Running Out of Power in a Tesla on the Side of a Highway
theatlanticcities.com
theatlanticcities.com
And then "The estimated 265-mile range of a fully equipped Tesla Model S has allayed some concerns about having enough juice to get where you want to go."
So they left on a long trip with a 93% charge knowing that it was going to be close and fell short by 3 miles... why is this news or article worthy?
I don't own a Tesla and don't plan on owning one any time soon, but articles like this really annoy me, as they don't provide any meaningful substance. The estimated ranges for gas cars are affected in the same way by driving conditions and the driver.
The person they refer to who did a CA to VA trip seems to have a much more pragmatic approach to this and as the result has a mostly uneventful trip. His good experience has nothing to do with being a Tesla fanatic. He's just charging the car all the way and makes sure the distance he needs to go is reasonably doable.
If you've thought about owning an electric vehicle, one of the first thoughts that pops into your head is "what happens if I run out of juice in the middle of nowhere?" Including questions like "How does the car behave? Does it just shut off?" to "How does AAA handle it?" to "How hard is it to get going again?" A lot of other commenters are raising points that the author was kind of a dumbass. But as it turns out, I too fuck up every so often, so I'd like to know what happens when I do.
The article goes a long way towards answering those questions. I'm probably not going to get an electric car anytime soon but I don't see anything in the article as particularly anti-Tesla or even anti-electric.
This is relatively common, though not in issue in most large metro areas.
It could also have given more driving guidance. I didn't see how they came up with 63mph, but the car itself should know about energy-speed tradeoffs could have told them a more appropriate speed to go.
And if knowing they wouldn't reach a charger it could have had them stop closer to somewhere that could help, or direct the AAA equivalent to them for the moment they ran out.
People keep mentioning the mental effort of electric cars for these issues, but the car itself can address those and do so far more effectively.
Trying to drive anywhere cutting it this close, in any car, is likely to leave you stranded.
Behind the scenes the mileage estimates are really a central number with error bars. In this case the electronics could definitely have come up with better numbers based on altitude changes. And it definitely could have given guidance on speeds to maintain whatever margin is comfortable.
The article also wasn't just about the range; it covered several other topics as well. One of those topics is the difficulty of getting towed. It doesn't matter how many times you say "well, duh, use it according to spec", humans are humans, and they will run out of juice in their vehicles, particularly in limited-range vehicles with little refueling infrastructure. They couldn't even get the car onto the truck without a Tesla guy to power it enough to shift it into neutral, so even if they had found an earlier towing service that didn't fob them off, not being able to be put in neutral is still a problem in regards to towing.
There's plenty of newsworthy stuff in this article, it talks about one experience of when things go wrong. There's also plenty of articles talking about how great the car is - should we say that they're also not newsworthy, because there have already been glowing reviews?
The next time the writers need to make a story I can loan them my Transit Connect. It never gets within 10% of its "distance to empty" for a full tank, even in optimum conditions.
Boom! Byline.
Range estimation is a lot more complicated than 'percent of max power stored' * 265. Obviously Tesla's algorithm was too optimistic in this case, even though there seemed to be some 'safety margin' at the start.
I'm sure Tesla has a great deal of data on the variance of miles covered on a full charge over all sorts of terrain, climate, traffic, driving patterns, etc. and I would assume they use that data to build a decent range predictor. Probably the hardest part is not so much getting a good prediction from given set of inputs, but actually getting decent inputs to feed into the predictor. Driver enters the destination address, but after that... is the car pulling temperature, wind speed/direction, elevation delta, statistical traffic patterns, live traffic, etc...
OK... After considering all that, maybe we just implement 'percent of max power stored' * 265 and call it a day. (It's like the 80/20 rule has become the 80/1 rule -- so often now I feel like you can make something 80% as good, for 1% the effort.)
...WIND ADVISORY REMAINS IN EFFECT UNTIL 3 PM PDT THURSDAY...
WINDS...NORTHEAST WINDS 15 TO 25 MPH WITH GUSTS TO 40 MPH WILL INCREASE TO 20 TO 30 MPH WITH GUSTS TO 50 MPH LATE TONIGHT THROUGH WEDNESDAY. ISOLATED DAMAGING GUSTS NEAR 60 MPH WILL BE POSSIBLE IN THE FOOTHILLS. WINDS ARE EXPECTED TO BE SLIGHTLY WEAKER WEDNESDAY NIGHT INTO THURSDAY.* ==============
Accurate range prediction in this neck of the woods also needs to account for the weather.
Sustained 30-40mph gust 60 is not uncommon in so-cal/desert.
Point your EV into that for an hour on i-5 and ...
You will lose some range for sure.
Estimated range is not fool-proof. Common sense should be applied to everything - even technology packed car.
Can you toggle the regenerative effect when you lift off the pedal? Like just when your about to crest a hill, they quickly toggle it on once you get + over the speed limit on the downhill side?
I can hit some pretty astounding numbers in my TDI Beetle vert with really simple tricks but that does rely on the fact that I can pretty much stay in high gear so engine braking isn't a factor, only danger being stalling. Engine uses no fuel without throttle input unless you put the clutch in.
In either scenario, I believe both the motor and disc brakes are utilized when the brake pedal is depressed; however, the general theme is that more kinetic energy is lost to thermal from disc brake friction in the Low scenario.
My thoughts are that regeneration of the battery is encouraged, but the marketing of the Model S appears such that Tesla Motors wants to deliver EV luxury & performance first, and foremost and then cleverly extend it behind the scenes in such a way that the user need not think about it. To me, it's not all that different from flagship smartphones. Actively 'hypermiling' a flagship smartphone for battery life would compromise many of the flagship features... which aren't suprisingly luxury and performance.
You can, but this would slow you down more than gravity, the transmission, and the friction of the road alone would.
The only thing you can really do is change to neutral. Dual-clutch transmissions in high-end sports cars have already been doing this for years, though they don't do it based on the angle of the road surface, since if you're going fast enough, it's beneficial to do it as often as possible.
Exactly. And this is through the desert. This approaches the darwin award threshold.
Also, the article neglects to mention the number of chevrolet cars that caught fire during his trip.
There's a neat part of the trip on the reverse route, when you're coming down the mountains and into CA. You can basically mark (in your head) a landmark across this enormous valley in your head. And then you drive towards it, cruise control at 85, (unless you're in a Tesla!) for roughly the next hour until you reach that spot.
You can just see so far and there is absolutely nothing. Except for that little oasis travel center / truck stop in that valley, which is almost exactly the halfway point between the two cities.
I bet they'll have a supercharger station there sometime soon.
It's almost like they wanted the trip to fail to write this article.
A model S uses 237.3 watt hours per kilometer.
If you turn them completely off for an hour, you can go an extra 50 meters.
I will. Let's see…
I just ran down http://standby.lbl.gov/summary-table.html and added up typical phantom loads at my house. It comes up to about 50 watts. 36 kilowatt hours per month. I use about 1500 kilowatt hours in a month, so thats 2.4% of my power consumption.
The Tesla has a 425km range. Our 50 meter savings works out to 0.01%, but to be fair that was for one hour of savings, so lets quadruple it to cover the whole battery trip… 0.04%
Unplugging all your phantom loads is 60 times more important to your electric bill than the backlight and stereo is to the Tesla's range.
Lets think about how tiny 50 meters is a different way… The probability of turning off your backlight and stereo making a difference in running out of power in the last 50 meters of your journey is 1 in 8500, once every 2.2 million miles (if you never plan your trip to have enough range). In 2009 there were 74 crashes with injuries per hundred million miles traveled, or 1.3 million miles per crash.
So you are more likely to be injured or injure someone else than to make any difference by turning off the console and stereo, but wait! If you do turn them off, you will likely be less distracted and may spare someone an injury.
(Um, I should confess that I played more than a little loose with the "running out in the last 50 meters" thing. It would require extensive modeling. One would probably never plan a 3000 mile trip on a single charge, likewise a 200 mile trip is never a problem, and realistically after the first few dozen times they ran out of power by cutting it too close, surely a driver would learn something and change their planning. I just calculated "given you ran full distance and ran out of charge, what is the probability that 50 more meters would have helped".)
It's a legitimate issue that you may take a detour--forced or by erroneous navigation--and so you might end up a bit short for whatever reason (that's not an issue specivic to ev/tesla). And it seems unrealistic to call the driver (any driver) 'stupid' 'ignorant' 'not worth of the car' ... which are basically lazy ad-hominem attacks. The issues is well understood by peole of experience in actual long-distance navigation.
A "normal" modern car's alternator can put out ~100 amps at 14.4VDC. That's a maximum of about 1.5kW, which will charge a Tesla at a rate of about 2 miles of driving per hour of charging, again assuming a magical efficiency.
The reality is that the car's own systems take at least a few hundred watts and alternators aren't built to run at maximum load all the time and are likely to fail after a while running near capacity. Plus, the Tesla doesn't, to the best of my knowledge, contain any sort of DC->DC hardware capable of bringing 14.4V up to the necessary charge voltage, and even if it did losses would be at least on the order of tenths.
So, even with a good, running car and a theoretical alternator, you'd be waiting a very long time next to the running car (and burning a lot of gas) before your Tesla would go anywhere.
Obviously it would be possible to build, for example, a tow truck with much greater charging capacity and as time goes on we'll probably see at least a few of those in real life. But the fact remains that gasoline provides an excellent energy density in a convenient liquid form and that for certain edge cases, its conveniences will be hard to beat.
Volumne would be the size of a a carry-on wheelie bag. The engineering solution would be to have a second charging point internal to the passenger/trunk--perhaps in the 2.0 version of the Tesla models currently on the board. The only engineering consideration would be that the li-ion powertrain would need to operate during charge (like a laptop or cell phone) but would presumably overcome.
Nobody wants lithium batteries in their passenger compartment, but I do suspect it might be possible to make a ~200lb pack that would fit in a Tesla's frunk and move it at least a few miles. Thermal management would be a tremendous challenge in that area unless the "auxiliary battery" port could somehow interface safely with the car's existing liquid cooling system.
In a way the Volt is just that, only in a single package.
http://en.wikipedia.org/wiki/File:Hybridserie.png
(in Europe it is called the Opel Ampera)
The Volt weighs an entire thousand pounds less than the Tesla, and it gets a much better watts-to-mile ratio. Even then, the Volt's generator is all the way up in the 55kW range.
That's a far cry from gas or diesel, where a five gallon can can be transported by a friend in any vehicle, poured into any car in a matter of minutes, and will get you 100-200 miles.
Towing to the nearest fixed rapid charging station seems more practical than the generator vehicle in many cases, provided tow companies will tow the vehicle.
One interesting idea might be for Tesla to provide a "towing mode" that raises the ground clearance as the batteries are dying in order to make it easier to load the car onto a flatbed - that, along with education of tow companies, could be very beneficial to improving the outcome of these "stranding" situations.
Not to mention that with such chargers you can charge you car for free, while hydrogen will still cost quite a bit (probably not much less than gas).
It IS the most abundant element in the Universe. Unfortunately, just not on Earth....
"drafting behind an 18-wheeler truck was tested and results showed that traveling 100 feet (30 m) behind the truck increased overall mpg efficiency by 11%. Traveling 10 feet (3.0 m) behind the truck produced a 39% gain in efficiency"
As charging stations become more common this will be less of a problem but somehow I can't envision the future equivalent of a gas station that processes a few hundred cars per hour when doing the same thing electrically with everybody waiting for much longer times. Also, a few hundred cars concurrently recharging their batteries will require significant power infrastructure, gas stations can be put up just about anywhere where roads go (and that's not always where the power lines are), in a pinch you can power a whole gas station off a small genset.
I'm sure smarter people that me have thought this through :)
Would a gas car running out of fuel get similar coverage?
Range anxiety at the time that gas stations were few and far between would have been a comparable item.
I think if electric cars become more common and commoditised, mostly standardised things, extra battery packs will become popular.
A single 50K liter fuel drop is the equivalent of 1.75 TJ, that's 486111 KWh (one liter of gasoline is 35 MJ). If a typical car has a 60 liter gas tank that's 850 cars give or take. A busy gas station will go through that in a few hours. Say 200 cars / hour, that's a bit over 4 hours. So the electrical equivalent of that 50K liter fuel drop delivered over the space of 4 hours (assuming equal efficiency, which probably is not correct) would require 125 MW of electrical input. As a check: One Tesla super charger is about 100 KW, running 200 of them in parallel would require 20 MW.
So it appears that even at roughly 5 times the efficiency of a gasoline engine the power requirements of a filling station capable with such volumes would be quite significant. You're talking high voltage distribution to the stations at a minimum to reduce the losses, even at 10KV that's still 2K amps.
Pretty sleepy here so apologies ahead of time if I messed up too much on the calculations.
Surely engines that combine multiple energy sources are more expensive and less efficient than engines that use a single source? My understanding of the current hybrids is that they exist because one energy source (petrol) can be quickly and easily replaced, whereas the other (electric battery) takes time to replace and isn't as easily available. Best I can tell, that makes hybrids a temporary solution to the lack of widely available better energy sources. Once most cars have settled on a non-petrol energy source, I can't see why hybrids would continue to exist.
I'm used to seeing them in almost every mall / bigger parking these days so it looks unlikely they're missing completely. Basically I'm trying to decide if this article is simply anti-tesla (they seem to really concentrate on superchargers only), or is it describing poor availability of charging points in general.
The article has a good discussion of the trade-offs between having a lot of charging stations and not having so many at the beginning of the electric car era. This paragraph especially caught my eye:
"Tesla won't offer any details about how much it costs to build and operate these Supercharger stations, but according to internal documents obtained by TechCrunch, they're an expensive effort, marketing or no. Each Supercharger station is estimated to cost between $100,000 and $175,000, and Tesla is picking up the entire tab — from installation to maintenance to the cost of providing the large amount of energy needed to charge their cars so quickly. Nicholas says Tesla is internalizing this cost and adding it into the price of the Model S, which can range from about $70,000 to more than $100,000."
It is unbelievable that this is expected to be newsworthy.