Hacker News Reader

Comment by petra | Hacker News Reader

The theoretical maximum is higher than gasoline(1700Wh/Kg[2]), and that's what matters:

rechargable Li/CuCl2 could offer 1166.4 Wh/kg[1], while recyclable, non-rechargeable aluminum-air could offer 5200Wh/Kg[1].

Some batteries may have 200K recharge cycles[3].

Innolith is talking about a battery with 50K recharge cycles[4], but for the grid.

They also talk abuot building a 1000Wh/Kg battery for cars.

But who knows , they're a startup, so we'll need to wait and see.

And phinergy, another startup working on recyclable aluminum-air battery - is planning to open a factory in India in 2020.

[1]https://electronics.stackexchange.com/questions/4328/whats-t...

[2]https://phys.org/news/2018-10-catalyst-high-energy-aluminum-...

[3]https://www.pcmag.com/news/343967/researchers-stumble-on-a-2...

[4]https://www.pv-magazine.com/2018/10/10/innolith-launches-inf...

View on HN

It’s important to realize gasoline is measured in thermal energy which is much harder to convert to useful work than electricity. A reasonable comparison cuts gasoline’s energy density in half at a minimum while a more practical analysis includes the much higher weight of gasoline engines.

> while a more practical analysis includes the much higher weight of gasoline engines.

Currently electric car batteries weigh far more than a gasoline engine - and the range is still worse. Saying the gasoline engine is heavy is weird when you're talking about replacing it with something significantly heavier.

The reasonable rough approximation ought to be comparing two complete power trains including stored energy.

For an AWD ICE car, that includes fuel, tank, engine, transmission, differentials, and drive shaft. For an AWD EV, that's two motors, batteries, and the rest of the cabling and infrastructure required for the batteries and motors.

I understand that and agree with you, but the wording of OP implied a gasoline engine is so heavy it would have a significant impact on the equation - with current technology that's really not true. You can find inline-4 engines in modern vehicles that weigh under 300 pounds, whereas the motor and inverter pack in the Model 3 weigh 350 pounds.

Even so, this is all trumped by the fact that the battery pack by itself weighs about 1,000 pounds more than the motor.

Those 300lb engines would be a poor replacement in a model 3 competitor. Plug in hybrids can get away with it, but at that point all 300lb could be used for larger battery packs.

> Those 300lb engines would be a poor replacement for a model 3 competitor.

Are you saying a 300lb engine can't compete with a Model 3 in terms of driving performance? Here's a 5+ year old engine that weighs 290 pounds, makes just under 300 hp and lb-ft of torque in a stock vehicle.

https://media.vw.com/en-us/releases/96

With a mild tune (software upgrade, about $1000) you can turn a Golf R into a 11 second 1/4 mile car (you can do the same with a GTI, but it's a lot more work because it's FWD). It will literally run laps around a Tesla at a racetrack because the Tesla will overheat.

https://jalopnik.com/here-s-what-happened-with-the-tesla-mod...

Anyway, back to my original point - OP made a comment that seemed to me like the weight of gasoline engines was really important in comparing efficiencies - it's not. Right now gasoline vehicles are lighter, have longer range, and for many people are more convenient. I look forward to the day electric vehicles have improved enough that they are lighter, have longer range, and for many people are more convenient. But right now battery storage density is a real limiting factor in making that a reality, and the weight of the gasoline engine is nothing more than a second order effect.

If you feel I am wrong, please provide numbers/calculations in your rebuttal - it will make for a much more data-driven discussion.

> It will literally run laps around a Tesla at a racetrack because the Tesla will overheat.

Wow dude you're the king of misinformation. Why not link to a 2019 article reviewing Tesla Model 3's Track mode? https://www.motortrend.com/cars/tesla/model-3/2019/tesla-mod...

The Model 3 has a dramatically different thermal management architecture and does not overheat like the Model S did.

Your article is reviewing a feature that hadn't been released to the public yet (your article is also from 2018) and strikes me as a bit of a PR piece. It does mention an impressive lap time scored by the Tesla - which is great - but I don't see anything that says the Tesla can drive multiple laps in a row without overheating.

So far nobody has posted an impressive time for any Tesla on the Nurburgring - the track considered the gold standard for performance lap times for decades - and that seems to be because nobody can make it a single lap without the car overheating. I am aware that in the past month Tesla has been making a significant effort to post a good time, and they still haven't done so even with a custom vehicle.

https://www.digitaltrends.com/cars/tesla-model-s-nrburgring-...

If you can show me a video or article of a Tesla racing 10+ laps in a row at any track without issue, or even doing 2+ fast laps in a row of the Nurburgring let me know. I just googled and couldn't find anything, other than anecdata of Teslas overheating and going into limp mode.

> article is reviewing a feature that hadn't been released to the public yet

Since that article is about "track mode", you should be aware that is has been shipping to the public since late 2018.

Your suggestion would be a horrific fit for a mainstream car in the model 3’s category. Fuel economy, emissions, and long term reliability are why they don’t put an engine like that in a Honda Accord.

In terms of real world preference, EV’s and plug in hybrids low speed handling characteristics are vastly more useful than top speed. So the 200hp Honda Engine is arguably undersized even in a lighter car.

A Tesla Model 3 usually sells for something like $50,000, I'd guess - it's not competing with a Honda Accord. This article is a year old so it's outdated, but it suggests an average sale price of $60,000 for the Model 3. This clearly puts it in the luxury/performance price bracket.

https://cleantechnica.com/2018/08/23/tesla-model-3-average-s...

Apart from that, the EA888 engine I mentioned is used in the VW Golf, the 7th best selling car in the world.

https://www.best-selling-cars.com/global/2018-full-year-inte...

The engine is considered highly fuel efficient, rated well above 30mpg for most variants and many people report >40mpg in real world usage (obviously this data will be biased towards higher numbers).

The emissions, however, are supposedly great but I don't trust VW at all on this front so I'll just assume they're doing something shady. They are being legally sold in California though, which is really the highest emissions bar there is, so I'm not sure what you'd expect from them (again, assuming they aren't doing shady shit).

And unusually for VW, this is considered a reliable engine. It's been around and iterated upon for much longer than any Tesla.

There is no apples-to-apples comparison between electric and gasoline vehicles, but claiming one of the most highly awarded engines in recent history is lacking in performance is like claiming Teslas are worthless in cold climates - it's more rumor than fact.

I personally know people considering a mid tier Honda Accord vs the base model 3. The numbers they have run provide similar lifetime costs, though clearly many early adopters are less price conscious. Historically, Tesla was not going to make a base model when their was more profit from manufacturing a more expensive car.

Back to that IC engine “With a mild tune” is an issue. You can get reasonable performance on a tiny hatchback with the stock 149 bhp EA888 but that’s not going to cut it on a midsized sedan. Want more performance from the same basic hardware and you will make some real sacrifices, weight, fuel economy, engine life, etc one or more things need to give.

Pretending otherwise is to suggest vast numbers of automotive designers are idiots.

I feel like you just ignore all the numbers I write and invent new ones. A stock Golf R with an EA888 engine makes 292 hp and 280 lb-ft of torque - significantly more than any Accord you can buy new. A Golf R with a mild tune (just software) will generate closer to 350 hp, and moderate-serious tunes start around 400 hp. They also use this engine in their Passat, which is a mid-size sedan which competes directly with the Accord and weighs about the same.

https://www.goapr.com/products/ecu_upgrade_20tsi_gen3_mqb_r....

I'm sorry, but at this point I'm just going to say you have no idea what you're talking about when it comes to IC engines and exit this thread.

You need to be making an apples or apples comparison for different numbers to be meaningful.

The base golf gets 25 city / 34 highway a Golf R gets 22 city / 29 highway. With a mid tune that golf R gets even worse mileage. These numbers drop further when you talk about a larger car with more wind resistance or weight. https://www.fueleconomy.gov/feg/PowerSearch.do?action=noform...

Meanwhile the larger Honda Accord is getting 30 city, and 38 highway. https://www.fueleconomy.gov/feg/PowerSearch.do?action=noform...

Taking a 25+% cut to fuel economy to use a lighter engine is a really bad tradeoff for most people. Especially when the point of this discussion was talking about fuel density comparisons.

On top of this IC’s also need a much larger and more complex transmission among other things which should be included in the weight comparison.

Yup. I drive a lot and don't care much about performance as long as I can get up to highway speed on the on-ramp. Fuel economy, maintenance, longevity all matter a great deal.

Could they not have meant all that is included with gasoline engine cars? I didn’t interpret their statement as to be taken literally to strictly including the weight of engine alone.

The parent was specifically talking about energy density of gasoline versus batteries in Wh/kg. The battery weight is already considered.

Parents implied point is that something like miles for functional equivalence / energy unit denaity is much more telling. The ICE is like 37% efficient while battery is >90%, so while hydro carbon fuel is more energy dense, the usable density is considerably less than it's raw energy content due to waste.

The battery is not the same thing as the motor. He’s saying that a motor is lighter than an engine, and this somewhat offsets the fact that batteries currently have lower energy density than gasoline.

Fascinating. I’m always intrigued in these style of answers. Where do super-capacitors fit in this?

Supercaps can provide higher peak energy flow, either for acceleration or regen. They can buffer the battery from rapid energy flux.

I doubt very much they are going to actually store even 1kWh of total energy, because mass storage isn’t their strong suit.

At least a 10x factor below.

A table you may be interested in:

https://en.wikipedia.org/wiki/Energy_density#Table_of_energy...

>> Fascinating. I’m always intrigued in these style of answers.

Me too. Does anybody know a technology site/community, focusing on the theoretical limits of what various technologies could do, and how do they work , instead of the day to day news?

The stackoverflow family I’d sited is good for discussions like this and is often nicer than the halo site when it comes to inclusion of questions (I use SO everyday but trying to give back by answering a questions or even asking one I have had not so great experiences). I really like the astronomy and physics exchanges.

That wouldn't be a huge problem, and probably much better for the world if we "gased up" with aluminum

Aluminum requires huge amounts of electricity to produce. Hopefully that electricity isn't coming from fossil fuels.

However it does appear to be highly recyclable, at a fraction of the energy cost.

> Aluminum requires huge amounts of electricity to produce.

That’s kind of the point. If you can produce aluminum efficiently (i.e. reduce Al2O3 to aluminum metal without using much more energy than the theoretical minimum) and then oxidize it back to Al2O3 efficiently in a car, you have a battery that stores a huge amount of energy.

There is a rather spectacular non-battery use for this energy: thermite.

I can't remember where I read this (busy at work so, sorry, no source), but:

There's a production/re-use cycle in Europe wherein Aluminium is processed and smelted in Iceland, turned into ingots which are sent to mainland Europe to be manufactured-with, and then resultant used product sent back to Iceland for melting down and processing back into ingots.

The point being: Iceland uses geothermal energy to work Aluminium, so the entire process is one grand re/charge cycle of a significantly-used and energy-dense process.

Next step - small, modular fission power sources for the likely-hydrocarbon based and heavily-polluting ships transporting it all back and forth.

The beauty of electricity is that it can come from anywhere. And you don’t have to change anything in your process to change source of electricity.

Electricity is also much easier to transport over long distances, using high voltage DC lines, than it is to transport oil or gas via pipeline or truck.

And oil is much easier to transport over really long distances on tanker ships.

Only if you neglect the huge externalities of tanker ships, with the extremely dirty fuels they burn and the occasional catastrophic spills.

If you can have batteries with higher energy density than oil, why not fill up a ship with batteries?

Because A/C lines are more efficient than either oil or batteries.

Not over ocean distances though?

and can then be burned to make electricity

Batteries will also burn under the right conditions!

What about volumetric density?

Reply on news.ycombinator.com