lithium battery arrays connected to a battery management system where each cell is individually monitored for temperature and charge state.
The Tesla article may be bogus, but it has a ring of truth for me because of my experiences above.
Having RTFA'd, it is a negative opinion piece full of half truths and misinformation. That article is a horribly blatant attempt to discredit the vehicles from somebody who does not understand the underlying technologies. It needs to be taken as an opinion piece - there are zero references on his five examples (it's simply a chat with a service manager), and he is stating things as fact which simply aren't.
:P
First: yes, there are cases with batteries where they can be discharged to the point where the cells themselves are damaged and cannot be recharged. This is the case with *most* battery chemistries and is not going away any time soon. The blogger calls this a "Devastating design problem" when it is simply a part of the technology, like not storing your fuel cans near the furnace or leaving fuel to sit in a carburetor. There is a pretty clear warning in the car's manual not to let the battery voltage flatline for long, our intrepid blogger even provided a a PDF file with that page out of the manual. It states that the battery must be charged immediately if the charge level falls to 0% and has a great deal of information on the care and feeding of the battery.
Even if we take the five failed battery packs as truth, that is 0.2% of vehicles with an issue - an issue that in each example was due to the owner not charging the vehicle with one possible exception. His extension cord example could present a possible issue with the Tesla chargers. A typical cheap "heavy duty" extension cord will have 16 gauge wires, which over that distance is going to have some noticeable resistance. I don't know the current draw of the battery, but if it is expecting to pull 10 or 20 amps, the charger will see a significant voltage drop and likely cut off the charging (unless it has a "trickle charge" mode, dunno...). If the vehicle didn't report that it wasn't charging (or inaccurately reported it was), then I could see this being a design issue. I'll also note that this particular example did not state whether the customer had to pay for the repair.
What it comes down to is all electrical systems have ways they could be improved. That doesn't make this a devastating problem, it is simply an aspect of this class of vehicle that the owner needs to pay attention to. This blogger has a bone to pick or wants to stir the pot with a sensationalist report. Apparently it's working, after all, we both read his article
+1 Disagree