Hacker News Reader

Comment by zackmorris | Hacker News Reader

I had no idea lithium ion batteries had become so inexpensive. Here is the top search for "kwh battery" on eBay right now:

http://www.ebay.com/itm/Li-ion-Lithium-24vdc-100ah-2-4kwh-NE...

So with shipping, $869.96 for 2.4 kwh, or about $362/kwh. A $3 gallon of gasoline has 33 kwh, but only 8 kwh of usable energy due to the inefficiency of internal combustion engines, so $0.375/kwh.

So batteries are somewhere around 1000 times more expensive than gasoline, but, if they can be recharged more than 1000 times they may actually be cheaper. 1 kwh of electricity costs about 10-20 cents usually (call it $200 per 1000 charges). That's a total of $1069.96 for 1000 cycles, maybe $1269.96 for 2000 cycles, etc, something on that order anyway.

So $1 per cycle per kwh is pretty close to a breakeven against gasoline, and batteries are only going to get less expensive.

View on HN

Ultimately, the batteries will win, but into your calculations consider the time value of money: how long will a given usage take to reach 1000 cycles. On the flip side, gasoline prices are volatile, and electricity from solar keeps getting cheaper, too. In a steady, non-volatile price march.

I just finished the Elon Musk biography — there's a good discussion about how Tesla was founded on the premise that the market did not yet recognize how cheap lithium ion batteries had become.

The original prototypes involved connecting hundreds of small batteries in parallel.

https://www.goodreads.com/book/show/22535480-elon-musk

The production cars are still hundreds of small batteries in parallel. I wonder if with Model 3 they have the scale to move away from the standard form factors.

Yup, the current battery uses thousands of 18650s, but they developed a more 'economically efficient' 2170 cell for the model 3 - https://qz.com/879121/teslas-tsla-cheaper-more-powerful-2170...

>> the market did not yet recognize how cheap lithium ion batteries had become.

How did that happened in the age of the internet ? and are there still a lot of such technological "secrets" ?

"Secret" is not the right concept. It's not that people didn't know the price, it's that they didn't see the implications of the price for opening a new market. That kind of insight takes imagination as well as knowledge.

Happens all the time. Kind of how smartphones drove down the prices of components useful for VR goggles, and made them a feasible product.

Manufacturers in this industry don't exactly publish price lists.

At some point, the tax structure for cars has to be changed to account for electric vehicles wearing the roads and not yet paying to maintain them? I'm not sure, but I think that's what fuel tax pays for? So, the cost will go up at some point in that sense.

Then, on the other hand, $3/gal is a US-only phenomenon among countries that can really afford electric cars, everywhere else it is ~2x that? Europe, japan/korea/australia? I could be wrong here too but I think that's how it is. So there electric may already be quite a bit cheaper, depending on generation cost.

The vast majority of road wear is caused by heavy trucks. Passenger cars including electric models have very little impact on existing roads. But it would certainly make sense to shift road infrastructure funding more toward vehicle registration fees.

I don't have a source for this handy and I can't look at the moment, so I could be misinformed, but I'm pretty sure I've heard in many different places that the current fuel taxes in the US don't even come close to covering road maintenance costs.

I wonder if some kind of weight + mileage tax would work better?

States already rely on a variety of revenue sources, such as vehicle registration fees and tolls, in addition to gasoline taxes.

Ultimately I don't see how one could institute a fair system without converting all roads to toll roads (perhaps with variable pricing to manage demand).

Gasoline taxes are a reasonable proxy for "usage", so heavy users contribute more.

Registration fees treat everyone of the same make and model the same, be it a livery vehicle with six-digit annual mileage or a proverbial old-lady whose car is being driven around town twice a week. And one can't exactly levy a tax on kWh.

Fixed fees are much worse than gasoline taxes, because they lower the marginal cost of driving, making the already fixed-cost biased cost structure even worse.

(This is bad because it makes driving more very cheap for people who already have cars, the exact opposite what's good for extracting maximal utility out of minimal fuel use)

> I'm not sure, but I think that's what fuel tax pays for?

The high taxes on cars and gasoline usually goes to improve public transport.

your smart meter will take care of that... (if you charge a car at your home, they will know and charge accordingly)

I'm just fine with gas taxes covering EVs until a carbon tax can be imposed.

But converting electricity to mechanical power isn't 100% effective either, and the weight of the battery affects efficiency too. Petrol is much lighter, and its weight naturally diminishes as it is expended.

So at least for transportation, the breakeven is probably somewhat further than your calculation suggests.

The problem is not the stored energy. It is the time needed to store it.

Low on gas? Get to a station, take 10mn to refill and you're gone for 500 more miles. Low on battery? Get to a station, wait hours to get it charged back up and you can go for 300 more miles. The fact charging takes time means a long queue before you can get to a charger.

Drive for two hours, rest for 30 minutes while the car charges enough to drive two more hours. While not exactly the same usage pattern as a fossil fuel car, it should be enough. We need more quick charge points.

but that is not how people travel, ever.

besides time to charge the real big drawback of battery EVs is the weight of the packs

I would say this actually fits the majority of traffic patterns: https://www.rita.dot.gov/bts/sites/rita.dot.gov.bts/files/su...

EG: The vast majority of travel is for errands and work commuting. Probably a lot of that social travel in the linked survey is short distance as well.

Average miles a day is 30, which can easily be covered by an overnight charge.

Agree that the time to charge EVs is definitely a drawback for long distance travel (whether for social reasons or for transport etc.).

But this is how they used to travel when cars were first introduced. The Ford Model T could only get 170-200 miles [1] on a full tank of gas and gas stations were certainly as rare in the early 1900's as electric charge stations are today. Yet Ford still sold a ton of Model T's, and look at where we are today.

[1] Source: http://www.mtfca.com/discus/messages/118802/146015.html?1276...

>It is the time needed to store it.

For now. Battery tech is increasing at a fast pace.

You need to take into account the cost to the environment of burning gasoline. The true cost is closer to something like $8 a gallon. So now it's up to about $1/kWh. As someone else has already mentioned, Tesla's packs are about $190/kWh. You'd only need about 190 cycles before that starts breaking even, which could probably be done in a couple years.

Then.... take into account the cost to the environment of mining lithium and all the other elements going into the batteries and associated power electronics.

True, but that's a one time cost. Once EVs have a significant market share, everything will probably be recycled.

And if we play this game, to get it even, maybe we should compare the whole car since gasoline requires a big engine too.

My impression is that analysis which takes everything into account puts EVs far ahead in environmental impact, and I think they're slightly ahead in terms of total ownership costs.

Every gallon of oil that's taken out of the ground (besides plastics) goes into the air. Mining waste is very local and tractable compared to that.

Tesla is under $190kwh for their battery packs and falling.

I think the price / performance / capacity of battery for now and the next 5 - 10 years roadmap are already very decent. What i want to know though, if there are any possible improvement to Weight / Energy ratio?

The current battery are way too heavy.

The battery doesn't generate any energy though, it just stores it. I mean, technically, the gasoline is just storing energy too, but it was generated naturally long ago. You still have to burn some fuel somewhere to recharge the battery each time.

Many shale-oil outfits, spend more energy on retrieving the oil, than the oil actually contains. Oil can be profitable, without being a net energy gain, meaning it's true value is in it's ability to store energy.

It seems obvious that a battery needs to be charged. But it's not true that fuel has to be burned to recharge the battery. There are many other, cheaper, sources than burning things to turn a turbine to generate electricity.

8kwh per $3 results in $2.67 per kwh for gasoline. $362/$2.67 results in a cost ratio of 136 between bateries and gasoline. It's in the magnitude of x100, not x1000.

Lithium ion battery capacity decays with each charge down to 50% after 500 cycles

capacity decays with each charge down to 50% after 500 cycles

Not true with modern EV batteries. Tesla's packs typically show less than 8% degradation after 100,000 miles, which would be approximately 500 cycles.

https://electrek.co/2016/06/06/tesla-model-s-battery-pack-da...

Tesla batteries only show very little degradation because their capacity is limited by software[1]. You can "increase" your battery capacity by ordering a $9,000 software update.

[1]: https://electrek.co/2016/07/26/tesla-upgrade-software-limite...

>You can "increase" your battery capacity by ordering a $9,000 software update.

Thats only for certain specific model and delivery time combinations.

i.e just before they introduced the 90kWh option in production, some 85kWh cars were shipping with the 90kWh pack.

If you ordered a particular pack before the next step size started shipping, then that's all you got.

>Tesla batteries only show very little degradation because their capacity is limited by software

Not true. It's mostly to do with active management of the battery, and keeping it within the optimal conditions for much of it's life. These are things your phone and laptop can't (or rather: don't) do.

Everyone tends to plug in their laptop/phone and charge to 100%, then leave it there for long periods of time. Plus, laptops and phones tend to get very hot, and there's no active cooling of the battery.

This combination of conditions tends to reduce the life of batteries significantly.

>Everyone tends to plug in their laptop/phone and charge to 100%, then leave it there for long periods of time.

Why is this a problem for Lithium batteries?

Leaving lithium batteries at their extremes of charge state reduces their lifetime.

Tesla recommends that their customers to charge to only about 80% for normal daily driving, and that if they need the full 100% range that they should start driving very soon after it reaches 100% charge. They also recommend not leaving it at empty for very long too as empty cells are more likely to drop below their minimum voltage, and die.

I own an electric car (with a lithium ion battery) and commute daily with it.

It's now four years old, has driven over 45,000 miles (well over 500 cycles) and the total degradation is less than 10%.

Me too. Just traded in a 4yr old Model S with 52k miles and saw <10% degradation. Fully charged, it was 210 miles new, and about 192 after 4 years.

That depends greatly on the specific battery chemistry, the temperatures the battery is exposed to, the charge levels, the charge rates, etc. You can't just make a blanket statement like that.

This is a bullshit statistic that has somehow become a meme on the internet. It just flat out isn't true. There's EVs with over a hundred thousand miles on the original pack still doing great. Please stop repeating nonsense as if it is a fact.

I understand how frustrating that can be. Do you have a reference at hand that busts this myth?

Well, here's some data from a decent number of Tesla Model S in the wild:

https://survey.pluginamerica.org/model-s/charts.php

I'd say it looks like they behave rather well in general.

Thanks! That is pretty impressive.

Guy #1: Shaking pinecones will scare snakes away.

Guy #2: No it won't.

Guy #3: Do you have any sources that disprove that claim?

The onus is on the person making the ridiculous claim.

I can tell you from personal experience I have had a Nissan LEAF for about 2.5 years and ~24k miles, driven daily and it has lost no range whatsoever. But if you want a study or something, you're gonna have to Google for yourself man.

Hey, your anecdotal experience is better than nothing. I'm reading this thread as an interested observer.

Both you and 'jjawssd are making claims, and you're very confident that 'jjawssd is wrong.

This is a bullshit statistic that has somehow become a meme on the internet. It just flat out isn't true.

I don't think it's unreasonable to expect someone who expresses that level of confidence to have something to back it up. And if the false claim is so widespread, I'd also expect someone to have busted it and publicized it. From my outside, ignorant point of view, both sides are unproven, with an edge to you because of your own experience.

We aren't both making claims. He's making a claim and I'm saying it's an unsubstantiated internet meme. My little joke example was meant to illustrate that.

I'm sure you have devices in your own life that disprove his claim. Look at your phone. Does it really go to 50% capacity after 500 days? None of mine have. Sure batteries degrade over time and you'll experience better performance one day or worse another, but he's claiming after 500 days of using your phone you should be at 50% capacity. That's obviously baloney.

You are both making claims, the positive and negative of a particular characteristic of batteries.

When I first read your comment, I thought you were a little over the top, but it sounded like you likely had something solid to back it up, so I was interested in learning something new. Personally, I don't make strong statements like that unless I can pretty quickly defend them. I honestly wasn't trying to catch you out. And seeing a myth busted is generally a good time!

As for phone batteries, I just had to replace mine, so that's not a particularly useful example. As I understand it, there's a lot of circuitry in phone batteries these days anyway, so I'm not sure what failed. I'm not going to use my anecdotal evidence for or against your case, for what it's worth.

Be well!

If I had said "No they maintain 100% at 500 cycles." Then I would also be making a claim. I made no claim, I just said his claim is an internet meme that is bullshit. There's a big difference. I'm not sure why this is so hard for you to understand.

I don't save a folder full of links to answer skeptical people on the internet and I'm not going to Google for you. If you're interested in reading articles, go look for yourself.

Anyway whatever. This has been a colossal waste of time.

What an unreasonably combative response.

Next time just say "sorry I don't have hard evidence for this opinion on hand". Nobody expects you to know everything but they do expect you to treat people decently.

My 2.25 year old LEAF with ~9000 miles on it has lost about 5% capacity. I'd be very surprised to learn yours has not lost any capacity.

Are you relying only on the 12 bar indicator or on the more detailed data download from the car's telemetry? My car still shows 12 bars, but has obviously lost some capacity (max range at 100% charge is down to about 95 from 102/103), just like my phone that still charges to "100%" but obviously has way less capacity after 3 years.

From my perspective, losing "no range whatsoever" after 2.5 years is the extraordinary claim.

I haven't downloaded the telemetry but we drive it every day and the range we get is extremely consistent. I'm sure there must be some degredation but my point is we haven't really lost any range. The weather and our driving behavior is the main cause of range variation.

If his claims about losing 50% capacity after 500 cycles were remotely true that would be the main cause of lost range ina 2.5 year old LEAF.

From my understanding, the LEAF has a primitive battery management system compared to other EVs, including passive air cooling of the battery. Lithium batteries are fussy about temperature (they have similar temperature preferences to humans) and outside of this the degradation can be faster. Other EVs have liquid heating and active cooling of the battery.

The LEAF does have a battery heater I'm pretty sure.

Reply on news.ycombinator.com