Tesla poised to reveal 'million-mile' battery as soon as this year | Hacker News Reader

Tesla poised to reveal 'million-mile' battery as soon as this year

cnet.com

22 pointsjonbaer46 comments

is this due before or after full autonomy?

I guess first you will be able to buy the chassis option that will support this future battery.

Full autonomy depends, among other things, on regulators. Could be years before the right people are convinced it should be road legal.

Hints I see re: Cybertruck point to new battery by end of 2021 (CT launch). Few/no regulators to persuade of exotic tech.

I’ll settle for self driving package that doesn’t plow full speed into stationary objects. What regulatory hurdle is that stuck behind?

There are no regulations currently in the US that would prohibit full self driving. They simply do not exist.

Regulation is a lie told by Elon to excuse a lack of progress.

If there aren’t regulations, there will be soon. Tesla needs to address the obvious before it becomes a problem.

But will the Cybertruck actually be produced by the end of 2021 or were the battery requirements too high for the spec and price point?

Point of this thread is new battery tech arriving soon, and hints during CT unveiling were it would be the first Tesla vehicle using the new tech. Notice the new “Gigafactory” will double as CT manufacturing, and Elon has referenced new batteries well before any of this.

Sure, but there were hints of that during the semi and roadster unveiling too. 200KWh in a car smaller than a 3? Almost 1 MWh in a semi costing less than 200k?

The CT was always going to be built at either NV or a new factory. All new factories are doing battery production. If there's a hint there, it is a really weak one.

Having said that, I do think the CT will make it in 2022 and maybe we'll even see the lower trim versions in volume production by 2023. That would still be pretty impressive.

Obviously after. Full autonomy is promised for 2017.

The hardware to achieve 'million-mile' battery is already shipping in currently selling vehicles, its just a matter of fine tuning the software at this point!!1

Fine tuning being setting max charge 55% and never going below 45%.

  def accelerator_pedal():
    if speed > 1 mph:
      return 0
    else:
      return accelerator_pedal_analog_value()

You'd think they'd have their hands full with the robotaxi network, building 1000 solar roofs per week, getting the cars to automatically call a tow truck, getting the cars to autonomously find a valid parking spot, powering the Gigafactory 100% with solar power, building a pickup truck that's a better sports car than a standard 911, outfitting all supercharges affected by blackouts in California with a Tesla powerpack, building more than two semis, building the roadster etc. etc.

Its right after the roadster and semi are in full production. They are scheduled for ~4 years after their announcement if they don't get pushed back again.

I see no reason to believe this is less of a hype followed by vaporware.

Aren't we now past the mark to call Tesla vaporware. Musk build a legit revenue stream and for both customers and shareholders this was a very welcome change. I wonder if any of the incumbents would have a similar electric car as Tesla in 2030. They haven't moved an inch since Musk is taking their industry by storm, thats almost already 10 years. Even Audi has mayor problems manufacturing their etron.

> Even Audi has mayor problems manufacturing their etron.

The Audi e-tron outsells the Model X and the Model S:

https://ev-sales.blogspot.com/2020/04/global-top-20-march-20...

And the e-tron is outselling the Model X and Model S combined in Europe:

https://ev-sales.blogspot.com/2020/04/europe-march-2020.html

There's not much reason to buy a model s right now compared to the 3 imo. The model 3 is #1 on your list and sells 4-8x more than the next 5 on the list, including the e-tron

The e-tron is not in the same price category as the Model 3. The Volkswagen ID.3 is in the same category. The ID.3 will become the best selling EV in Europe, maybe as soon as this year, definitely by next year. Have a look at it:

https://www.youtube.com/watch?v=03n2H_EY2zA

I'm not calling Tesla vaporware. They make the best EVs on the market and I do like them.

Having said that, they have a bad habit of announcing products with _unbelievable_ specs that are just a little too good to be true and then delaying them constantly. For examples, see FSD, the Semi, the Roadster, and maybe the Cybertruck as well. Obviously we do not know for sure if the Cybertruck will be delayed, but it would be an amazingly fast program if it were to be produced in volume by the end of next year.

I still love the SX3, and Y, though. They are great products, IMO.

For context, a Tesla X or S battery degrades by 5% 50K miles and roughly 10% at 160K miles.

Good to know. My 2011 Honda has 61k on it, battery life has been a concern in an EV as I picture my next car sitting around on-and-off for a decade or two again. This one smelt new until a few years ago. My iPhone SE has around 15% degradation, for comparison. I like The Volt’s onboard generator approach, feels robust. Hopefully another while to think about it, as cost of maintaining a 9 year old/61k mi car is pretty good still.

If only Chevy had been able to deliver on the Volt’s original concept (with the electric motor connected only to a generator)— it turned out that that arrangement was actually pretty inefficient, and the production car ended up being more of a traditional series hybrid approach similar to the Prius (with the ICE and electric motors connected through a planetary gearset so both can drive the wheels).

BMW does something similar to the Volt’s original concept with the Range Extender in the i3– that actually is a gasoline engine coupled to a generator; it never directly drives the wheels, and it can’t be used as the car’s only power source (once the battery’s dead, you’re done, even with a full tank of gas).

The i3 range extender kicks automatically once your battery range gets low, and uses gas to maintain that range (while reducing power). So theoretically, your battery should never die while you have gas.

Duplicate. See

https://news.ycombinator.com/item?id=23182097

I was hype until the part about it being total miles, not range on a single charge. And then I was disappointed. And then uninterested as I thought to myself "yeah that makes more sense." And then suddenly angry, as I thought to myself "wait why they hell AREN'T they trying to develop a million miles battery on a single charge!"

I'm curious what the total energy storage would be, and what kind of destructive damage it could do. A million miles is a lot of miles to move a full Tesla, passengers, and luggage around.

Anyone want to do a back-of-the-envelope sketch and give some colorful units-of-measurement given instantaneous discharge (e.g. tons of TNT)?

Well using the most optimistic numbers I could find the Model 3 LR does ~429km on 270 MJ battery pack. Assuming efficiency of system remains the same for 1 million miles that is approximately 1x10^12J of energy, roughly 240T of TNT or around 30,000L of Jet A-1.

Using your figure of 1e12 joules, I thought, what about a hypothetical miniaturized fast breeder nuclear reactor?

Googling, I found a document that says regular light water reactors produce power with something like 1.8 tons of fuel per gigawatt per year. And fast breeder reactors potentially could be 100 times more efficient, they say.

So if I did the arithmetic right, and if Google unit conversions didn't fail me, you could get 3.2e16 joules from 1632 kg of uranium fuel, which ends up being about 2 ounces per 1e12 joules.

So, sure, it seems doable to me. The "battery" would be about the same weight as a regular alkaline 9V. A bit smaller though.

Dividing 1000000 miles by the up-to 390 mile range of the 100 kWh Model S indicates 256.4 megawatt hours for a million miles. That's about 221 tons of TNT equivalent.

Based on this[0] thread on the Tesla forum, I'll use 0.31 KWHr/mile. If that's the case, a million mile battery would be equivalent to 267 tons of TNT.

[0] https://forums.tesla.com/forum/forums/miles-kwh [1] https://www.wolframalpha.com/input/?i=%280.31+KWHr%2Fmile*1+...

Ok so.. 100kwh = 350 miles ish

1,000,000 miles would need a capacity of 285mwh...

Or... 245 tons of TNT...

I don't want to be in the same zip code as that vehicle

Lol so apparently this sounded like a good enough math problem to get a lot of people interested

Assuming 500 watt hours per mile for Cybertruck (likely the first Tesla vehicle using it), a million miles is 1.8e19 ergs.

1 kiloton TNT is 4.184e19 ergs.

Million mile battery is 0.43 kilotons TNT, about 1/35th of a Davey Crockett man-portable nuke.

Well, a more reasonable value for a tesla is 300 wh/mi, and a tesla could pull off 1,000,000 miles of range with a mere 12.91 gm of plutonium.

If the the cybertruck really needed 500 wh/mi, it would require a bigger million-mile battery - say 21.53 gm of plutonium.

But you know the environmental folks would complain bitterly, so maybe tesla should offer a "green" cybertruck option with a heavier 22.66 gm thorium battery. It will probably require retuning the suspension for the extra weight.

https://en.wikipedia.org/wiki/Energy_capacity

Just have to efficiently & compactly convert thorium to electricity at a suitable rate in a sub-1000 lb package priced under $65,000 (cost of battery + 500,000 kWh).

I picked 500Wh/mi because Cybertruck will likely be first to use the new batteries, and will require more power than the cars/SUVs.

Trying to develop a battery that can do a million miles on a single charge right now would be like telling a Chinese kingdom circa 500 AD to develop a working moon rocket because they have fireworks.

At least we now know that a moon rocket is even chemically possible

If you think about the energy required to drive a million miles, all concentrated and stored somehow in a car, it would be extremely dangerous. If that energy is suddenly released in an accident, it would probably destroy a city block.

1e6 miles at 3kwh per mile = 3e5kwh = 1.08e12 joules = 0.25 tons of TNT. Not sure how many buildings that would destroy, I have no practical experience with explosives. But it's a decent sized truck bomb.

Yep lets hope Telsa cars stay expensive and we get a public hyperloop instead and we can just ride bicycles around in the clean air.

Remember when Richard Hammond crashed the Rimic, it burned for 5 days!

I don't know, energy doesn't need to be that unstable, hidrogen fusion gives a lot of energy and don't need much mass of it, not more than current fossil fuel methods.

PS.: Of course no fusion reactor that fits in a car will probably de developed in the next years

You joke, but a while back there was a guy who had developed a new aluminum air battery technology.

"Jackson claims that if you replaced a Tesla Model S lithium-ion battery with one of his aluminum-air batteries of the same size, range would increase from 370 miles to 1,500 miles."

search for "jackson aluminum air battery"

There were articles where it was "a conspiracy" that he couldn't get funding for his breakthrough battery.

which leads to the downside: when the battery discharged, you had to recycle it and replace it with a new one.

Just to be clear, what Jackson is claiming to have developed is a new, better, electrolyte for aluminum air batteries. Aluminum air batteries themselves go back to the '60s. They have around 8 times the energy of a lithium ion battery of the same weight, or around 4 times the energy of a lithium ion battery of the same volume.

He seems to be doing better now on getting funding and interest [1].

I believe that the claims of 1500 miles in a Tesla Model S are based on the same volume. If you went for the same weight, you should be able to get an EV with something approaching 3000 miles.

That's enough that having to swap batteries when it gets low would be acceptable to many people, assuming it is reasonably quick. Jackson is claiming they could build swap stations that could do the job in 90 seconds, which if true is surely fast enough.

The discharged batteries can be recycled.

I believe I read somewhere that recycling the aluminum anodes from a discharged battery does take a lot of electricity, so you'd want to do that somewhere where electricity is cheap and green.

[1] https://www.automotivelogistics.media/electric-and-autonomou...

you have to admit - having to replace the battery, even only once a month, is a huge impediment to adoption.

Works as long as you can stop by a suitably naive civilization every so often to strip-mine dilithium crystals for a few bottles of synthahol.

Reply on news.ycombinator.com