Stalled Out on Tesla’s Electric Highway
nytimes.com
nytimes.com
This article is _exactly_ why we're not getting one for the fleet anytime soon, and it's basically the answer I've been giving people since the Roadster came out (and the questions started).
Tesla is an awesome, awesome company. I'm a fan. I'm thinking of getting a Model S for myself. I think (and hope) they're going to be huge.
But they're not ready for the truly mass-market quite yet - and car rental customer (even luxury car rental customers) are the mass market.
I could see the exact scenario outlined in the article happening to renters constantly - except for one difference: We would be yelled at by the customers, not Tesla ;) ("Why is my car saying 'battery dead'? I did nothing wrong. Send a truck to pick me up or I'm writing a bad review about you guys.") Sigh.
Anyway, I'm still a fan despite the headaches - and I know they'll grow out of it.
Luxury car rental company? Please tell me more (like do you have a website??)
I ask because I know of a couple of luxury car rental companies, but there is one ore more things about each of them that make them exceedingly unpleasant to deal with. I would be interested to try doing business with a luxury car rental company that is run by people who also read HN - seems to be a good signal for not-totally-fucked-up business practices.
The car works great.
The problem is that there are very few places to charge it.
If there were more than two cities in the whole area to charge, he could have stopped sooner at one of those.
A relative gauge or percentage would be one thing, but if you tell the driver it's going to get 100 miles, they need to know it can make that to the next charging station.
-- why not engineer a stable micro-climate for the battery?
I wonder if Tesla chose to optimize for other scenarios in the design of the car: Tesla distributes the battery across the entire body, which makes for great handling, fault tolerance, and safety, but may make it more difficult to insulate than in other cars, which typically just lump the battery in the back, surrounded by insulation.
An "inside" battery also takes advantage of the solar greenhouse effect from the glass-enclosed interior. On cold-but-sunny day, a parked car will heat up substantially simply due to the ~1kw/m^2 of solar energy delivered to the vehicle. Of course, you don't want the battery getting to hot, either. In addition to better thermal insulation, perhaps Tesla could add solar panels to the roof to aid in heating and cooling.
I hope this problem is solved, as it would be a bummer not to be able to take a Tesla on a ski trip.
I remember back when I used to have a travel-intensive job that I was flying into airports all over the United States and getting into unfamiliar models of cars rented on the corporate account. That was sometimes no fun at all. But all of those cars fit everyone's familiar mental model of displaying fuel remaining. The Tesla Model S, by contrast, reports miles still able to drive in a dynamic model that can rapidly get out of touch with reality. And of course it takes a good bit longer to fully recharge a Tesla, even with high-speed recharging, than to pump gasoline into any other kind of car. This will take some major getting used to for most drivers.
If EVs become common, people will have home charging stations. Demand for non-at-home charging stations will come from (1) folks who drive a lot and need to charge intraday (and who cannot charge at work) and (2) road trippers. The sum of both of these categories is fairly small and less demand means less supply. So if you do fit into one of those two categories, you will need to plan your driving carefully instead of figuring that you'll be able to pick up some gas when you need it. This can be overcome by arranging for the proliferation of public EV charging stations even absent the demand. Perhaps the manufacturer or municipalities will subsidize it.
I don't care how many charging stations there are - adding multiple hours of charging time to a short trip like that is just not going to work.
Although there's one other conceivable option for a technical solution - if you could add charging circuitry to the freeway and recharge while driving, that would be a game-changer.
I agree with you that this is the only realistic mainstream future of electric vehicles as long as batteries remain in their current low-density state.
Also a while a digital camera does get to spend some time outdoors it spends most of it's life in fairly people comfortable environment.
And that sucks.
However, I'm annoyed that Tesla's UX people let a car with this much design intelligence leave without connecting range estimations to a simple thermometer.
I understand that to HN readers batteries in the cold is "simple physics" but if the battery indicator jumps around like a BitTorrent download time estimation, people will quickly learn to not trust what the car tells them.
Consumers that are highly neurotic about the road trip problem are much more likely to remember these sorts of horror stories when making a purchasing decision.
Would you trust your laptop or cell phone's battery life estimate to that extent?
The question I'd have is, how much current would they have to divert to keep the batteries warm? It seems like the juice expended to hold the battery pack at 20C might be less than the capacity lost by letting the batteries go down to an ambient -5C.
Or, they could provide what a lot of east-coast cars have, a built in engine-block heater that you plug in to a 110V extension cord overnight. Except in this case it would be a battery pack warmer, but the same idea, keep the important bits toasty over a cold night.
Showing that the range is zero when the battery is dead is not hard. The problem is when the displayed range drops faster than you're traveling, and if that wasn't the problem, he wouldn't have been stranded. If it was adjusting its display for temperature, it wouldn't have been wrong.
Or, they could provide what a lot of east-coast cars have, a built in engine-block heater that you plug in to a 110V extension cord overnight.
Had he plugged the car into a 110V outlet overnight, the battery would have been charged in the morning.
Full disclosure, I own a Zero motorcycle. The thing is amazing in SF city traffic. It's less amazing on the highway. You should read up on Terry Hershner. He's taken a Zero motorcycle cross country by driving for 45 minutes and charging for 45 minutes. So, it can be done, but road trips are not the optimal use case for an electric vehicle. yet. :)
A Tesla S with the 260 mile battery, I would treat like a car with 130 miles range. Plenty for an around town commuter and errand vehicle. A Leaf, I would treat like a 50 mile range car.
Also, why don't those things come with a propane heater? Electric heat makes no sense.
You can do electric heat via heat pumps, but it requires more equipment and it doesn't work if it's too cold. You can get about a 4 to 5 advantage over plain resistance heat.
Gasoline engines lose a lot of energy to waste heat -- but sometimes that waste heat is exactly what you want.
They're a disaster from a range point of view.
Battery capacity is, though, just like it is something you think about when driving a petro fuel car. Drive a diesel? You need to think about gelling fuel in Vermont in January. For a supposedly well educated technological society, an awful lot of us seem to engage in magical thinking.
Would he have been able to get where he wanted to go, or would he have not tried to go as far as the car said he could go?
Yeah, either the author was kinda ignorant about how batteries react to weather, or he was trying to achieve a bit of sensationalism.
So, I've been reading the forums at teslamotorsclub.com recently, and there are a lot of questions (and answers) about the Model S's range. From what has been written over there, it sounds like the car's speed has the biggest impact on range. In the NYT piece, I notice that the only time he mentions his speed is after it started to look like he might not have enough range to make it to his destination--and then he only mentioned how slow he had to drive. I would submit, then, that an "experienced early adopter" would know that his driving speed matters and therefore not try to drive down the expressway "upside-down at mach 3 with his hair on fire."
I also notice that the author made no attempt to charge the car at his destination in Connecticut. Instead, he planed to rely exclusively upon Tesla's "Superchargers" for the entire trip. Again, an "experience early adopter" would know that the Model S can be plugged into a clothes dryer outlet and fully charged overnight (i.e., no waiting). He could have started his second day with a fully-charged battery, if he chose to. Actually, the author probably would have had enough range for the rest of his trip if he had even plugged into a standard 120V domestic outlet in Groton.
Frankly, it sounds like the author was trying to make his life difficult. Almost a third of the verbiage is devoted to range problems that were a direct result of his decision not to charge in Groton. When drivers choose not to fill up before driving across Death Valley, do we blame the driver or the car?
This isn't to say that the car is ready for prime-time. After all, the author may be representative of the general public. However, I think that anyone with even a modicum of curiosity about his new car--and more importantly, someone who, by purchasing one, had some sort of incentive to make things work instead of writing a diatribe in the NYT--would have no trouble figuring out these things, and avoiding the range problems described in the article.
I also just checked plugshare.com: there are at least half a dozen public J1772 charging stations (which are relatively high power) near the route segment where the author plausibly had range anxiety. Why he chose to charge at Butch's Lunchonette when there was a high power station nearby is baffling. What is doubly baffling is that he launched on a 209 road-mile trip with 185 miles of estimated range, and then made no attempt to recharge the car until he was nearly stuck (if you believe the numbers on the little map inset). So, I'm afraid that I can't escape the conclusion that this is a problem of the author's making.
That is just not the modus operandi with an electric vehicle. You plug it in everywhere you can, and frankly thats almost everywhere theres main power.
Frankly, I dont get all this discussion about mileage. It seems like a perfectly good car for driving in the city already, which is where most of fuel is wasted.
nb: I am a huge Tesla fan.
Fixed above. Also, my impression from the article is that Tesla is telling people to derate the mileage estimate by more like 10%, not 25% or 50%.
Early adopters in any age will have a rough time, that's kinda the definition of early adopter. However, in a hundred years people are going to be as familiar with this technology as we are with liquid fuel. They'll probably look back at us and chuckle about how primitive we were with our "battery" technology. :)
Underestimate the remaining miles. Even if that makes you look bad up front. Its an eerie form of cognitive dissonance when your reserves are dropping faster than you consumption and the wheels in your mind spin wondering "will I have enough". You never forget this feeling. Even if its just a test drive, you'll associate Tesla cars with that feeling forever after.
I suspect they already do, hence the comment that he'll still have some range even when it says zero.
The problem is what number to display after a full charge. If it doesn't strongly resemble the specification range then people won't be happy.
Because the world understands completely how to store and handle and deal with liquid fossil fuels.
However, in an ideal world, I would think that carbon neutral hydrocarbons would become a niche fuel for things like long haul trucks and bulldozers. (And for auxiliary space heaters in people's electric cars.)
I know people who have done family road trips with the Tesla S already and didn't have those problems in normal weather. One could surmise that pushing the edges of the range AND pushing extreme weather conditions might be risky, and in an EV it is certainly riskier than in a gas car, but like the Top Gear "driving a roadster on a race track drains the battery horrendously", it somewhat reads like an attempt to intentionally push an EV to the breaking limit, and then write a negative article about it. The political angle of trying to make Tesla look like a Steve Chu/DoE failure increases my suspicions.
Could you imagine a reporter writing an article about a gas car that he intentionally drove at the limit of its mileage, got stuck on an interstate, and then turned around and roped in government loans to General Motors?
The Tesla S deserves criticism in order to shake out bugs and improve the software to reduce customer surprise, to better predict battery life, no doubt about that. But this should not be used by the anti-EV crusaders who have an axe to grind to score points against ex-Obama administration officials.
Why nobody seems to understand how stupid charging an EV in real time is, is something that I find astonishing.
(To clarify: I can understand why most people don't understand why non-swappable batteries are stupid, but I can't understand why someone like Elon Musk doesn't understand that. I'll admit that it makes me wonder if I'm missing something.)
I'm probably just scratching the surface with a minute of thought. I'm sure that the folks at Telsa have been over this ground, and much more.
Swapping my car's battery should be as big a deal as swapping the propane cylinder on my gas grill.
And yes, batteries these days are perfectly capable of monitoring themselves in a leased-usage scenario. Your laptop battery has its own CPU and EEPROM, for instance. To understand why, see any recent news story on the Dreamliner.
Yes, and we've just read about them in the New York Times.
The tradeoffs due to price point, available technologies, time to market, etc leave a lot of "should" on the cutting floor due to compromise. I don't think we're particularly close on hot swap batteries as a stand in for real time charging while they still weigh 1000lbs.
I'm betting that folks like Telsa will drive hard on the necessary advances in battery technology over time, or energy storage technique more broadly.
Personally, I think having to swap the battery is just as stone age inefficient. I want some sort of super-capacitor with instant charge capability and fine-grained control over the energy release --- not to mention avoiding the nasty chemical mix in traditional battery tech, waste energy in the form of heat, etc.
Maybe some amalgamation of materials science, nano-tech, etc will get us there.
Yes, that part of the article struck me as a real head-scratcher. This kind of "engineering" is becoming all too common in the auto industry, even among companies that have been around for decades and should know better.
For instance, the battery in the new-generation Porsche 911 is in the front trunk... which can only be opened electrically.
WTF, people? Are all the smart guys and gals working for Google these days, or something?
Of course, without an industry-wide effort to standardize a few core battery sizes for different applications, it's a moot point. That is the problem that I wish Musk had set out to solve.
Tesla's battery packs are distributed throughout the car, both to allow for cargo space (not having a trunk would be something of a minus) and to distribute the weight of all those batteries. I am having a very hard time envisioning a way to achieve both of those goals while making the batteries replaceable.
It's just that for the majority of people, battery swapping makes no sense. They can charge their vehicle over night at very very cheap prices and have enough range to serve them all day. If you insist on taking a cross-country roadtrip, rent an ICE - the infrastructure is just not there yet (or peoples minds have not accomodated - see this article).
Pic of the battery: (which is really the base of the whole car) http://www.cartype.com/pics/10143/full/tesla_model_s_battery...
To some extent there is a standards problem: there is not a common battery pack (like there is a gas tank fill point). But even if there were, the batteries would continue to run down after delivery (unless supplied with a non-trivial amount of power). And the space required to physically store these batteries would occupy a lot of volume. After all, gasoline/diesel is far more energy-dense than modern batteries.
My thinking is that they would be kept on trickle charge at the gas station after delivery. That wouldn't take a huge amount of power. The actual recharging process would happen at a depot that has high power connectivity to the grid with low distribution losses.
I suspect it would be more efficient to do that and have the charged batteries trucked to the service stations, than it would be to have each service station do their own primary recharging. Especially since, as you say, the service station owners aren't going to want to devote a huge amount of space to battery inventory.
I'm vaguely aware of Better Place but they certainly don't get Tesla's mindshare (good, bad, or otherwise.)
Storing the batteries at the station will require much more volume than the analogous amount of gas/diesel. And a dumb tank beats a smart battery charger with software and EEPROMs if your concern is resiliency. Some EV problems are pretty clearly chicken/egg adoption problems. But there are some real infrastructure problems, too, and other changes may bear far more fruit towards the goal of more efficient energy exploitation.
Maybe there will be swappable batteries eventually, or one part of the battery will be swappable, but as far as I understand thats not coming soon.
What's a bit disturbing about this article isn't the laws of physics, but that he was being coached directly from Tesla. If Tesla HQ can't coach a reporter 1:1 out of a devastating result, what hope does the average early adopter have? Either the reporter twisted the facts, or the car has horrible temperature problems, or the Tesla coaches seriously screwed up.
I hope it's not a real issue. A hotel manager might be able to let one Tesla plug in overnight, but there isn't infrastructure to let ten or twenty Tesla owners do the same--yet. However, just as wifi became a competitive advantage to offer at hotels, so will EV charging stations. Perhaps even pay parking lots will begin to offer them.
It's an interesting and broadly relevant article, but it feels like we're having the wrong discussion - we might be single males, but the target market for the time being is well off families considering a new second car... ;)
NYTimes article about Tesla range in cold is fake. Vehicle logs tell true story that he didn't actually charge to max & took a long detour.
source: https://twitter.com/elonmusk/status/301049593385340928Cold weather and running something that is power intensive (the heater) will drastically effect your mileage if you are relying on a battery to power your car.
Tesla probably should do a better job of communicating this but it doesn't seem like a flaw with the car.
Can the flaw be corrected? Maybe, maybe not. But it's still a flaw.
I bet he probably could have gauged it by jumping in the car turning it on and then starting the heater as well. His mileage would have declined before he even started driving.
Think of it this way, batteries always improve in 2nd / 3rd gen devices (look at apple) and I am sure that is what Tesla will start to address.
No, its not perfect. Neither is the choke on your motorcycle or the people in siberia waking up every 2 hours to start their combustion engines because blowing up what was generated in thousands of years refuses to happen in a cold engine block.
But I guess everyone with a modicum of knowledge has by now realized that a) we can't keep blowing up that stuff and b) there won't be another magic energy material like oil, ever again.
Damn gasoline engines - so unreliable! You'd think after 100+ years of development they'd have sorted this stuff out.
Batteries suck in the cold. Did no one think to test the Model S in the east's freezing temperatures before they turned one over to a journalist?
What gives, Elon?
The reason for the big losses in the cold is the battery management system that is trying to keep the pack warm and has to expend energy to do so. You can opt not to do any battery management, but then you get thermal runaway and a life-threatening Boeing situation, or (in the cold) your battery pack is damaged.
They picked the best option there is under the circumstances.
- Have the reporting system look at current temperature and change the reported range.
- More supercharger stations closer together. (200 miles apart for a reported range of 268?! Ballsy. Especially in the extreme cold of the northeast. Reckless when you are giving a car to a NYTimes reviewer.)
- Don't have the parking brake LOCK so you can't even move the car without a flatbed truck.
- Have the car warn/inform the driver when they park somewhere cold to keep it plugged in to warm the battery up
and that's off the top of my head.
Many things can be done. This isn't necessarily a 'physics problem.' It's communicating and setting expectations. When the car says "100 miles" and you are stuck at the side of the road after 50 miles, it doesn't matter that "the physics make sense". What matters is you were screwed by completely false information.
The fix is simple. In fact, it's so simple, I would be SHOCKED if the TSLA guys aren't already including it in a future firmware update.
Obviously, lithium batteries are going to lose charge in colder weather - that's just physics, but the problem here was twofold: one, the inherent limitation in the battery itself, but perhaps more importantly, how the car reports available charge to the driver in those conditions. If the car tells me that I have 100 miles to drive, I should have 100 miles left to drive. That it does not is a design flaw, and one that should not have been overlooked, plain and simple.
Even with batteries going down sooner than you expect, you could always recharge, if there was a charger in the same _city_.