This article makes it seem as if the Toyota president came out to Elon Musks California home, and after a ride in the Roadster, suddenly saw the light and decided to start producing electric cars.
And just how did he decide to execute his grand electrified vision? By backporting electric propulsion onto an existing SUV platform. Yeah, this is exactly the moment where this particular piece of journalistic spin comes crashing down.
The RAV4 EV was sold only in California precisely because it was a compliance car. Toyota simply decided it was cheaper to leverage their existing investment into Tesla to produce an EV on the cheap that could get them the required environmental credits so they could continue to sell their actual vehicles without having to buy the credits at a (at the time) large cost from other manufacturers. It was a simple localized business decision, not some grand strategy.
Of course Musk doesn't just ridicule fuel cells, he affectionately calls them "fool cells". As someone who builds his businesses by reasoning from first principles, he justifiably doesn't understand the idea of swapping a 20% efficient (if you're very lucky) combustion engine with a 10% efficient, experimental fuel cell when you could instead use a dead simple, clean and 80%+ efficient electric motor.
It's not the motor that Musk is so invested in, but the mature technology of the 80%+ efficiency lithium-ion battery. A fuel cell powered car will use the same motors as any other electric car.
From a non-technical perspective, the battery capacity is what sets Tesla apart from every other EV available today. Musk seems to understand that quite well.
I think you're missing the point about the motor. The BMW i3 uses an electric motor for propulsion, but that doesn't mean it's suddenly 80% efficient if the electricity is coming from the internal backup combustion engine. If anything, you are now only 0.2 * 0.8 = 16% efficient [1].
What I'm interested in is the efficiency of the whole process in the car that turns the fuel into motion. In the Tesla, the electricity is stored in batteries which have few losses and goes directly into a highly efficient electric motor, so it gets that 80% overall.
1: In practice, the setup electric engine for propulsion + combustion engine for electricity tends to be somewhat more efficient than just the combustion engine for propulsion, as you can run the combustion engine at a constant, efficient range of RPM.
I think you're focused on the motor, when that's the least important part of the equation. As you point out with the BMW i3, the source of the electricity is important. In battery mode, it is actually more efficient (27 kW-hrs/100 mi) than the Model S (35 kW-hrs/100 mi), but that's obviously not true when the range extender kicks in.
Sure, the motor is more efficient, but energy density of gasoline is around 50-100 times that of lithium-ion battery packs[1]. Sure, you can get around that in electric cars because of the more efficient motor and simplified car design. You can remove a bunch of conventional car parts and replace the weight with batteries. Still, I don't think you can dismiss outright research into fuel cells, which promise to be much more energy dense than batteries.
“A little bit of slope makes up for a lot of y-intercept”
So, sure, right now, the energy density of gasoline makes terribly inefficient combustion engines practical. But their energy density isn't going up (while lithium batteries are improving and coming down in price), certainly gasoline isn't going to get cheaper, and as for the combustion engine, well thats pretty played out after 100 years of trying to improve it.
But there is probably a hard ceiling to the efficiency of safe and cost effective Li-ion batteries. We can't be entirely sure because we don't fully understand how Li-ion batteries work, but my best layman's educated guess is that those are close to tapped out as well.
The strength of the battery+motor electric drivetrain is that it's modular. Come up with another energy storage method, build it into a power pack, then plug it right into existing Tesla S cars, get the over the air software update, and you're back in business. There's also all of that frunk space.
Obviously batteries will not surpass 100% efficiency and are already reasonably close to that so yes, batteries are nearing a hard limit on efficiency. Energy density on the other hand is no where near its theoretical limits.
> Obviously batteries will not surpass 100% efficiency and are already reasonably close to that so yes, batteries are nearing a hard limit on efficiency. Energy density on the other hand
Energy density is what I meant. I found some graphs to look at. You are probably right about there being lots more room for improving energy density.
Only that we've been trying for awhile now. I just looked at some graphs online, and maybe they'll get 50 to 100% times better in the next decade or so. That's pretty significant.
I don't disagree (at least w.r.t electric cars), but to play devil's advocate: a factor of 50-100 is A LOT. We're talking about two orders of magnitude here, which would require a fundamental breakthrough in battery technology to achieve parity with gasoline. Even if we make batteries twice as efficient, we're still 25-50 worse than gasoline. And breakthroughs in battery technology are hard to come by. So if Toyota wants to hitch their wagon to fuel-cell technology instead, all the power to them! Batteries are not necessarily the future of portable energy.
But of course you're not just worrying about fuel/battery energy density difference. You should look at energy deliverable to the road per pound of fuel + engine + exhaust system versus electric motors + batteries, which I would guess is nowhere near a factor of 50-100.
If you now consider that the only way to use that energy is through combustion, and that an combustion engine is only 1/4th as efficient as an electric one, that factor suddenly shrinks to 12.5 - 25 already.
Frankly, energy density is a bit of an distraction. If we go by energy density alone, we would presumably all be cruising on Uranium, which energy density is... well off the charts.
But there a plenty of good reasons we don't fuel transport with Uranium, and there are just as many good reasons why we shouldn't continue to use gasoline.
It's a dumb saying. Let's say Bill Gates stopped making any more money tomorrow. How long would it take for you to catch up to his wealth?
Lithium-Ion battery energy density has about doubled in the last 25 years: http://futurist.typepad.com/my_weblog/images/2008/03/11/batt.... Even if gas technology stops improving tomorrow, it'll take 100-200 years to catch up on the energy density front.
Battery propulsion doesn't need to get anywhere near the energy density of fossil fuels in order to have equal utility, due to the huge difference in efficiency (both theoretical and practical). Ignoring utility and just thinking pure kWh for a moment, do you have a source for your 100-200 year claim, assuming maximum Carnot efficiency for fossil and 100% efficiency in electric motor conversion?
From my armchair perspective, the Tesla Model S would be equally useful to any gasoline car for all but a handful of edge use cases, with only the minor change of adding a metal-air battery as a range extender for excessively long trips. And that's today, not 100 years from now.
What existing infrastructure can hydrogen use? Sure, there are natural gas pipelines in place, just like there are electric transmission lines in place; what other infrastructure can hydrogen take advantage of? Why would that infrastructure need less upgrading than the electric infrastructure? (Serious non-rhetorical questions to which I don't know the answers)
The grid and power generation plus the fact that most us homes are 110V. and are limited in how much power they can charge.
Also My local small petrol station can fuel say 60 cars an hour have you any idea what sort of HVAC plant is needed to store and deliver that sort of power.
You can convert IC engines to burn hydrogen and they can just slot right into existing models.
You can also use intermittent wind and solar power to create Hydrogen for use later.
Most US homes have 220-240 volt service coming in that is split at the breaker panel.
Most A/C and heat, dryer, range are 240V.
An electric car can charge at night when load is low, so the power infrastructure that is engineered for high demand due to cooling in the middle of the day in summer will work just fine if everyone is charging at night. In other words, the grid is already mostly robust enough to handle this load.
If you're going to use electricity to make your hydrogen, you are certainly no better off than using it to charge a battery. Also, the energy density of hydrogen isn't exactly awesome either, and it's hard to store.
> As someone who builds his businesses by reasoning from first principles, he justifiably doesn't understand the idea of swapping a 20% efficient (if you're very lucky) combustion engine with a 10% efficient, experimental fuel cell when you could instead use a dead simple, clean and 80%+ efficient electric motor.
You're comparing apples and oranges. That electricity isn't generated via a particularly efficient mechanism after all.
Sure, I mean this is where it gets complicated and subjective.
If we consider a solar cell with top of the line 15% efficiency and use that to charge our electric car, of course the overall efficiency takes a sharp drop. But I don't think it is useful to include the efficiency of the solar cell in the calculation here. After all, the 85% that the solar cell couldn't convert into electricity aren't useable by any other way known to man either; the loss is neither here nor there because we can't do better utilizing the sun than with current gen solar panels.
If you consider, however, that oil refineries consume significant amounts of electricity (from the grid, no electricity equivalents here) to refine oil into something our cars can use, that is certainly a factor we need to include in the efficiency calculation for gasoline powered cars. That electricity could have been used to power an electric car, after all, so there was an actual loss here.
> Of course Musk doesn't just ridicule fuel cells, he affectionately calls them "fool cells". As someone who builds his businesses by reasoning from first principles...
That's almost an ironic statement, given that his whole game-plan with Tesla and SpaceX has been aggressively commercializing extremely efficient implementations of proven, well-understood technologies.
I love Elon Musk, but there's an additional wrinkle here. Fuel cells are much further from commercialization than lithium ion, and needs a lot more basic R&D, and Tesla doesn't have the time or resources to go down that road. Toyota does. This is not to say whose approach is right, but rather that there is a lot more in play than "first principles."
It's a time-honored tradion in business: ridicule and marginalize the things that you d not do or that compete directly with you - until you so them.
Examples:
Apple: Nobody wants native apps, theywant the mobile web.
Oracle: You don't need row locks (row level locking appears in Oracle 7, as a premium add-on called the Transaction Processing Option).
I would hope Musk's position is rooted in the knowledge that we need to get off of fossil fuels in a few decades or we're done. If hydrogen as a non-thrust fuel has any use, it has so much research ahead of it; there isn't enough time for that.
This is mainly driven home by the idea that hydrogen is very much in abundance in the universe, but here on earth it really only exists as hydrocarbons. Hydrocarbons are a tricky thing to exploit without contributing to global warming.
Sorry, I should have mentioned water. But that bond is very strong, and the methods of breaking it are thus very poor.
Another compound worth mentioning is ammonia. A lot of the current hydrogen harvesting research revolves around ammonia. Unfortunately, ammonia is very scarce in nature and certainly soon exhaustible as a global fuel.
Could you explain how you got the 10% / 20% efficiency numbers you mentioned? Your combustion numbers are in the same ballpark, but you got a much different result for fuel cells.
> Toyota engineers also rejected Tesla’s proposed designs for an enclosure that would protect the bottom of the RAV4 EV’s battery pack... Tesla ultimately added a titanium plate to ... better protect its battery.
That might explain why Tesla was so willing add said plate to the Model S.
I know this is very meta, but the page is so cluttered... I would link to the Readability version but it redirects back to the article (previous complaints "you're stealing our visitors" caused that). So instead I can recommend this: https://readability.com/bookmarklets
It doesn't sound so much like a culture clash to me. Sounds more like a difference in the direction each company wants to take (EV vs hydrogen).
Just because they didn't share code (since they were both doing it, it makes it a common mindset for both and not a difference in culture) and one thought the brakes should be less jerky isn't some massive difference in culture. Any complex engineering project will have many difference in opinion you can nitpick. But I saw nothing in the article that shows these differences were what led to the breakup.
"Back in May 2010, as the emerging alliance took shape, Palo Alto, California-based Tesla’s chairman, who’s also CEO, called the partnership “historic” and said Toyota was a company he long admired."
I think a more interesting culture clash is the one between Toyota and GM. There was an episode of This American Life dedicated to the story[1], and it was eye opening (to someone not versed in the ways of car production). Interestingly, the plant that Toyota sold to Tesla (mentioned in the article) is the Nummi plant, around which the GM/Toyota partnership was centered.
45 comments
[ 34.9 ms ] story [ 1409 ms ] threadAnd just how did he decide to execute his grand electrified vision? By backporting electric propulsion onto an existing SUV platform. Yeah, this is exactly the moment where this particular piece of journalistic spin comes crashing down.
The RAV4 EV was sold only in California precisely because it was a compliance car. Toyota simply decided it was cheaper to leverage their existing investment into Tesla to produce an EV on the cheap that could get them the required environmental credits so they could continue to sell their actual vehicles without having to buy the credits at a (at the time) large cost from other manufacturers. It was a simple localized business decision, not some grand strategy.
Of course Musk doesn't just ridicule fuel cells, he affectionately calls them "fool cells". As someone who builds his businesses by reasoning from first principles, he justifiably doesn't understand the idea of swapping a 20% efficient (if you're very lucky) combustion engine with a 10% efficient, experimental fuel cell when you could instead use a dead simple, clean and 80%+ efficient electric motor.
From a non-technical perspective, the battery capacity is what sets Tesla apart from every other EV available today. Musk seems to understand that quite well.
What I'm interested in is the efficiency of the whole process in the car that turns the fuel into motion. In the Tesla, the electricity is stored in batteries which have few losses and goes directly into a highly efficient electric motor, so it gets that 80% overall.
1: In practice, the setup electric engine for propulsion + combustion engine for electricity tends to be somewhat more efficient than just the combustion engine for propulsion, as you can run the combustion engine at a constant, efficient range of RPM.
Did you read your parent post? "What I'm interested in is the efficiency of the whole process"
[1] http://en.wikipedia.org/wiki/Energy_density
“A little bit of slope makes up for a lot of y-intercept”
So, sure, right now, the energy density of gasoline makes terribly inefficient combustion engines practical. But their energy density isn't going up (while lithium batteries are improving and coming down in price), certainly gasoline isn't going to get cheaper, and as for the combustion engine, well thats pretty played out after 100 years of trying to improve it.
The strength of the battery+motor electric drivetrain is that it's modular. Come up with another energy storage method, build it into a power pack, then plug it right into existing Tesla S cars, get the over the air software update, and you're back in business. There's also all of that frunk space.
Energy density is what I meant. I found some graphs to look at. You are probably right about there being lots more room for improving energy density.
Can you elaborate on your reasoning?
Frankly, energy density is a bit of an distraction. If we go by energy density alone, we would presumably all be cruising on Uranium, which energy density is... well off the charts.
But there a plenty of good reasons we don't fuel transport with Uranium, and there are just as many good reasons why we shouldn't continue to use gasoline.
Lithium-Ion battery energy density has about doubled in the last 25 years: http://futurist.typepad.com/my_weblog/images/2008/03/11/batt.... Even if gas technology stops improving tomorrow, it'll take 100-200 years to catch up on the energy density front.
The saying also deserves an addendum: "... if you've got the time."
From my armchair perspective, the Tesla Model S would be equally useful to any gasoline car for all but a handful of edge use cases, with only the minor change of adding a metal-air battery as a range extender for excessively long trips. And that's today, not 100 years from now.
whereas upgrading both domestic and the grid to support widespread electric charging has a lot of cost and infrastructure issues.
Also My local small petrol station can fuel say 60 cars an hour have you any idea what sort of HVAC plant is needed to store and deliver that sort of power.
You can convert IC engines to burn hydrogen and they can just slot right into existing models.
You can also use intermittent wind and solar power to create Hydrogen for use later.
Most A/C and heat, dryer, range are 240V.
An electric car can charge at night when load is low, so the power infrastructure that is engineered for high demand due to cooling in the middle of the day in summer will work just fine if everyone is charging at night. In other words, the grid is already mostly robust enough to handle this load.
Also what happens when a large number of people put their cars on charge when they come home from work.
You're comparing apples and oranges. That electricity isn't generated via a particularly efficient mechanism after all.
Unless you're talking about converting natural gas or coal directly to hydrogen, which is just as asinine.
If we consider a solar cell with top of the line 15% efficiency and use that to charge our electric car, of course the overall efficiency takes a sharp drop. But I don't think it is useful to include the efficiency of the solar cell in the calculation here. After all, the 85% that the solar cell couldn't convert into electricity aren't useable by any other way known to man either; the loss is neither here nor there because we can't do better utilizing the sun than with current gen solar panels.
If you consider, however, that oil refineries consume significant amounts of electricity (from the grid, no electricity equivalents here) to refine oil into something our cars can use, that is certainly a factor we need to include in the efficiency calculation for gasoline powered cars. That electricity could have been used to power an electric car, after all, so there was an actual loss here.
That's almost an ironic statement, given that his whole game-plan with Tesla and SpaceX has been aggressively commercializing extremely efficient implementations of proven, well-understood technologies.
I love Elon Musk, but there's an additional wrinkle here. Fuel cells are much further from commercialization than lithium ion, and needs a lot more basic R&D, and Tesla doesn't have the time or resources to go down that road. Toyota does. This is not to say whose approach is right, but rather that there is a lot more in play than "first principles."
Examples: Apple: Nobody wants native apps, theywant the mobile web. Oracle: You don't need row locks (row level locking appears in Oracle 7, as a premium add-on called the Transaction Processing Option).
This is mainly driven home by the idea that hydrogen is very much in abundance in the universe, but here on earth it really only exists as hydrocarbons. Hydrocarbons are a tricky thing to exploit without contributing to global warming.
Uh, water?
Another compound worth mentioning is ammonia. A lot of the current hydrogen harvesting research revolves around ammonia. Unfortunately, ammonia is very scarce in nature and certainly soon exhaustible as a global fuel.
https://en.wikipedia.org/wiki/Fuel_cell#Theoretical_maximum_...
Comparing tank-to-wheel efficiency of the whole vehicle, fuel-cell vehicles average 36%, compared to 22% for diesel.
https://en.wikipedia.org/wiki/Fuel_cell#In_practice
Could you explain how you got the 10% / 20% efficiency numbers you mentioned? Your combustion numbers are in the same ballpark, but you got a much different result for fuel cells.
Genuinely curious, thanks!
That might explain why Tesla was so willing add said plate to the Model S.
Just because they didn't share code (since they were both doing it, it makes it a common mindset for both and not a difference in culture) and one thought the brakes should be less jerky isn't some massive difference in culture. Any complex engineering project will have many difference in opinion you can nitpick. But I saw nothing in the article that shows these differences were what led to the breakup.
This sentence, hurts, my brain.
[1] http://www.thisamericanlife.org/radio-archives/episode/403/n...