Hacker News new | comments | show | ask | jobs | submit login
Tesla Model S battery degradation data (steinbuch.wordpress.com)
267 points by dabeeeenster on Aug 31, 2017 | hide | past | web | favorite | 251 comments



Doesn‘t anybody find it curious that the source of data are range estimates displayed on the cars built-in dashboard?

For all we know, Teslas might be programmed to show more „optimistic“ estimates as the batteries get older.

I‘m not saying this is actually the case, but these stats are effectively manufacturer-provided numbers. Would you trust data like this if it came from any other car manufacturer?

If you have a gasoline car, do you trust the fuel consumption rate displayed on the dashboard? Should journalists trust it? Or should they measure how much gas they put in the tank and divide it by distance driven?


When I was reverse engineering car ECUs, one of the interesting little projects was finding fuel consumption display calibration data and calculation algorithm, on 2008+ Subarus. And then I did some measurements trying to get accurate real world consumption, and "sync" the dashboard with it. And while there was ~3% error IIRC, I think it has more to do with getting the most accurate average consumption over the car's lifetime, than fooling the customer on purpose. The calculation is based on the effective injector pulsewidth (ECU can't count actual fuel flow))), and as injector performance deteriorates over time, you get less and less optimistic values with more miles.


> ECU can't count actual fuel flow

Thanks, this is pretty interesting.

It's the first non-bullshit explanation I've seen for why cars tend to display inaccurate average MPG on the dash. I always assumed there was a fully accurate measure of fuel flow into the engine and there was something else broken (or dishonest) about that calculation.

This has come up in the press occasionally and the manufacturers never admit that they don't have a totally accurate measure of flow: https://www.edmunds.com/fuel-economy/your-fuel-economy-gauge...


> ECU can't count actual fuel flow

That doesn't seem right. Flow sensors exist and could be fitted to the fuel line to the engine for a precise measurement of how much fuel is consumed.

So I'm assuming manufacturers just estimate fuel flow in the ECU because a flow sensor would be additional cost for negligible benefit?


> because a flow sensor would be additional cost for negligible benefit?

Exactly. It's basically unnecessary given the other sources of information that the ECU has at its disposal.

It knows the target fuel pressure (some cars even have fuel pressure sensors) and it knows the flow characteristics of the injectors with respect to pressure and pulse width. That alone is enough for a decent calculation of fuel mass flow.

In addition to this, the ECU is also calculating the mass air flow into the engine, either directly via a MAF sensor or indirectly via manifold pressure, VE, and RPM (known speed density).

When it goes to target a specific air/fuel ratio, this is all the information that is really required. However, since injectors tend to deteriorate (clog, stick, etc.) over time, this initial calculation is usually “trimmed” by means of feedback from the oxygen sensor(s).

At the end of the day, there’s a fair amount of information available to the ECU which can be used to refine the fuel flow calculation.

The fuel consumption calculation on my 335i is usually within 0.5 MPG. However, my 2002 Maxima was about 3 MPG off.


Never saw a separate fuel flow sensor in an automotive application, but fuel injector is technically an orifice flow meter in this calculation method (pressure differential is constant and you calculate how much fuel is flown through orifice by measuring time it is open). Sure, when orifice size is reduced by lacquer depositions or dirt it will cause measurement error. But it will also cause increasing fuel trims (to the point when CEL goes on) and bad driving performance when fuel trims don't apply


I currently drive a 2015 Opel Astra. It shows an average of 5.1 liters per 100 kilometers (sorry, German, so it's metric units for me) over multiple months of driving in the dashboard. At the same time, I'm tracking every one of my gas refills using an app which also calculates consumption averages over arbitrary timespans. By that data, I am actually consuming about 6 liters per 100km, which is an almost 20% deviation. I thus stopped to trust the fuel consumption measurements of this particular car.

Before that, I had an old and small 2005 Peugeot 206. The measurements of that car were like exact matches (at least to the first decimal digit) with my manual consumption tracking.

What really baffles me is that the newer and more expensive car is the one with the huge deviation, not the older and cheaper one, as you might expect.


I'd put this in a typical market drift. New generation are built with a lazy cheap mindset, not the old generation "lets try to make something good (probably because there was more love and motivation for said product at that time)".


An independent third party can do accurate measurements and determine the difference between the dashboard measurement and the actual capacity; I like to think that if the difference was too high, people would've noticed by now. If the difference is too high and beyond the range of a simple measurement error, then I'm pretty sure it's fraud like VW's emissions scandal and Tesla would get millions if not billions of fines from the EU. But I'm sure Tesla wouldn't let it get that far - it could probably be fixed via a software update.


I fill up every time I refuel and I take notes of the quantity and odometer and the average matches the dashboard.


There is a website you can use to keep track of this: http://www.fuelly.com/


I take a look at my estimates in the dashboard on my Mazda2 (Germany) and also calculate the metrics manually every time I fill the gasoline. Funny thing is, that I tend to use less in the real world then my car makes me believe.

My car tends to value the long range drives on a highway way higher then my average driving habit while commuting. So the higher consumption outliers on the highway tend to skew my results.


What manufacturer and model is that?

I had a 2008 VW Golf TDI Gt Sport (the 140PS PD engine) and that consistently lied. My newer 150 Euro6 (I think) engined VW Beetle TDO gives even better MPG so perhaps it is lying again... Potentially time to measure it versus how much I put in.


How do you measure the distance driven? Perhaps the tachometer is also lying?


Very good point. Odometers (tachometer measures RPM, odometer measures distance) are not as reliable as some may believe, and even if they're perfectly accurate, there are so many variables that it's hard to control. Smaller tires spin faster, which drives up the miles read. Tires get smaller as they wear, so the odometer can be calibrated for new tires but will quickly become inaccurate.

But the odds that your odometer is accurate is very slim in reality: http://www.azfamily.com/story/19769079/tests-show-odometers-...


I trust my range indication in my petrol car to be conservative. It has about 30km more range than indicated - imagine the uproar if you planned a journey it said you could do but ran out of fuel? It’s quite handy as there are some areas I’ve driven through have no fuel stops for over 100km, and I’m sure the US has quite a few areas like that.


I have a Volkswagen Golf GTE. My dashboard claims it can run ~47km on a charge, and it adjusts whenever you change your AC settings. When I bought it, VW claimed 50km of range.

In the 60.000km's I've driven that car, I NEVER were able to get even close to 40km on a charge.


That's because of the "fuel reserve" in your gas tank. You can Google that term to learn more.


I know it's anecdata, but I found the dashboard estimate on my VW (heh) strangely correct. It's off by +/- 5% in most cases. Since it doesn't take upcoming traffic/terrain/wind into account, I find it more than reasonable.


Anecdotally, I run a 2016 model year ford fusion 4cyl turbo and its fuel economy gauge has never gotten above 14.5mpg in the year I've had it. Not exactly numbers to boast about. I trust it.


I trust the fuel consumption in my car, yeah. If it says I've been using 12l/100km on average and I've travelled 500km, you can bet I'm filling up exactly 60l at the pump.


That really depends on the car. I don't own a car so often use rental cars. Most cars' consumption is within 10% of the shown values but at least every fifth car I drive has vastly different numbers.

I think it is not necessarily due to the car but because of the calculation methods they use. If you drive motorway and city, consumption is accurate. But as soon as you start driving in the mountains or a mix of different terrains, data often vastly differs.


If the car has a Mass Airflow Sensor (MAF), then it can more accurately calculate the fuel usage. The majority of older / cheaper cars use a Manifold Absolute Pressure Sensor (MAP) and have to make some guesses at what the actual fuel usage is.


I've had a car that consistently showed 0.5 l/100 km less than the real consumption.


The Tesla is strange because that number on the dash is "rated miles", or the best you're going to get. So you can't just drive to empty and get a number you can compare with others, unless you drive it in quite a certain way.

My S60 says something like 215 miles when fully charged, but when driving on the Interstate the actual range is more like 160 miles, with the cabin heater on. It can vary significantly due to rain as well, a recent trip that had a leg almost entirely in very heavy rain caused a significant drop in range.


were you driving it manually, or using Autodrive? Were you driving it conservatively, or heavy on the accelerator?

Did you stay at a consistent 55 - 60mph, or were you going faster?

There are a lot of factors in play for a electric vehicle.


Can't you measure how much energy is used during charging?


> The red fitted line has a slope above 60.000 km (say 40,000 miles) of 1% per 50.000 km (30,000 miles).

I'm going to bring this up because I think it's both interesting, and sometimes it comes up on its own and detracts from the conversation with conjecture (and sometimes harsh assumptions) until covered adequately[1] - different countries have different conventions for whether they use a comma or period for a thousands separator, and in those languages if they also speak in English sometimes, they may choose to continue using a period as a thousands separator as is convention in their country, even if the predominant language in their country is not English, and there are circumstances which might make that more common for some people.[2]

In this case, we have what appears to be switching convention when switching unit systems, which is interesting. I'm not sure if it's more or less confusing in the end, but there's not really any standardization of this as far as I know, so it's all good.

1: https://news.ycombinator.com/item?id=14970002

2: https://news.ycombinator.com/item?id=14862695


>but there's not really any standardization of this as far as I know

There is. Resolution 10 of the 22nd General Conference on Weight and Measures [1] (the body responsible for the SI system) forbids the use of comma ordot as thousands separator. Other institutions like NIST have followed suit, forbidding or at least recommending against it. The recommended approach is to use spaces as thousand separators (or ideally: thin, non-breaking spaces).

You are still seeing a lot of people do it wrong because the resolution is from 2003, and these things take decades to overcome entrenched preferences.

1: http://www.bipm.org/utils/en/pdf/Resol22CGPM-EN.pdf


Interesting reading. Resolution 10 only declares the decimal separator, leaving that to language convention still whether to use comma or dot (as opposed to finally unifying that, wouldn't that have been something?). As for thousands separators, it just reaffirms Resolution 7 of the 9th CGPM from 1948 which made the declaration of spaces, never dots nor commas.

So it's actually been since 1948 that the GCWM/SI system has been recommending spaces as grouping separator? Wow.


[flagged]


As a side note (not to you particularly--I see this a lot), you can't delete a comment once a reply is made in order to preserve context to the replies. This is probably for people like me, who come after the replies are written.

With this in mind, why side-step the inability to delete your comment by just replacing it with something else? It's technically allowed, but isn't that against the spirit of the original rule?


The comment was losing points and thus obviously incorrect. I'm not going to let it sit around and burn internet points!


You have thousands of points so who cares, and disrupting the conversation is only going to get more downvotes. How about editing in a retraction, without deleting?


I'm curious at to when that example would ever be suitable; memory sizes should be grouped in 3s as well.

The only one I can think of is phone numbers, e.g. Australia is a 4-3-3 grouping for mobiles. But a space has been the standard delimiter there anyway.


Dashes were "the standard delimiter" for a long time for Australian phone numbers. 451-4153 was my phone number growing up (and yeah, I'm showing my age with a seven digit phone number... A friend of mine was on 99-0066 up until the early or mid '90...)


Interesting. As a contra data point, I grew up in Sydney with 86 2050 as our landline (as if there were any other types of line! :) and I never saw hyphens used within phone numbers. I worked for Telstra during most of the 1990's, and didn't see it in use there either. Might be a regional thing? I note that our North American cousins enjoy splattering hyphens through their phone numbers. I like the space approach.


Fair enough. I grew up overseas, so my local knowledge only kicks in in '99.


Frankly it might be a bit confusing to Canadians, here we use the U.S. numeric conventions, but we tend to use a mix of SI and imperial units, and in this case we'd use kilometres for ground distance, with thousands separated by commas. The UK also seems to prefer commas to separate thousands.

I'm inoculated against confusion about this because I've done a lot of software and content localization, and have a grasp on the specifics for a lot of different cultures.

I think the problem is exacerbated by the underuse of the SI system's prefixes. Why even talk about thousands of kilometres when you can speak in tens of megametres?

Added:

I figure it's getting to the point where there are too many genuine English locales to keep up. It takes a lot of energy for normal people to retain fluency in dialects and locales like en_US, en_IN, en_CA (thankfully rapidly merging with en_US), en_GB, en_AU, en_JM, en_MY, etc. etc. etc.. all at the same time. Maybe at some point there will be a great and natural convergence.


I'm always surprised that en_CA and fr_CA are so different. We use spaces as separator and comma for the decimal. Also we use the 24h format while the ROC uses 12h. We do use 12h colloquially though. Dates are officially y-M-d but most people use d/M/y.

It's funny how in Canada we have some specific remnants of the imperial system. Most people under 50 have no idea what a Fahrenheit is (myself included), except for pool water and oven temperature. Go figure.


The funny thing about the imperial remnants is that for almost all general tasks, Canadians still use pounds for weight (groceries, body mass, vehicle weight), but we don't use ounces. We use feet and inches to measure people, but metres for almost everything else, except nautical speed. Litres for gas in cars, sometimes gallons or pounds instead of litres for gas on boats. Pints for blood and beer, ounces for hard liquor, litres for all other service and retail fluids.

People complain about inconsistency and lack of standards in the use of measurements in the U.S, but it is actually much more consistent there. It turns out that decimal units aren't really better for any practical purpose in a typical person's life, and it's certainly worse if you need to know both!

Even in the U.S. of course, engineers, mechanics, plumbers and others still have to deal with a mix. Think metric-based and imperial-based bolts, metric v. imperial lumber sizing.


> Even in the U.S. of course, for engineers, mechanics, and some others they still have to deal with a mix.

s/U.S./the entire world/

It's very common to have international U.S. based companies offer N types of fasteners/whatever in imperial sizes, but only N/2 in metric. So we frequently build stuff in imperial sizes because of greater flexibility.

And don't get me started on (non)tapered pipe threads... I've heard some horror stories of huge subassemblies built in Europe using BSPT and shipped to the U.S. for final assembly, where of course everything else is NPT, so some poor sod has to spend a month replacing all subassembly fittings.


Dates are a complete mess, yeah. I've seen absolutely every single convention used in Canadian government and business documents. I personally use ISO 8601 wherever it's an option (contracts, personal documents, casual communications) because of how ridiculous all of the other options are.


Agree about 8601 except that not ALL other options are silly. Writing the day with words is kind of nice in some writing, like "September 1 (Monday), 2017" is not too bad.


> Why even talk about thousands of kilometres when you can speak in tens of megametres?

Probably because when you're speaking about them colloquially, you don't say "kilometres" you say "kays". That probably varies by locale too, I'm in en_AU. How is it over your way?

I do like the nerdiness in the idea of saying "two megs" instead of "two thousand kays", but it's never going to happen.


> I'm in en_AU. How is it over your way?

It's mixed here in en_CA (I've lived in Calgary, Vancouver, and Toronto at different points). In my experience, kay is used for kilometres when talking about manual travel/exercise (bicycle, running) as in "couch to 5k", but it's almost always klicks or kilometres when talking about driving, and usually only kilometres when talking about most other forms of transport.


We occasionally use 'klicks' interchangeably with 'kay' or 'kays', but it's pretty rare.


It can be more confusing. US date - m/d/y. Distance in miles. If you are in the US military - d/m/y and klicks (km). Was in UK a few weeks ago and rented a car, mph for speed, everything else (except beer) in metric.


The UK is more complicated than that. Most of the signs are in metric but some distances are in yards and heights are in feet. Petrol prices are in litres but consumption is in mpg, even though there's no indication on how many gallons you actually fill up. In stores, everything has to be metric because of EU regulation but packaging for liquids is often in pints (e.g. milk, beer) and then gets translated into litres.

Temperatures are usually in Celsius but are shown in F once it gets hot (100 degrees sounds better than 38).

Even though I'm a fan of the metric system, being in the US is easier than the UK, at least measurements are consistent...


Eh, not everything is consistent. Soda cans are 12 fluid ounces, but bottles are 18 ounces, or 1 or 2 liters. Hard liquor comes in 750ml bottles, but is served in 1.5 oz or 2 oz shots. Beer comes in 12, 16, or 24 oz cans or bottles, milk in gallons, quarts, and pints, etc. British pubs tend to serve beer in imperial pints (20 imperial fluid ounces or about 568ml) instead of US pints (16 US fluid ounces or about 473ml) but other bars serve US pints.

There are other oddities and mixed units. Dates are in "drunk endian" with month/day/year format, indoor areas are measured in square feet and outdoor areas in acres... Small units of length are in fractions of an inch unless you're a machinist at which point they're in "thous" or "mills" (same thing, 1/1000 inch) or "tenths" (1/10000 inch). Drill sizes are fractions of an inch for some common sizes and large diameters (27/64 and up are always fractional), but otherwise are numbered starting at 80 and decreasing down to 1 as diameter increases with the spacing between numbers being inconsistent, except for a series starting at 2340 thou and ending at 4130 thou which is lettered with 40 thou between letters (mostly, some inconsistencies here). But you can still get drills in some of the fractional inch sizes that fall between the letters. Seriously, look at a tap drill chart[1]. And of course if you're a machinist you'll also have to deal with metric items imported from the rest of the world or made for sale there so you'll also need all the metric drill and tap sizes.

[1]http://www.brokenbolt.com/images/starrett-inch-metric-tap-dr...

Electric systems use SI units, except that PCB designers tend to use thou for trace sizes and spacing. Some components have pin spacing in thou, others in mm. Getting the PCB to look nice with a mix of pin pitches and grid sizes is more of a chore than it should be.


Oh yes... I design PCBs. <bangs head on desk>


And stone for weight of people.


Western Canada, we use a short form pronunciation similar "Kay" too (pronounced like just saying the letter "K").

"K" is short form thousands as well written and pronunciated "k".

Ex:

20,000 = 20k.

But when talking about distance specifying units, you say the thousands and "k" as in short form kilometers.

20,000km is pronounced: "twenty thousand k" or kilometers.

But if not specifying the unit, you could say I drove twenty "k", as in "I drove 20 thousand (kilometers)"


>I do like the nerdiness in the idea of saying "two megs" instead of "two thousand kays"

But they are not the same!!oneoneone. Unless you are in the dog food prefixes camp. Are you?


2 000 km = 2 Mm?

You're not thinking of the mebi-, kibi-, etc debacle, are you?


Yes. I'm old school, 1K is 1024, 1M is 1024K, 64K is 65536 and I refuse to use kimblebits and such.

Ah... screw it: https://www.xkcd.com/394/


I'm with you, but it's a non-issue when it comes to non-computer units, which is what we're discussing here.


Decimal prefixes are standard prefixes for SI units.


Born in canada, West coast, lived here still for 30+ years.

English as first language, but 2nd/3rd language for immigrant parents.

I have very rarely seen space separator for thousand seperate, maybe one some bank drafts.

Everyone I know, and I have used the comma seperate.

Officially may be a Canadian thing, but not used really


It's not just about the thousands separator, the same character is used to separate the decimal part.

It's better to use no punctuation at all if you want to internationalize.


Actually, if you want to internationalize for the thousands separator, spaces are likely pretty safe as they are preferred by numerous standards bodies, as well as AMA.[1] That also neatly sidesteps any comma/period confusion for the decimal in large enough numbers, since whichever you see, if there's space grouping thousands, it's that one.

1: https://en.wikipedia.org/wiki/Decimal_mark#Digit_grouping


When the period is used to separate thousands, the comma is used to separate decimals. Usually...

https://docs.oracle.com/cd/E19455-01/806-0169/overview-9/ind...


Of course, that page is completely wrong about the large number convention in Canada. Anglo Canada (including government documents) uses the U.S. convention, and at very least uses period for the radix point. The shown convention only applies to Franco Canada, and maybe not even all of Franco Canada.


I see both in Quebec, so it isn't universal - even there. Meh, it's close enough to demonstrate. I prefer the US method, but that's just due to familiarity.


That's quite impressive. What I am wondering is why they do do well, barely losing 10% in 5 years, compared to phone or laptop batteries that last maybe 3 years on average. AFAIK, they use the same technology, in fact, I wouldn't be surprised if the same batch of cells end up partly in Teslas and partly in laptops.


In order of importance:

1: Tesla cells are hard limited to a lower voltage, reportedly 4.1 V rather than the 4.2-4.4 V maximum. Phone batteries run much closer to their maximum voltage

2: Tesla cells stay at much lower temperatures and have active cooling. Phone batteries are used to sink heat from the processors and wifi.

3: Teslas have a much lower normal depth of discharge. In addition to the derated voltage, the average American's daily driving distance is ~30 miles or 10-15% of the battery capacity. Phones are often almost totally discharged every day. Tesla's are also typically only charged to 80%, although you can turn that off.

4: Teslas have a low average power use: 300-600 horsepower on tap but cruising on the highway uses very little. In technical terms their current rate is ~.2-.3 C, while phones can have much higher C rates.

5: Phones use an LCO chemistry, Teslas use NCA which is more stable over the long term. Tesla also has custom additive and cathode/anode materials that improve lifespan a great deal.

Phone and laptop batteries almost all come from china, Tesla batteries are korean. They're pretty premium as far as 18650s go so they're used in power tools and really good laptops. However Tesla has their own proprietary chemistry so nobody else gets to use their blend.


Tesla uses batteries from Panasonic (Japanese) [0].

[0] https://electrek.co/2016/11/02/tesla-panasonic-2170-battery-...


Thermal regulation is the most important difference, followed by low depth of discharge.

There are also some (rather weak) reports that if you take the individual 18650 cells out of a tesla battery they last significantly longer than the NCA 18650s you can actually buy from Panasonic, which indicates that they might made to better manufacturing tolerances or something, but it's really all about Tesla's treating the batteries much better during use.


I think this is a good list, but I think reason 3. should be higher up. Rule 3. worded differently:

3. Rule of thumb is that a battery lasts around ~1000 cycles. Phone batteries are discharged fully about once a day, so that leaves a lifetime of almost 3 years.

Tesla range is ~ 300 miles, so using this rule of thumb it should last around 300,000 miles. This + reasons 1,2,4 and 5 are why the battery barely degrades over the lifetime of the car.


A lot of these could be mitigated if phone makers were willing to put in a slightly larger capacity battery and apply more conservative max-min limits in their battery management system. But I guess companies are happy to keep sales up via planned obsolescence.


Tradeoff between weight, battery life, and battery lifetime.

I'm glad phone manufacturers heavily optimize for the first 2 and not the latter.


I'm sure that one of the Tesla founders stated that one of the reasons that device batteries failed after a couple of years is because longer life wasn't prioritized by the companies buying them and so not much effort was put into that dimension.

You can spin this in a bad way ("planned obsolescence") or a good way (they were optimising for other things that customers valued), but the point remains the same, if car manufacturers (or grid storage units) require longer lifetimes, then it's not a hard physical limit at the current lifetimes that's been hit yet.


My phone rarely gets below 50%, and I'd say it only gets below 20% a handful of days a year.

But yes, I've often thought it would be nice to have a toggle to say "Prolong battery life (as in years)... I only want a 90% charge since I'm not going to be using much tomorrow."

I've wondered too whether I should only use a slow (1A) charger for the overnight charge rather than the full 2.4A. But then again, I only keep my phone for 2 years before passing on it, so I've not had to care that much.


I have a Lenovo laptop with this feature. It only charges the battery up to about 60%, which is fine since that laptop is plugged in a lot. When I am planning to take it out I'll turn off battery safer mode and let it fully charge.


It looks like Sony's Android phones have this option https://www.phonearena.com/news/Sony-Xperia-Marshmallow-upda...


I know ThinkPad laptops have this (mine is set to charge up to 90% and only start charging if it's below 80%), and I think I've seen it in some custom kernels for Android ROMs a few years ago, but I'm not aware of any popular devices supporting something like this.


Sony phones have that, I have an Xperia Z5 and you can select an option to "prelong battery life", where the phone only charges it up to 80% and stops(the scale shifts, so 80% shows as 100%, but internally the battery is never at 100%).


Laptops charge fully, discharge fully, get hot, and spend most of their time fully charged. This is the worst possible combination for Lithium-Cobalt/polymer batteries. Tesla batteries are never fully charged or discharged, are often charged at a lower C rate, and are thermally managed. A laptop with a cooled battery that only charged from 15-90% and took 3 hours to charge would hold up quite well, but it would weigh a ton and you'd hate to use it.


Now you made me want to build a luggable computer, maybe with wheels like the large carry-on-baggage with a ton of well-managed batteries in it!


A portable and clean replacement for a small gasoline/diesel generator in luggage form would be nice. So your idea is not too unrealistic.


This product genre exists: check out GoalZero Yeti, Anker PowerHouse, etc.


The real killer for laptop batteries is the fact that they are kept at 100% charge plugged in while under extreme thermal stress. Temperature + full charge kills li-ion batteries. Do your laptop a favor and never charge to 100% unless you really need to.


Really? If this was the case, I would expect laptop providers to allow you to set a top charge rate for those like me who have their laptop plugged in 90% of the time.


Some do (ThinkPads).


To Apple MAC developers: can we get this feature?


I have a broken charger that will do this for you. It only provides power and doesn't charge the battery anymore.


If you have a MagSafe connector, cover the middle pin with electrical tape.


Which does what exactly?


blocks the data bus (i2c or whatever it is) used to negotiate with the power adaptor. Will cause it to draw power, but not charge the battery.


My Toshiba has the option of only charging it to 80% while plugged in, called Cell-Saver. Just feels better not having to worry about cell degradation.

I mentioned this to a colleague lately who looked at me like a strange animal for having:"worry about cell degradation" even on my list.


That's completely false. It was true though with NiMH batteries


>That's completely false. It was true though with NiMH batteries

No, it's not. NiMH have practically no degradation from cycling, which is why Toyota still uses them in the Prius. The tradeoff is in energy density.

Li-Ion batteries are not meant to be held at 100% State of Charge (SoC). There is a non-linear relationship between average SoC over the life of the cell and the rate of degradation. Optimal charge for extended life in a typical 4.2v cell is around 3.9v, or 65% of available stored energy. Keeping an average SoC near 100% can cause up to a 5x loss in battery lifetime. See the link below [0]. Specifically "Table 4: Discharge cycles and capacity as a function of charge voltage limit."

http://batteryuniversity.com/learn/article/how_to_prolong_li... [0]


I believe this is due to the fact that by default Tesla recommends that you don't fully charge or discharge the battery if you don't need the range. For phones people charge them all the way everyday (or more) and often discharge them all the way.


How can we prevent a battery from fulling charging? We plug a car/phone in at night, and wake up to it 100% charged.


Tesla's cars have a charge limit you can set, after which the car simply stops. I believe they come by default at 90%. You're supposed to override it if you want to go on a long road trip.

Would be cool if there was a similar feature available on phones. Maybe you could rig up something with Tasker or the like on an Android.


It looks like Sony's Android phones have this option https://www.phonearena.com/news/Sony-Xperia-Marshmallow-upda...


Interesting fact.. with the last software update they changed what was considered trip, and local pretty significantly. I wonder if that is aligned to a SW improvement in battery management?


I'd imagine it is down to the number of recharge cycles. A phone tends to go through an entire recharge cycle every day to two, whereas I'd imagine a Tesla wouldn't go through it's entire range as often.


If you check the interpolated cycle life at the source, the oldest Teslas have gone through <800 full discharge cycles, so only ~3 years of a phones life. However the discharge decreases nonlinearly over time so the Teslas have lost <10% of capacity while a phone loses >30%.


This is pretty surprising. What it means is that most owners will not need to replace their batteries in the time that they own the car. In my 'total cost of ownership' models, I've assumed a battery replacement at 100,000 miles, but that appears to not be necessary. This changes the calculation considerably. Not to mention that if the drive train can go 400,000 miles without needing major repair, I can keep the car 2x as long as an ICE, further reducing the cost of owning an electric vehicle.


Musk has made a comment on an investor call that they're targeting a million mile lifespan for the motor.


I think ICE cars are just not optimized for prolonged life. Many countries have policies discouraging long lifespans, more than ever since emmissions laws. Also, EOL for a car is several owners and 5-10 years removed from first purchase.

This adds up to a realtiy where doubling the expected lifespan is not worth much more, in sales.

Basically, if Tesla want to make this a goal then beating benchmarks should be relatively easy.


x2. Modern vehicles have a targeted minimum lifetime based on certain usage models. That's why people who are driving a pickup truck around all day for work with a few hundred pounds of equipment (e.g. medical equipment service technician) get 500k+ out of them and oil field trucks are beat before 100k.



This is fundamentally flawed because lithium batteries mostly degrade not because of amount of use, but because of time (and their charge state and temperature during this time). No matter if used or not, they will lose capacity. The rate of capacity loss is mainly dependent of average charge (best is 40%, worst is 100%) and temperature (best is 5 degrees C, the higher the worse).

(This is also why phone batteries fare so badly: they are often near 100% charge at 36.6 C in your pocket)

http://www.sciencedirect.com/science/article/pii/S0378775316...


I do not believe that the passage of time is the main cause of lithium ion battery degradation. The source that you linked, http://www.sciencedirect.com/science/article/pii/S0378775316..., does not seem to support that idea either.

Figure 3 from the paper, "Cause and effect of degradation mechanisms and associated degradation modes," shows time as just one of 8 degradation mechanisms affecting lithium ion batteries. Other figures/tables from the paper do not emphasize time as the primary degradation mechanism, nor does the body text. Nor is there any explicit passage-of-time term in the final diagnostic model that the authors develop.


Phone batteries do poorly because the SoC generates a ton of heat when used which gets dumped right on the battery pouch.

It's the same reason that you see early Leaf cells die out in Arizona while Tesla's have held up to 100k+. Thermal management makes or breaks longevity.


yes... I said that...high temp is bad


Sure, but you lead with:

> not because of amount of use, but because of time

Which makes it seem like time is more important that heat.


heat != use

think of it more as

BADNESS = integral(abs(temp - 5), over time) * integral(abs(charge_amt - 40%), over time)


heat = (discharge + charge)

use = (discharge + charge)

hence: heat = use ;)


To be fair for phones you focused on: 100% charge and pocket (body) heat.

You did mention 100% and temperature being the problem generally though...


It seems like the opposite might be true, and that batteries degrade from cycle count and not because of time. This is why they are rated in cycle counts (i.e. cycles until 80% capacity = life time of battery). Also important is the charge/discharge rate, but because the batteries are so oversized in Teslas they can discharge as low as .25C even at highway speeds which helps increase their lifetime.

source: http://batteryuniversity.com/learn/article/what_causes_lithi...


They are rated in cycles so consumers do not need a degree in physics to understand the reality. Feel free to inspect a REAL spec sheet for a LiPo battery from a manufacturer (I think ATL provides some online)


Can you provide links? You clearly feel we have never seen a "real" sheet, perhaps more hand holding is order. I would appreciate the experience and would not do it without the link.


Here's a "real" manufacturer datasheet. The spec the battery life in cycles. Op may be correct, I don't know battery chemistry, but I've never seen batteries spec'ed the way the op describes in any "real" datasheets I've read.

http://www.omnitron.cz/download/datasheet/NCR-18650PF.pdf


I am trying to find some non-NDA specs online for you. I saw this when I was working with these for a job and the specsheet was NDA (and I no longer work at the place this was)


Lithium batteries lose charge with use as well. Also rate of charge/discharge. It's more complicated than just one thing.


For about 16 hours a day, my phone is somewhere between 100% and 0% charge, probably around body temperature or a little higher.

For the remaining 8 hours, while I'm sleeping, it's at close to 100%; which is not ideal. I'd be curious how effective an overnight charging scheme would be that charges up to 40%, then holds it there until say 4 a.m. (or some other time based on the phone consumption, capacity and charger output) then charges up to 100% so that when you wake up it's fully charged.


So if you were optimizing your phone for low battery degradation, when would you charge it, and to what %? If you were trying to keep it around 40%, would a good strategy be to let drop to 20%, and then only charge to 60%, or would you bias it even lower to keep it away from 100%, like a working range of 10-50%?


The ideal state of charge for a lithium battery is 50%. So charging between 20% and 80% would significantly lengthen battery life. If you wanted to completely optimize for low battery degradation you'd charge between like 49% and 51% and only have a few minutes of usage, but that kind of defeats the point of keeping a lot of battery capacity.


Maybe schedule it like so:

Between 6AM and 8AM: Charge to 100% All other times: Charge to 75%

So the battery is topped off in the morning right before you put it in your pocket for the day, but otherwise doesn't fully charge to save the cell life.


They have a chart for calendar degradation. It's not weighted by mileage, but there are none of the datapoints you'd expect for that proposition (low miles, high age, high degradation).


Wow, I've never heard anybody say that before or model it that way before.

Though these tests are evidence of that, it's definitely not a common belief.

Before the Tesla data, what informed you?


In the RC world, the '50/50' rule of thumb is common for lithium battery storage. 50% charge at 50˚F. It's close enough to the 'ideal' values, and easy to remember (for those of us used to a human-centric temperature scale)


> human-centric temperature scale

One point (100 °F ~ 37 °C) doesn't make a scale.


Heh, sorry, that was just a tongue-in-cheek joke, not a central point of my comment.

0 to 100 on the Fahrenheit scale is "Really Cold" to "Really Hot".

0 to 100 on the Celsius scale is "Kinda Cold" to "Unsurvivable Hot".

When I say it's "human-centric" I mean that it maps well to the range of human comfort (Celsius is obviously vastly superior in just about every other use).


I'm not so sure I would call 100° C "unsurvivable". People go in saunas with 100° C all the time for enjoyment. If the air is dry enough, it's not that hot.

Maybe it's unsustainable, you can't survive that indefinitely. But you can't survive 0° F indefinitely either, unless we bring protective clothing into the game.


100C would send you to the burn unit. I think you mean 100F.

Edit: 100C is boiling point for water - is the ambient temp of the Sauna 100C?


I'm thinking of finnish-style saunas, which operate at 60 to 100 °C [1] (with 100°C being a frequent option). You usually stay in them 10-15 min at a time, but I guess if you keep hydrated you could stay a lot longer.

Finnish saunas are kept very dry. Sufficiently dry air makes very hot temperatures bearable because evaporation cooling from sweat gets more effective the dryer the surrounding air is. Water on the other hand makes sweating useless, which is why you can't tolerate water temperatures above body temperature (depending on how much of you is submerged).

1: https://en.wikipedia.org/wiki/Finnish_sauna#Types_of_sauna


"Air temperatures averaged around 75–100 °C (167–212 °F) but sometimes exceeded 110 °C (230 °F) in a traditional Finnish sauna." [0]

Apparently so! TIL.

[0] https://en.wikipedia.org/wiki/Sauna


I'm not sure that the 0-100 thing matters that much.

I'm in the UK we switched to C for temperatures back when I was a kid (I still vaguely remember both been used on weather charts back when they stuck the things on by hand..), 0 is cold, 10C is brisk, 16C is perfect, 20C is warm, 30C is hot, 35C is "kill me now".


I provided a citation. Feel free to google around to find more info. This is basically how everyone models LiPo and Li-Ion cells. Tesla only publishes marketingspeak (which makes sense - that makes them money). Look for actual science and you'll see


Weird, was that there when I replied? I definitely don't remember seeing it there... Perhaps I was just rushed.

In any case thanks for the source!


I've recently had to become pretty familiar with this subject for work, and your comment is only partially correct.

Li-Ion battery degradation is generally modeled as two (roughly additive) components, called calendar and cycle aging. Calendar aging is what you're talking about in your comment - it's basically determined by the temperature and state of charge. Both of these impact the rate of chemical processes that lead to loss of lithium and active material. This is what happens when the battery is sitting on the shelf.

Cycle aging, on the other hand, happens when you charge or discharge the battery and is driven mostly by the actual volume change in the anode and cathode when they get lithiated (it's quite significant, up to 15% or so in some materials). This introduces mechanical stress, which can break the protective film that forms between the electrodes and electrolyte and allow chemical degradation to proceed at a faster rate. The degree of mechanical stress is mostly determined by the depth of discharge, although rate of charge and discharge is also believed to be important. Cycle degradation is also impacted by the temperature at which the battery is cycled, in a similar way to calendar aging.

With high temperature variance (low temperatures will wreck your batteries too, and I can't recommend keeping them at 5C), calendar aging is the dominant mode. This is absolutely the case with cell phone batteries. However, electric vehicles, especially higher end ones like the Tesla, have an active cooling system that keeps the battery at a constant temperature, even when the car is "off". This is why many EVB warranties will be voided if you let them run out of charge for more than 14 days - at that point, the cooling system isn't working and your poor battery is at the mercy of the elements.

Under closely controlled temperatures, cycle aging is the most important factor, and it's dominated by depth of discharge. If you drain the battery as far as it will allow you every day, it's going to be hosed. If you only go down to about 80% of the allowed charge every other day, it'll last for quite a while. These numbers are roughly correlated, obviously, with the mileage on the car, but the relationship is not simple, and I'd caution against considering mileage to be a good determinant of battery degradation.

Some good sources if you want to learn more:

A fairly simple model of degradation (ignores charge/discharge rate): https://www.researchgate.net/publication/303890624_Modeling_...

A thesis on the subject, goes into great detail and has an excellent bibliography: http://webfiles.portal.chalmers.se/et/Lic/JensGroot.pdf


Battery science is fascinating. Here's in an-depth look at degredation: https://www.youtube.com/watch?v=9qi03QawZEk

The speaker (Jeff Dahn -- he now works with Tesla) pioneered a faster way to tell how fast a battery will degrade (which means you can iterate your design faster!)


What could explain the graphs and cycle stability is that Tesla does not deplete the battery pack below a certain percentage of the real battery capacity or over a threshold... basically they keep the battery pack well beyond the critical ranges where batteries degrade from simply idling, and they "undersell" the pack capacity.


Tesla undermarks their sticker capacity by ~3% usually, which is just to keep you from totally destroying the battery by overdriving it. For normal wear and tear they have a charge limit, which sets a soft maximum charge. 80% is standard IIRC, but it can be set up to 100% (which is just slightly below the actual capacity).


Yup and that's totally fine in my book since they're selling a rated range.

My single datapoint at 60k is I've seen only 2% degradation which is phenomenal.


I've always wondered how they do it.

They use the same 18650 cells that power laptops, yet laptop batteries degrade badly over time. My 1 year old XPS13 reports a battery "wear level" of 29%.

I wonder if Tesla's battery technology could be used in laptops and phones?


I wonder if they're particularly good at balancing the cells. When people recycle 18650s from laptops, I think it's often because one or two cells went bad while the rest remain fairly healthy.

Similarly, if you put things like batteries or capacitors in series, you typically want circuitry to balance their voltages, because a pair of 2.7V elements charged to 5.3V could really be at say, 2.75V and 2.55V, which would be unhealthy for the former cell. I think those imbalances are also caused by small differences in the individual elements' properties (capacitors are typically less precisely-tuned than resistors) which will drift further over time with wear and temperature changes and overvoltage conditions and whatnot. So that 2.75V cell would degrade very rapidly compared to the 2.55V one, probably causing even larger imbalances until it eventually fails.

Most lithium cells have basic protection circuits built-in against over/under voltage and overcurrent (e.g. shorts,) because they tend to explode if you leave those things out, but they definitely last longer if you pay attention to them as individual cells rather than a system.


A Tesla focused Youtuber recently did a video [1] where he accessed some of the data on the CAN bus, and it showed excellent balancing between the cells. The highest voltage cell reported 3.87V, with the lowest voltage cell reporting 3.864V. So the highest voltage cell has only a 0.1% higher voltage than the lowest. That's in the middle of a discharge cycle, so this may not be representative of the real capacity difference of each cell, but they're being discharging in a very balanced manner. I don't work with batteries often, so take my analysis with a grain of salt, but it seems to me that (the lack of) battery degradation is one of Tesla's biggest strengths when it comes to their battery technology.

1 - https://youtu.be/F3tF0i98MpI


Presumably they have a lot of cells in parallel? This would help with balancing, since individual cell differences would average out over all the cells that are parallel.


I wonder if Tesla's battery technology could be used in laptops and phones?

Sure, if you want a laptop which you can only recharge once per week.

Laptops and phones go through charge cycles much faster than cars; I'm sure Tesla has tuned their designs to maximize performance in that usage environment.


I think the idea is if you programmed your laptop to never charge the battery past 80% or discharge it past 20%, it would degrade much more slowly, but you would also get about 40% less battery life out of a single "full" charge. Or you would need to increase the battery mass by 66% to get the same battery life. Laptops need to be light enough for a human to carry, while cars don't, and cars are generally expected to last longer than laptops, so the second solution is probably more viable for cars.


My Thinkpad supports limiting the max charge to 80%


You can replicate it to a certain extent by not discharging below 10-20% (don't know the exact range though). Limiting upper bound is likely either impossible or requires reprogramming the battery control unit.

What is also an advantage: Teslas can cool (and I believe also heat) the batteries so they don't enter temperature ranges where degradation is accelerated. In a laptop and certainly a phone there is neither space nor weight available for a thermal management.


Pretty sure all newer thinkpads can be configured to only charge to a threshold


It's far less likely real technology and far more likely massively undersold capacity in a Tesla.

Laptop and phone batteries are physically close to heat sources + are used frequently to substantial degrees of discharge + frequency of charge cycles after substantial discharge.


I thought Tesla switched away from the laptop cells some time ago?


I think most/all EV's do this.


They do it to different degrees.


The Prius did this - fanatically keep it's NiMH battery at between 20 and 80% charge.

I'm pretty sure cell phones do this too.


A NiMH would have to, wouldn't it?


Average life of ICE vehicles can't be 266kkm, that's absurdly low. Most of people I know have bought vehicles with that mileage and used it for another 100kkm after that at some point in their life (think first/second car, second family car, etc.). It's important to know that especially in Central/Eastern Europe, you can never trust the car's odometer - even the cars I'm talking about had probably more than 350kkm when bought.


I have seen plenty of cars not make to 165_000 miles, those that do are often near the end of their life. Mine is beat and barely holding together and it is at 97_000 miles.


My 80s diesel Benz is barely broken in at 249k miles. It all depends on the car.

Ironically, more modern cars will probably not last as long! I can rebuild just about anything on this car, electronics included (most complicated part is perhaps a common opamp). On modern cars, as soon as the advanced ICs (say an ECU) are NLA you’re SOL.


Nah, older cars which weren't made well, or maintained well are all sitting in a landfill right now.

It's only the few that were both made better, with no production flaws, and great maintenance that have lasted long enough for us to see then with 150k miles before being purchased today.


165 000 miles is 266 000 km, for the readers who might wonder.

My car is at 350 000 km now, and I'm not planning on selling it anytime soon. It doesn't have any special problem, except for a malfunctioning electric window lifter on one side. It's only one data point of course.


What percentage of car owners do this?

There is some percentage of cars crashed in the first week. It only takes a few of these to cancel out the average of extreme outliers like your car.

For reference I have never seen an odometer with more than 200 000 on it before. Even most classic rebuilds I have seen like 69 Corvettes have some high 100K count on them, some event have reset odometers.


I don't know, but if the average is 220 000 km (according to the article) I'd say a fair share of cars must go beyond that.

If I had not bought my car (it was at 300 000 km) I'm pretty sure it would have been sold in Eastern Europe instead, or maybe Africa since it's a Peugeot (a 406).

In the countryside, it's also very common to have both a nice recent car, and an old reliable one like mine for the dirty work. Most people I know do have one car that is a few 100 000s of km. Also, we mostly use diesel here, and these are supposed to be much more reliable than the gasoline engines used in the US (and I'm not going into the purported lack of reliability of American cars to begin with). I'm talking about everyday cars that are 15 or 20 years old, not classic cars.


Gasoline engines can be plenty reliable. I think the main reason why a lot of cars from the 80s and 90s are able to go to 400k and more is (1) cheaper and easier maintenance, so more inclined to do it regularly (2) no fancy shit going on, no turbo chargers, no high-pressure fuel injection, thus far less wear on parts, no unreliable high-pressure pumps etc.

Replacement parts on these are also far more expensive. New turbo is a four-figure number just for the part. Injection pumps are bloody expensive, too.

Less optimized designs may also be a factor. CAD and FEM modelling has only become better with time, so back in the day engineers would rather make a more conservative design, nowadays that's not so necessary any more.


Aluminum heads are part of why some newer cars don't last as long. It's easier to blow a head gasket. Replacing those in a tightly packed front wheel drive engine bay often exceeds the value of the car. Or, driving with the blown gasket causes overheating that damages other components.


I'm up to 235,000 miles on a 1985 Ford Bronco as the original owner, but that's a 2-ton truck. High-mileage trucks aren't that unusual.


We had a (1988?) family Peugeot 505 station wagon that made over 600,000kms back in the day.

Perhaps they don't build them like they used to, but our 1980's Honda sedan also easily clicked over 350,000kms too.


Huh, I drive a Toyota with 205,000 miles on it. Still in great condition after tons of cross-country road trips.


My 1972 VW Beetle has done 6000 miles apparently.

Turns out it has gone round the clock...


Poke your head in any work van or truck and you'll see 200k and up.


Or any taxi. My trip back from the airport last week had 440k on a Prius.


> except for a malfunctioning electric window lifter on one side.

There's probably a Youtube video that'll walk you through how to replace it. I replaced the one on the driver's side door of our 2001 Honda Odyssey last year. I think it was like $120-150 (USD) for the part and 3ish hours. I know almost nothing about fixing cars (just generally how to use basic tools), but have used info from forums and (especially) youtube to fix several things like that on our cars.

[EDIT] great thing about the Youtube videos is you can get a good idea of whether it's something you can manage yourself before you tear your car apart.


People making YouTube videos to teach strangers how to do things is one of my favorite things about the human species :')


97 000 miles, just barely 160k kilometers. That's... really not much at all.


I had a 98 Jeep Sport Cherokee at 170k miles. One time I ran into a guy with the same car at a Walgreens (same model, same color, same year), we said hi and laughed about it. His had recently hit 300k and was still going strong. Cars can last quite a long time with regular maintenance.


I had a friend in high school whose family's Jeep just hit a million miles and was still chugging. Well, that's what Jeeps are for.


That was very true of cars in the US before 1990, say. Today's are much better.

I have a 2004 Corvette with about 140,000 miles. Stuff breaks, but it is a vette. My last F150 had about 160,000 miles when I, uh, wrecked it. Still looked and drove like new.


My automotive longevity anecdata:

'99 Toyota 4Runner, 232K miles (currently own); '85 Honda Civic, 197K miles; '59 VW bus, ~490K miles (with a rebuild of the 40hp engine at ~240K miles and a transaxle from a junker at ~300K miles).


Uhh... My own car has around 98k miles (165kkm) and I consider it pretty new.


Yup, when I donated my car at 145k mi I expected to to be resold but got notified that it was turned into parts for $350(they had to report back to me for tax reasons).


Buy a subaru next time.


> Mine is beat and barely holding together and it is at 97_000 miles.

You are ridiculously hard on your capital equipment. It must be nice not to have a care in the world for your expensive possessions. Personally I couldn't bring myself to torture fine machinery like that.


Lots of assumptions in this comment.


As a kid being driven around in carpools, and being one of the smaller kids in the neighborhood, I ended up in the middle-front seat a lot (Back when cars had bench seats in the front). I saw exactly one car with over 160 000 km (100k miles) and it was a manual transmission Volvo.

I don't know why my friends parents had lower mileage cars, but I know what it was with my mom:

At this point most cars were automatic, and they crammed lots of parts of the transmission into a hard-to-service single-part-number item. Three times while I was growing up (~32 000 km per year driving) she had the transmission fail at around 170 000km and the mechanic quoted her a number approximately equal to the street-value of the car to fix.


>Average life of ICE vehicles can't be 266kkm, that's absurdly low.

It appears to be the consumer reports data.

[1]: http://www.nbcnews.com/id/12040753/ns/business-consumer_news...


Add one or two zeroes at the end and i'd guess we'll be more in the ballpark..


what...? You're saying cars can drive 1.5 to 15 million miles on average? They barely ever make a million miles[1].

[1]: https://www.autotrader.com/car-news/ive-been-using-a-950000-...


Well, after googling "ICE car" i can see my mistake :)

Being german, i immediately associated ICE with the train of Deutsche Bahn[1] which probably do around that mileage...

[1] https://en.wikipedia.org/wiki/Intercity-Express


> Average life of ICE vehicles can't be 266kkm, that's absurdly low.

People buy a lot of Chryslers, Fiats, Fords, and Chevys. You'd be lucky (or cursed, depending on how much you hate the car) to make it to 266kkm with a Chrysler Sebring, for instance.


Sebrings, as I recall, require either completely removing the front left tire, or pulling the wheels to the left as far as they go in order to remove the battery, which is housed under the wheel well.

Not everyone builds Corollas.


That's interesting because I know a lot of people with European Fords (Mondeo, Focus), myself and my family included, that went well over 350k km. Are the American Fords so incredibly bad?


I think it's more that people who never maintain their vehicles don't expect them to last past 100k miles. If you follow the manufacturers suggested maintenance regime in the manual it will last much longer.

Unfortunately, this means the used car market is also full of moderate mileage vehicles that require a ton of preventative maintenance as soon as you buy them, giving the used cars a bad name.

If you buy a used Chrysler with 100k miles, its probably 40k miles overdue for a timing belt replacement, transmission filter change, radiator flush, alignment, brakes, tires, etc... On the plus side, you probably got it pretty cheap.

For one point of reference, Ford was notorious for "value engineering" the old US Rangers so they would die around 150k miles, but I kept mine running for 250k miles and 23 years with mostly just regular maintenance and the occasional repair. The thing that finally killed it was an aftermarket slave cylinder for the clutch that failed and blasted brake fluid all over the inside of the transmission.


Not expecting a ~$30k vehicle to run after 100k miles seems absolutely crazy to me. It's a nice showing of how different (wealthier and not caring in this case) the USA are from Europe.


It will be interesting to see what the useful lifetime of electric vehicles will be. In theory, they should last a lot longer because they have fewer mechanical parts. So, maybe the older ones will keep running fine but upper-middle-class people won't want to drive them because they look worn and shabby. Which is sort of what happens with gas cars now, but to a greater degree. Maybe 30, 40, and 50 year old cars will become common, assuming the frames don't rust out and someone still makes replacement parts (including batteries, when those eventually wear out).

And then there's the software... I don't have a lot of confidence in any company to supply software updates for multiple decades. I hope we don't end up in a scenario where wealthy people have secure vehicles and the less well-off just shrug and accept malware on their car as the price of living in modern times.


Or maybe they will go to the junkyard despite being perfectly fine because no one manufactures batteries for that model anymore (and installing a third-party battery is blocked by some DRM).


Sounds like a case of survivors bias. Cars totaled in accidents likely bring down the average.


And Fiats.


Then there is the average being pushed down by cars scrapped because of accidents or massive failures such as big engine failures with low mileage.

Also: Here we have a very strict road legality checkup (joints, breaks, rust, lights, belts, ..) for all cars. Once a bad/cheap car reaches the age when it starts failing the checkup, it's often not worth fixing. So while it could probably do another 100kkm with the dodgy wheel bearing, it's not allowed to without immediate expensive repair. So these checkups effectively purge every car that would need repairs to any vital parts costing more than its worth every year.

I have a 10 year old peugeot worth around $1-2k that has gone 18kkm. If it were to fail a checkup for even something simple like a wheel bearing or uneven handbrake appliance - I'd probably hesitate to repair it because I could just as well get another one (A recently checked one for $1k would limp along at least 1-2 years to the next checkup while mine is banned from the road immediately).


Remember, the average American has one car accident every 17 years.

Car accidents are random in nature and significantly reduce average car's lifespan. Further, many cars age without being driven very far per year and 15+ year old cars are often junked even if they are relatively low mileage if they are part of even a fender bender.

You also have issues with rust independently from mileage.


> Average life of ICE vehicles can't be 266kkm, that's absurdly low.

I'm not sure it is...

I mean, in my country (Netherlands) the average car user drives 13k km per year, in very car heavy countries that's probably around 20k. But here it'd mean a car would last on average 20 years, that's not absurdly low by any standard.


OK, maybe I misunderstood. To me it seems like it's comparation of incomparable though - you can't fault the vehicle that you didn't use it enough to broke it and had its life ended just because of age. With proper care, the vehicle could go much farther than 500k km with no noticeable performance or range degradations - just like the heavily used Teslas.


Pretty close. In the US, it is commonly assumed that average mileage per year is 15,000mi (24,000km).

Maybe someone could write a quick script to scrape AutoTrader.com or a similar site and see what the mileage is on cars over 20 years old that are offered. ;-)


Not sure about the US, but in Europe, you can never trust the odometer. Displayed mileage is often a third or a quarter of the real mileage.


I've heard from mechanics that ICE cars are designed to last 250kkm reliably.

It doesn't mean that car can't last longer than that, but more and more repairs are to be expected, and it is more likely to fail at any time. If you are a professional, keeping such a car doesn't make economical sense.

Furthermore, in high income countries, they actually want you to destroy these old cars to promote the new car market.


I have family who regularly drive old bombs. (Do they call them that elsewhere?) They apparently spend half their life at the mechanic. (Slight exaggeration.) Yeah, I'm not doing that.


It depends a lot on whether the ICE is diesel or gasoline. They have very different lifetimes.


It is more about when the car is costing you enough per year to fix that paying for a new one would be cheaper overall.

Also perceived safety/ convenience, a car might be good for another 100k km but appear unreliable enough that it might break down at any time.


My sister spent $2000 on a Daewoo sedan like 6 years back, it needed constant repairs and didn’t have working climate control - the thing only had like 80K miles on it. Total waste of money, but she kept fixing it up and wasted thousands of dollars before eventually junking it.

Meanwhile, the 2006 Prius I bought two years ago just hit 110K miles (purchased with 80K miles), cost $10K, and has needed nothing more than new tires and a replacement cooling block on the engine (which was deferred by the previous owner). I plan on running this thing until they no longer can service it, a full swap of the hybrid transaxle is relatively cheap as is replacing the engine, but with regular maintenance they’ll probably last 500K or more and I’ll only have to replace the battery at some point.

Always do research on the reliability and cost of repairs on a car before you buy it, this alone was why I picked the Prius over other cars on the lot (there’s very little that can need serious repairs because of how simple the ICE design is, and the electric drivetrain components are pretty long lived as well).


kkm, is that kilo-kilometer? Could we also say megameter? I like that word.


With the side bonus that you can pronounce it "muh-gah-muhteh"[1] to really confuse people.

[1] I've never got the hang of IPA, sorry


ICE = internal combustion engine


I've never seen anyone put more than one kilo before the unit before.


For me, it's normal to write "its mileage is 165k" - and the "km" is assumed. I just wanted our friends that use imperial units to understand, you're true it's non-standard, and it's actually the first time I wrote it like that.


Well Americans did (sometimes still do) use micro-micro, as in µµF.


I guess it's less confusing than using "mm" for "megameters". "I drove my car 266mm" doesn't sound like much of an accomplishment.


mm is millimeters though, so that would be about a quarter of a meter. I guess Mm would be what you are looking for.


The capital letter doesn't add much to the sound of it.


You lose about 5% capacity in the first 45k miles, then it's essentially flat- 3% in 120k miles. It seems like keeping your charge level at 100% has some impact on lifetime, but lower than 90% looks to have very limited returns. I'll be interested to see how the cycle graph evolves, but right now it looks like batteries will on average make it to 2500-3000 cycles, which is great.

There are a number of early outliers that had more significant capacity loss. They're all from the US- Europe & Asia have over 800 reports compared to 170 in the US, but the lowest reading there is 88% compared to 85% in the US. All with ~<35k miles, <4 years old, mostly 60 kWh models. I'm not quite ready to call fake, but the numbers from the US are definitely unlike other countries.

Those cars weren't driven nearly as hard as their counterparts in other countries. The only real explanation other than false reporting is heat, but unfortunately there is no indication of location besides one guy in Michigan. Keep that in mind when looking at all the graphs- the lowest data points are almost exclusively very strange outliers from the US that weren't driven particularly hard.


IIRC no high voltage car batteries are actually charged/discharged 0-100%, even if it reads so from the user's perspective. This is due to significantly increased wear on the batteries when fully charged/discharged, and why a cellphone battery only lasts a few years.

The batteries need to be significantly over-provisioned to extend their lifespan. A Prius, for example, will keep the battery's SOC between 40-80%, and is one of the most reliable hybrids out there.

The same is true for EVs, so charging it till it reads 100% does not mean the battery is at capacity. The same is true when it reads empty.


I don't think any lithium ion battery is ever truly used at 0-100%. True 100% would be just before it explodes, and true 0% would be just before it completely bricks. You get horrible degradation near those areas so they're always avoided. I'm sure consumer electronics push it farther than cars do, though.


The "sticker" capacity on a Tesla is usually ~3% lower than the real capacity (the one in EPA filings), but the car lets you chose any charge level up to the sticker rating. That's the one I'm talking about- I just mean as relevant to an owner.


FYI: Tesla batteries are literally laptop batteries with cell balancing. No secret battery magic going on.

You could in theory crack open the pack and replace inefficient cells to get your capacity back. But, I'm assuming this is too huge a PITA for most people.


I thought laptop batteries are also a bunch of phone batteries stuck together. Is that assumption incorrect?

Edit: apparently so.... I would appreciate a reply if anyone has time. I have a difficult time figuring out what about my question could have caused disagreement.


Not an expert and I didn't downvote but from my examination of laptop batteries, they appear to be composed of LIon cells of a standard size, something like 3V cells. Phone batteries OTOH seem to be pretty custom designs to fit a specific phone or range of phones. For e.g. https://electronics.stackexchange.com/questions/156928/repla...

Edit: added link.


Phone batteries are more commonly lithium polymer, and laptop batteries are more commonly lithium ion (using 18650 cells).


> No secret battery magic going on.

They have a complex cooling system. Your laptop battery definitely doesn't have that. If it did, it wouldn't need to be replaced every few years.


Most people need to replace their technology about as often as their batteries, so there is no incentive to cool them.

But the reasons they need to be replaced often is more about 1. Not keeping them at/above 40% charge, and 2. Laptop batteries don't balance the cells. Most laptop batteries can get 80% capacity back just by replacing one or two crap cells.


The "technology" involved here is #1 understating the raw capacity of the battery and #2 managing the charge/discharge levels and rates to maintain that capacity. This gives a good user experience and perception of longevity. Apple does the same thing with the iPhone and it makes the iPhone user experience noticeably better than comparable Android handsets where the charge controller just charges the battery as high and as fast as it can.

Essentially, this is good UX, but for battery charge indicators.


> Apple does the same thing with the iPhone and it makes the iPhone user experience noticeably better than comparable Android handsets where the charge controller just charges the battery as high and as fast as it can.

iPhone battery life and lifespan has been a much-hyped thing in certain circles for a while, so back when I got my first iPhone 5S, imagine my dismay to discover it was actually not noticeably different than my old HTC. Same with the 6.


What exact models are we comparing? HTC's current phones weigh more than the iPhone 6 and much more than the iPhone 5s. Note also that since Android follows the Windows/PC model, there exists hardware that does this correctly, and there exists hardware that does it incorrectly.


Sharing a form factor does not mean those two things are the same. The only similarity between laptop batteries and tesla batteries is the can they come in, which is... not important.

Please don't open a Tesla battery, the bus is at 600 volts and will kill you stone dead, then start a fire


The cells inside a Tesla S are off-the-shelf Panasonic catalog parts and are identical to parts found in some laptops (and flashlights, and so forth). Tesla's contributions are entirely in the controlling electronics, integration, cooling, etc.


That hasn't been true for a very long time. Tesla uses their own chemistry[1] and has put in extensive work improving it. One of the higher-profile changes was that they use 1-2% silicon (Musk mentioned this in a stockholder call), which panasonic still hasn't rolled out to the rest of their batteries.

[1]: https://electrek.co/2017/05/04/tesla-battery-researcher-chem...


I think people get confused because it's widely (and correctly) reported that Tesla uses 18650 cells. The confusion is that people think that's a model number, but it's actually just a size and shape. It's literally 18x65mm and the 0 means it's round. It tells you nothing at all about what's inside the cylinder.


NCR18650B is a specific series of parts from Panasonic(/Sanyo). If you want literally the exact battery that Tesla uses in the S, call Panasonic and tell them you want a shipment of ten million NCR18650BF (the new one with the silicon anode. The identity of these parts has been confirmed multiple times by people tearing down the battery packs (where they get this kind of disposable income, we don't know) and comparing the size/weight/discharge characteristics of single cells.

The "advanced chemistry" that Tesla moved to after a year of S production was a SiO-doped anode, exactly the difference between the B and BF part numbers from Panasonic-Sanyo. The cells weigh a gram less. This makes the charge density appear higher when the denominator is in terms of mass.

Tesla has done a wonderful job of selling their brand and convincing people that their batteries are other than off-the-shelf technology. In reality the same chemistry is available to anyone who calls up Panasonic and orders a hundred million batteries.


Do you have links? All I'm turning up with search terms like "NCR18650BF tesla teardown" are people using that as the closest point of comparison or speculating.


I don't think those have ever been used in the model S.


Tesla has used a customized chemistry since just after the launch of the model S. Tesla literally has their name on the talk from that link. Do you think they're just doing research for fun?


It's a PITA and also kind of dangerous.


Elon would personally come to your house and hit you with a stick.


Presenting the Telsa graph with the full x axis (to demonstrate the rate of degradation is very small), and then the Leaf with a reduced x axis (to suggest it's got serious problems) doesn't really lend confidence.


There is one thing about batteries: Watt-hours battery will provide highly depend on current (electric current) draw. To some extent this also applies to deterioration. Which means that as batteries get more and more used, one should expect to see higher and higher disparity between range in sporty and economic driving. In other words as battery gets older, the higher consumption driving aggressively.

Tesla, however, has one redeeming factor. While battery voltage-capacity@current curve will change over time, Tesla can collect real world usage data and with software updates push updated curves for older batteries to adjust range estimation models for battery wear.


If it were the same quality as my laptop battery, you would have to keep it plugged in while driving after 1 year.


Correct me if I'm wrong, but Lithium Ion batteries have exponential degradation, right?

The data here assumes constant degradation, which is going to throw this off.


What all these dots above 100% mean, besides moving the average up? And I'd be worried about the owners of all these dots below the line


>Want to buy a Tesla? Use my referral link!

Man, this only needs to work once for it to easily outstrip all other forms of revenue a blog might generate.


The current referral program offers only non-monetary awards (toy car, fancy wheels, entry in contest to win car). Better than a poke in the eye, but not as good as cash.


There's not much point in Tesla offering a lot. A search for "Tesla" in Google gets you to the right place to buy one with a couple of clicks.


Is there a non-linear decline at the end? i.e. do the batteries at some point just give up and capacity starts declining very quickly?


In the comments it is explained that this is because there is only one user who made it that far so the graph is skewed towards that users data. That users battery happened to be slightly worse than the mean so the graph goes down. It doesn't really mean anything.


No, loss slows down over time and continues to drop approximately linearly. The batteries develop a protective effect as they degrade, which slows further loss.


So you can conclude your car willl be worth nothing after 200k miles


I'm just over here patiently awaiting some startup to unseat Tesla as the king of speedy, luxury, electric automotive. Nobody is quite taking the cake, or even trying for that matter.


There are at least a dozen companies trying to do exactly that. Thing is, Tesla was exceptionally lucky, and it is still incredibly hard to build a car company. Tesla managed to squeak through to the Model S, which was absolutely critical.

For companies that are trying to make a Tesla killer, they have to hit the price point above a Model S with zero volume. That's means no factory and hand building almost everything. That destroys any ability to price reasonably. Tesla killers start at 300k and range up to 2 million. Nobody at that price point wants electric cars right now, or at least very few people. It doesn't come with the same bragging rights as a Ferrari or Koenigsegg. Those have brand and absolute, vastly supreme performance respectively.

Also, I suspect people are aware that as soon as Tesla announces a new roadster the game will be over. Nobody will want to buy anything else.




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

Search: