This just shows me that Tesla is able to execute. The Gigafactory, opening up patents, rapidly growing charger network, and the eventual $35k Gen III tells me they have tons of potential in the future.
Tesla is thinking big. They want everyone in the world to eventually have a electric car. I think they will own a significant share of premium market after they release the Gen III.
However, they will be cars that people will desire at a cheaper price point than the Gen III, from different manufacturers. Also some people love certain car brands regardless of anything else, and mostly will stick to those brands. It will be great that those other car brands will be able to more easily make electric vehicles.
Those electric cars from other manufacturers will most likely have a Tesla battery and charge on Tesla charging station. I think this is true disruption.
There is still no truly legally enforceable promise, that explains in detail, exactly what Elon is giving away and what he isn't, and what the terms are.
It's now 1.5 months later, and despite many people asking such a thing, nothing has been forthcoming at all.
So before you give them amazing amounts of credit for this, could we please wait till they actually do something more than PR?
Do you know for sure that they haven't been negotiating the promised agreements in secret? I'd be very surprised if behemoth automakers managed to finish an agreement in less than two months, when it can take years in other cases. I'd also be surprised if the automakers wanted their names thrown around by Tesla before anything was official.
Okay, but when does the factory first come online? The projections are for by 2020, so where do any possible cost savings for the III come from? Unless we can expect it to slip a few more years like the X did.
I am hoping for a 200 mile range mid size, but I doubt Tesla will be first. I am also expecting range extender vehicles to have much more popularity initially because to be honest, I don't have time for 30 minute charge cycles when I am on trip and most people I know just want to get to their destination.
I think pushing the mainstream launches back would be a wise decision. They've just begun their hard knock education in designing for longevity and reliability with the earliest examples of the Model S passing the 2 year mark.
Traditional auto manufacturers have decades of experience driving the continual refinement of their designs. These efforts have resulted a general expectation among consumers of unfailing reliability and service life measured in decades. The issues they're experiencing with drive units are troubling but acceptable for most owners who know they bought a bleeding-edge testbed for future technology. My concern is one of time, things move relatively slow in the world of automotive engineering and these issues show Tesla has a lot to learn and a lot to change once they've discovered the points of failure. That kind of time doesn't fit with their expected release schedule and owners won't accept the kinds of failure rates experienced in the Model S in a mass market car.
I have been driving the 80 mile "real range" Leaf for three years and it gets me to work and back without recharging. Because it charges when parked, I experience zero charge time.
If you are single without family (so only one car) and travel with car a lot (skiing in state or whatever) then the current electrics aren't for you. But you are in a minority part of society (although probably over-represented on H N :-)
> Okay, but when does the factory first come online? The projections are for by 2020, so where do any possible cost savings for the III come from? Unless we can expect it to slip a few more years like the X did.
Tesla has already broken ground in several states for the Gigafactory (with plans to discard the sites they choose not to move forward with). I think 18-24 months is a reasonable timeline for them to be online.
I feel this comment is getting downvoted, not because it is wrong, but because it is crude. I actually hope Tesla "Fs them in the A", because the entire industry has sat on its laurels for pretty much 20 years. Instead of innovating, they have been, essentially, polishing a turd.
Are you comparing a naturally aspirated engine with a turbo-charged engine? Also, I don't think it's fair to compare truck/utility engines (which generally run on regular gas) with the engines that go into BMW's highest performance cars.
And, the 2.7 liter V6 which is going into the new F-150 supposedly makes 325hp.
Why are most of the lithium cells cylindrical, while most of our devices are cuboid? (I think of phones, laptops, external battries... but phones and most laptops have flat lithium cells nowadays). Is cylindrical less expensive? Is it more secure? It seems like a waste of space to me.
It's an efficiency matter. The cylinder gives good electrode surface relative to the amount of matter contained. For a cube that ratio would be much worse, for a sphere it would be optimal, but spheres are harder to manufacture and connect to, as well as harder to stack in arrays. So we take a medium between the sphere and the cube and end up with a sweet spot in the cylindrical shape.
Indeed it does not, this is about Tesla's and Panasonics 'gigafactory' which will make Lithium-Ion batteries in a cylindrical form factor.
Lithium Polymer batteries (or LiPo for short) are a special case of Lithium-Ion batteries. They are sold in a 'pouch' format (and they are not the first to use a format like that, for instance, in the polaroid cartridges there used to be a battery in a similar form factor (polapulse)).
The reason why LiPo batteries are not so constrained in shape as their siblings is that the electrolyte is a liquid. Solid electrolyte batteries are typically much more constrained in terms of shapes.
This is also why regular car batteries (lead-acid) can be block shaped, the acid is also a liquid.
The polapulse was an interesting battery, it only had to fire the camera 10 times but was quite high current. The packaging constraints of the cartridge dictated the battery geometry and so a flat pouch like battery was born.
LiPo batteries exist in all kinds of interesting shapes and sizes too.
Also, and this is entirely speculation and I'd love to hear a battery engineer confirm this, but: if you're packaging together cylindrical batteries, it inherently gives you more air channels between the batteries to promote cooling. When you're talking tens of kilowatts, both in discharge and charge rates, the percentage that ends up as waste heat suddenly becomes a whole lot of heat. You probably wouldn't want a maximally stuffed cavity with these packs turning into a single, massive chunk of thermal mass.
EDIT: looks like I was firsted somewhere below, doh!
This is exactly backwards. Traditional Li-Ion batteries have a liquid electrolyte, and Li-Poly have a solid polymer electrolyte. Both batteries consist of anode/electrolyte/cathode rolled up. In the case of a "pouch" cell, they are rolled up flat, and a cylinderical cell has them rolled up in a cylinder. There is actually no limitation for the shape - you can totally make a Li-ion battery with liquid electrolyte in a prism - many cell phone batteries do this.
Li-ion, due to the liquid, needs a hard metal case, whereas li-poly can often get away with a foil pouch. The surface area per volume of a cylinder is less, as well as easier to make structurally sturdy, which means less weight is spent in the casing. Tesla cares more about weight than volume, so this is an acceptable tradeoff.
Interesting! I just spent the better part of today reading up on all this because I found it hard to accept that I'd remember something so terribly wrong. Turns out that the names have changed over time to denote various packages and gel/liquid/solid combinations. Wikipedia has a whole page on the subject and it almost seems as if they made the naming of these battery classes confusing on purpose:
Enjoy the read, I certainly did and it turns out the situation is a lot more complex (also historically) than I remembered it, but my memory wasn't quite as faulty as you made me feel initially (fortunately!).
Because it would be more expensive and the space gain would be very small.
Battery geometry is mostly dictated by physics and economics, even though a flat plate battery would be denser the gains in density would be off-set by the increase in cost and in the end the wasted space doubles as a cooling channel so you might end up having to separate your flat plates by periodic air channels anyway.
The reason why flat plates are better in some applications (for instance, for starter batteries in cars) is because those batteries are optimized for very high current during very short bursts of time. 300 Amps @ 12V is 3.6KW out of a very compact little box. The power generated from that little package is extremely impressive (if you pull the distributor cap on your car (if you have an old car) you can actually move the car for a bit on the starter engine in first gear or reverse!) but because of the short duration the power absorbed by the battery is relatively small (current is absorbed by the internal resistance of the battery during charge and discharge leading to a warming up of the battery).
But it does not last for very long so heat removal is not usually a problem, and the charge currents are reasonably low, so again no big problem in removing the heat (low current, longer time).
With electric vehicles the situation is reversed, the discharge time is relatively long (hours) but the charge times are getting ever shorter which means packs have to be actively cooled to avoid overheating, and the easiest way to do that is to flow air through gaps in the cells.
I think they use standard 18650 batteries (might be wrong) and there have been discussions from Tesla saying that the round batteries also have the benefit of allowing an air gap to insulate and help mitigate overheating problems.
Plus they are mass produced, cheap, and have a relatively high energy density.
The new factory might work on creating some new battery technologies though, not sure. Ideally yes, you want the battery as compact as possible and lots of round batteries aren't the most efficient for that, but when you factor in cost and the natural heat insulation, it might have been the best option at the time. If it works, roll with it and work on one of the numerous other problems they have to solve.
The Panasonic 18650B cell has the highest energy to weight ratio of any commercially available lithium-ion cell. This is by far the biggest factor. It's also the cell used by many solar car teams for the same reason.
This is very significant for Panasonic's new direction. They now partner with a startup that takes on a IPR policy radically different from the Panasonic of the past. Panasonic's R&D engineers used to be measured internally by the amount of patents filed and the company had a large presence in every significant patent pool.
In addition it means that one of Panasonic's key manufacturing expertise will be located overseas, which is also a radical departure to the mindset of keeping core manufacturing as 'black boxes' in Japan. However this may still apply to components that are integrated in the Gigafactory.
> which is also a radical departure to the mindset of keeping core manufacturing as 'black boxes' in Japan.
I saw some of this up close recently and it was quite interesting how Japan was still 'fighting the last war' in this sense.
They were trying to avoid a repetition of Japanese companies doing an end-run around American and European companies by doing their manufacturing and then turning around to innovate on the processes and principles in the end resulting in a wholly new industrial base.
But the world has long moved on from that time and even though there is plenty of innovation in this sphere the threat of a manufacturer turning into a competitor is still present (for instance, in China) but between the US, the EU and Japan that threat is all but non-existent.
Japanese companies are slowly coming around on this and are becoming more free with respect to sharing technology and IP with other partners. This Panasonic move is another step in that direction.
Has there been any analysis on the amount of joule energy pumped out of all of the current gas stations on a daily basis?
I'm worried the current electric grid is woefully insufficient for a world full of electric cars. I know their are plenty of electric lines out there running at greater than 100% of their rated capacity already.
- The US used 134.51 billion gallons of gas in 2013.[1]
- There are ~33 kWh/gal in gasoline [2]
134.51e9 gal x 33 kWh / gal = 4.44e12 kWh (4.44 trillion kWh / year)
- The US used a total of 3.856 trillion kWh of electricity in 2011 [3]
Ignoring improved efficiency and excluding other factors, converting all gasoline usage to electric would more than double US electricity consumption.
Under reasonable assumptions, PV solar generates about 0.75 kWh/m^2 per day. [4]
4.44 trillion kWh/year / 0.75 kWh/m^2 per day x 365 days/year / 1e6 m^2 per km^2 = 16,219 km^2
Which means you could generate all of the required electricity with about 16,200 km^2 of solar panels - which is just about exactly the total land area of Hawaii [5], or 405 million average home PV systems.
Ignoring improved efficiency and excluding other factors, converting all gasoline usage to electric would more than double US electricity consumption.
Not a bad analysis, but there's a lot we're glossing over here.
For one thing, that 33kWh can get you about 100 miles in an average electric vehicle. That's compared to the average new US car which gets 25MPG, and existing cars out there average worse than that. So right away, there's a factor of four that we're leaving out here.
Further (as others have mentioned) it takes a huge amount of electricity to refine oil into gasoline. The electricity savings from not refining a gallon of gas are enough to drive an electric vehicle pretty far on its own.
Also, our grid and generating capacity are built to handle peak load, which (in most parts of the US) will occur on hot days when lots of people are using AC. On the flip side, the majority of electric car charging happens at night, so there's a lot of spare capacity already there.
If everyone switched to EVs, we'd probably need to build out more generating capacity, but probably not as much as your analysis suggests.
"Further (as others have mentioned) it takes a huge amount of electricity to refine oil into gasoline. The electricity savings from not refining a gallon of gas are enough to drive an electric vehicle pretty far on its own."
This isn't as much as you would think. The manufacturing electrical usage of the US (which includes this) is separated out on the EIA site, and you can simply subtract it :)
One fact everyone seems to forget is that it takes approximately 6.5 Kwh of energy to refine one gal of gas. That is JUST refinement and not the energy to pump, store, etc.
My Volt drives around 22 -> 28 miles on that power.
Under those conditions, converting all vehicles to electric would increase net electricity usage by ~20%. (Better than 100%!)
Eliminating gasoline refining would reduce the needed electricity by ~850 billion kWh / year, and the improved efficiency (ignoring charge/discharge cycle efficiency) would reduce the net energy to ~594 billion kWh / year - or 54 million average home PV installations. Quick googling shows that it will be about 10% higher for charging losses.
Breaking even eventually. I doubt they're going to suddenly just stop refining all of that oil. There will have to be a trend before the oil companies start lowering their output.
A given car will either use refined gas or electricity. The point is those both use a similar amount of electricity, either way the impact on the grid is similar. Oil companies are not going to refine more gas than is used. They might refine more than is used on a particular day or even month, but they are not going to outpace demand on any long time scale.
Keep in mind we'll still be refining oil / petroleum for other uses (lubricants, ink, plastic, fertilizer, diesel, etc). If this is independent of the refining into gas, then all's well with your math.
Do cars typically run on plastic and fertilizer where you're from? If it's true that a Volt can drive 25ish miles on the electricity used to refine a gallon of gas, then substituting electric miles for gas ones is a wash in terms of net electricity usage. Or are you trying to say something more subtle that I'm just missing?
If this is independent of the refining into gas, then all's well with your math.
I think we agree ... that if there's less gas being refined, then there's less energy being used to do that ... if and only if gas refining is independent of refining into all the other products. If, on the other hand, refining into the other products efficiently ends up making gas anyway, then that's different.
>Ignoring improved efficiency and excluding other factors,
But these factors are absolutely enormous, particularly for the comparison you set up, as electric vehicles are about 3x as efficient. EVs on the mass market get about 3m/kWh, whereas most gas vehicles get about 1m/kWh, if you use your 33kWh/gal.
So instead of doubling US electricity consumption, 100% EV vehicle fleet just raises consumption by 33%, an amount that could nearly be made up for with simple efficiency gains in electricity use.
That's one of the down sides to napkin math. I posted a follow up on another comment. Including some of the efficiencies leads to a net increase of about 20% in electricity usage.
If I read this right this is assuming that all current gasoline usage is converted into electrical usage. It's not like that would happen over night though, or possibly ever. Also, can PV help at night when a majority of charging should occur?
You are correct - it assumes total conversion, but the numbers don't change if you project some ramp for the transition.
A more reasonable assumption for charging is that it will take place wherever electric vehicles are parked, regardless of time of day; in that case, large parking lot installations for PV may be more sensible than individual home systems. In either case, it seems likely that increased adoption of electric vehicles will change what we consider "peak" electricity use hours, since it could consume half or more of total electricity in the US.
My assumption is that most of the charging for electric vehicles (aside from stopping at a supercharger) will go on at night which is off peak for our grid now. At least for the short term future. I don't know that solar will help in that case. The adoption of electric vehicles seems to be slow enough that it should give us time to identify the problems and attempt to come up with solutions. That rate could significantly increase if the model 3 delivers on it's promises. It probably still take major investments in our electrical infrastructure to support a majority of vehicles becoming electric.
Is this where a Solar City type of setup would come into play? Tesla's charging stations generating their own electricity, thank you very much, partnered with SC. Or would that be feasible? Of course, Tesla could also use vertical windmills [1], geothermal, etc at its stations. I'm a big fan of distributed generation; that makes the electric grid's shortcomings a moot point.
What technology generates the current has nothing to do with whether or the GPs concerns are valid. It's all about transportation, not just generation.
Do the math. It will not work. Sunlight in full sun with the sun at 90 degrees relative to the panels, in winter (cool panels) will give you about 200 watts / square meter of surface of electricity at the exit of the inverter.
You'll need a little bit less grid, definitely not a lot. Those buildings are typically multi-layer and don't have a roof over the top layer (though not always) so it's not a trivial modification either. You could design that in from day 1 of course and it will help a little bit but that's borderline making money rather than a significant reduction in infrastructure.
Say your carpark is 100 meters long and 30 meters wide, that will house 50x4x4 = 800 cars to charge but you're collecting (on a good day) no more than 100x30x200 = 600 KW. It's not nothing, but if you take into account that a single normal car charger is already 3 KW to charge a full park of such cars would require 2.4 MW. And that's the slowest rate of charge for those cars, go faster and the requirements shoot up. Best case with only 4 decks of cars you're looking at a 25% reduction. Worst case not more than 10%. Good but not great.
The grid is definitely going to see some structural modifications before we can push the equivalent of our current gasoline consumption through it.
I found some of the business arrangements interesting, though maybe not surprising. I think those are the real new details of this release.
First, they'll be producing 50GwH of packs, but only 35GwH of cells; so they'll still be dependent on external cell manufacturing for a sizable portion of the packs.
Second, that it sounds like Panasonic will basically own and run the cell part of the factory; as opposed to Tesla or a joint venture purchasing the equipment and making the cells themselves.
Panasonic already has one of the largest production complexes in the world for battery production; but with the huge demand from Tesla they have had a hard time keeping up with the market. I see this as a way for them to scale up to a capacity that can handle Tesla's orders while still being able to produce 18650 cells for the rest of the electronics market. Over the past few months I (and many others) have had a very hard time getting a reliable supply of the NCR18650B cells Tesla is using in their packs.
I don't see Panasonic wanting to give up any of the control it has over this sector. The agreement is probably positioned to obligate Tesla to purchase X amount of cells a year to make the new gigafactory economically feasible for Panasonic.
Again as with my previous response, I still dont get why the need for a Gigafactory. Battery Industry are an extremely automated, high volume market and with profit margin in single digit. Cutting those off aren't going to automatically made electric cars affordable.
Unless Telsa and Panasonic has jointly developed an revolutionary battery that is going to need some funding and mass produced. But any new battery will need years in time for safety testing. And the news would properly have leaked by now.
Making things in volume reduces the price. The profits on the batteries may be low (I'm not an insider so I have no idea what the profits are on these), but the profits on the cars (which Tesla makes) are reasonably high. It's like investing in infrastructure when you make something that uses that infrastructure, akin to a hosting facility investing the fibre optic lines that connect it.
Electric cars are already affordable, that's not the main sticking point (a fast way to charge them and a widely rolled out network of charging stations is).
If they manage to shave a few percentage points off the cost of production and create jobs in the process I don't see the problem.
"In cooperation with strategic battery manufacturing partners, we’re planning to build a large scale factory that will allow us to achieve economies of scale and minimize costs through innovative manufacturing, reduction of logistics waste, optimization of co-located processes and reduced overhead."
I guess that and also securing the supply chain for themselves as the rest of the car industry starts/continues to compete with them for the battery supply.
This is perhaps too cynical, but it's to Tesla's advantage to be supporting as many workers in the United States as possible. They benefit tremendously from wide array of state and federal incentives, and having jobs directly depending on the continuation of those incentives makes them politically more likely to continue.
Of coz, i am not saying it is bad to have it built. And all the advantage of local workforce you mention. I am just not getting the business and technical / tactical advantage of it.
They have a existential need for batteries with the Model 3. If they don't get the batteries they need when they need them, they go out of business. Vertically integrating the batteries helps prevent them from being screwed by problems elsewhere in their supply chain. They've had problems like this in the past. For the model S, some suppliers didn't actually believe that Tesla would actually want the scale of parts they had optioned for. When time came to deliver, the factory was ready, but the parts were just being produced half way across the world. Tesla had to pay substantial air freight costs.
Tesla still uses 18650 cells exclusively, unless I'm misinformed. They are still chemistry and form factor agnostic and only care about performance and price. This announcement is part of a plan to further reduce battery cell and battery pack cost, as well as increase production volumes by a factor of 10-50 over the next 10 years. I don't think there is any information on how these new batteries will look yet.
I am curious about the choice of lithium-ion vs. something like lithium-iron-phosphate (LFP) battery chemistry.
Obviously, there is a good reason for Tesla and Panasonic not to have chosen LFP, does anyone know why? I know the energy density is lower, but are there other reasons?
Lithium iron phosphate has much lower energy density (energy per weight). This is the most important factor in the choice of battery, along with the cost per energy.
LFP's advantage is in power density, which is not an issue with the size of pack that Tesla is using.
In terms of investing (buying Tesla stock), is Tesla's probable-seeming big success with their next two cars already built in to the price? In other words, how overvalued is Tesla stock? Would it be worth making it a significant fraction of my portfolio?
Out of curiosity, with the recent release of their patents, is Tesla going to be enabling other car manufacturers to purchase these batteries? Standardising on a housing module across manufacturers seems a 'good thing' for the general consumer.
I don't see why they wouldn't, they already sell their powertrains to other companies.
It would probably depend on how much extra capacity they have. The Tesla vehicles are in high demand and only going to be in more demand when they start rolling out the Tesla 3 (which is really the 4th Tesla isn't it, I guess they're referring to it being the 3rd generation with the Model S and X being the same generation).
Elon gave a funny explanation for the new name: he would have wanted a Model E to compose S E X, but the name was already taken. So he chose 3 to at least have S 3 X.
I don't know why this is being downvoted. Elon Musk himself confirmed that S, E, and X was supposed to make fun of how sex is used in marketing to sell.
They plan to have capacity of 35-50 GWh at 2020 [0]. IMO they will not be able to produce so many cars, to use all the battery capacity in Tesla cars they would to make ~0.5 Million cars per year. This means that they do plan to sell the excess capacity.
They say so in their announcement: “and for the stationary storage market”
A huge growing market will be batteries in homes. Germany is trying to push it to go from 30% of photovoltaic self-consumption up to 70% (when the sun shines, people usually are away at work). The excessive electricity is sold to the network.
This requires between 5 and 10kWh storage capacity (compared to the 50+kWh of a Tesla) for a single family home.
The current price is about 20€cts/kWh for the storage over the whole lifetime. So it is not economic yet, but it should be within a few years and I hope that Tesla will contribute to it.
Solar costs about 15€cts/kWh and electricity from the network is around 25€cts/kWh. When solar will reach 10€cts and storage 10€cts/kWh, the market will be very important.
Well I live in the UK, I've specced out a mid range Model S with a few optional extras so this comes in at £67000.
Lets assume by some miracle I manage to save up a £7000 deposit, this leaves me with some £60000 to find. At a quick google around its pretty hard to get car finance over £25000, but lets have a crack at it and try autotrader. They'll only let me borrow £50k over 60 months which gives me monthly payments of £958.
My monthly take home is £1800, so that leaves me with £852 to play with, my rent is £450 and my council tax is £101, so now we're down to £301 and I haven't paid for my utilities yet.
So an electric car in my lifetime (or at least one with a useful range like the Teslas) is a distant dream. At least until the Model 3.
That's like saying "I badly want to own a car in my lifetime", then speccing out a brand new Range Rover (think Escalade if you're american) and deciding it's impossible.
The Vauxhall Ampera is £29k, you can get a used 2012 Chevy Volt from around £17k.
I'm not sure how the limited lifetime of batteries will affect the used car economy, but the kind of people who buys a brand new Model S generally aren't keen on driving 10 year old cars.
I briefly considered hybrids, but the question was concerning an 'electric car', and I don't really feel like hybrids fit the brief.
Like I said, this might all change with the Model 3, and I genuinely hope it does. I'm tired of driving petrol dependent cars, and I want my next new-car purchase to be fully electric.
How about that, I actually thought the Ampera was a full electric, but they've thrown in a small combustion engine. They hide it well in the specs, though.
I need to do more research on the Leaf but yeah that is a close contender.
The Zoe is a problem. I love Renault cars, my Clio was my favourite thing in the world, but the Zoe has DRM in the battery which you have to rent from Renault for a fee based on your mileage, which as far as I could tell, worked out to more than I would pay in petrol.
I believe, although it may have been FUD, that if you go over the mileage you have pre-arranged with Renault, they will disable your battery remotely.
It seems like it would be FUD, as the capability to remotely disable the car (and dealing with the liability from doing that in a inconvenient or even dangerous place) seems rather expensive, when you might as well just check the milage when it's serviced (not servicing your new or new-ish car at an authorized dealership is a great way of slashing its resale value).
The reviews I've found claim they remotely disable your ability to recharge the battery if you go over the mileage you've paid for, so the car keeps going until the battery runs out and is no more at risk of dying at a dangerous time than if they hadn't disabled it.
Give it a few years, things are changing very fast. Useful electric cars will be affordable in 20 years for sure; Tesla seem to be aiming much sooner.
Anyway, even if nothing changes, in 20 years your £67K Model S will be on autotrader for £6K or less. You can pick up a 20 year Ferrari 456 GT for the price of a new Ford Focus and I'm sure they hold their value better than a Tesla will.
Tesla is thinking big. They want everyone in the world to eventually have a electric car. I think they will own a significant share of premium market after they release the Gen III.
However, they will be cars that people will desire at a cheaper price point than the Gen III, from different manufacturers. Also some people love certain car brands regardless of anything else, and mostly will stick to those brands. It will be great that those other car brands will be able to more easily make electric vehicles.
Those electric cars from other manufacturers will most likely have a Tesla battery and charge on Tesla charging station. I think this is true disruption.