Hacker News new | ask | show | jobs
by avernon 1735 days ago
The problem here is the form factor of the cell as much as the chemistry. Pouch style cells are very difficult to regulate temperature in. This is a key reason why other automakers like Tesla use the small cylinder cells. Chinese companies like BYD have always used lithium iron phosphate. It has less energy density than lithium nickel batteries like NCA and NMC, but is cheaper. The patent was never taken out in China and the key patent expires for the rest of the world in April 2022. As much as 3/4 of vehicle batteries might end up being LFP. Improvements in the chemistry and less need for cooling systems and packaging mean its disadvantages aren't so glaring. Plus there aren't enough nickel mines to supply the coming avalanche of demand. Lithium, iron, and phosphate are plentiful.
2 comments

FYI nickel is plentiful and unlikely to ever be in shortage. It's just so cheap right now that most nickel deposits are "non-economic", and thus not counted as part of proven reserves. The same thing is true with uranium for example.

see here for the difference between resources and reserves,

http://minesqc.com/en/informations-sheets/what-are-mineral-r...

We are currently extracting 2.5 MT of nickel a year, from reserves of 94 MT. Resources are estimated at 300 MT, which is definitely a lower end, as prospecting tends to concentrate profitable ores. If the price of nickel went up, we'd see more resources becoming reserves and more discovery of new resources.

If batteries are easily swappable with minimal tools, then a lot of range concerns go away as well. I think battery ownership and care needs to be decoupled from the vehicle. I also think the federal government should enforce interoperable batteries between vehicle brands.
The cars shape and structural components are designed around the battery. The cars software, motors, charging electronics, software, cooling systems, etc are all designed for the specific battery. The battery eats up a significant fraction of the cost of the car, and the systems designed around it a significant fraction of the remainder.

Asking for interoperable batteries between vehicle brands is like asking for interoperable engines between vehicle brands, it's ridiculous. The government should not be saying "hey, that $xx,xxxx item with a million constrains and optimizable variables, why don't you make it meet this arbitrary standard so people could theoretically put it in another chassis that wasn't designed to let you optimize it as well as possible".

Certainly no one should expect to be able to swap the main batteries that are embedded into the chassis of the car. The total weight of the batteries in a Tesla are around 1200 lbs.

But it seems like there could be a hot-swappable portion of the battery, kind of like a separate "reserve tank" (although it wouldn't actually be reserved).

Some back of the envelope math says that if 1200 lb battery gets you 250 miles, a 200 lb hot-swap battery would get you 40 miles. Possibly enough to get to your next destination.

> a 200 lb hot-swap battery would get you 40 miles.

According to one report, for a Tesla Model 3, a supercharger can add ~100 miles of range in ~10 minutes. It's hard to see how any improvement over that could possibly justify the immense additional complexity of physical battery swapping for only 40 additional miles of range.

https://insideevs.com/news/506520/tesla-model-3-supercharger...

Does the Supercharger network come with a fleet of drones that can autonomously repair a charger in the middle of nowhere, that broke down one late evening of a particularly snowy winter Sunday?

That said, it's probably a UX problem. Once EVs stop competing on range so much, it'll make sense to just designate the last 10% of battery as "reserve" and not count it in the battery level indicator.

Somehow gas pumps get repaired/maintained in the "middle of nowhere" today. Eventually the economic incentives will align that that the charger networks get maintained no matter where they are.

Arguably it should be far easier to get those economic incentives aligned as chargers are far simpler mechanically (they are just plug sockets with weird over-engineered male adapters) and most of what breaks on them is either vandalism or a small subset of existing problems of gas pumps: credit card reader malfunctions, display/screen problems, internet connectivity issues for account management/credit card transactions. (All the human UX points of contact.)

> Once EVs stop competing on range so much, it'll make sense to just designate the last 10% of battery as "reserve" and not count it in the battery level indicator.

Most already do (even while still competing for range) because it's a battery maintenance requirement. Li-Ion cells generally don't like being 100% full, especially not for long periods of time, and sometimes have a preferred "directionality" (ie, a cell should only be charging until it hits 100% and then you can draw from it and vice versa once you start drawing from the cell you should keep doing so until it hits 0%) so battery controllers already have to do a bunch of math to keep a "reserve" so that they don't violate "directionality" (you always want cells in the "charging" direction available even while driving for regenerative braking storage, for instance) and don't generally hit 100% charge for long rest periods, but instead 95% or so.

I thought we were comparing fast charging to battery swapping. Surely a machine that physically swaps out a battery is going to be significantly larger and more complex than a supercharger, and therefore also much more likely to break down..
No but there are always multiple supercharges at each station.

> it'll make sense to just designate the last 10% of battery as "reserve" and not count it in the battery level indicator.

Very questionable? Why would you do that?

We don't do that for gas cars either.

What about small generator and a canister of gasoline?
That’s what BMW did with their i3 REX models: a scooter engine and a tiny tank, hooked up to the electrical system. Turns out though that the extra weight impacted the performance without adding much in the way of range to the point that it wasn’t really worth it.
A 2000W inverter generator could give a Chevy Bolt an extra 14 miles of range after 3 hours of charging. A 240V generator could shorten that time or lengthen the range but would also be larger and take up more room in the car. Carrying gasoline in the back of your car all the time in case you run out of battery is somewhat dangerous. The generator means less cargo space as well. For a long trip to the middle of nowhere it may make sense to carry a generator and some gas. For a trip into town it may not.
> Asking for interoperable batteries between vehicle brands is like asking for interoperable engines between vehicle brands, it's ridiculous.

Engines are interoperable to a large degree; you can switch out the engine+ecu of your car for the engine+ecu of another car more easily than you'd think because the majority of the effort will be in changes requiring an adapter plate and shaft for the transmission.

There are hurdles that make it harder (for example, auto transmissions have software that expects a particular set of engine characteristics), but by and large most engines are isolated enough from the rest of the car and the drivetrain that you don't need to worry.

Batteries are to the same degree. if you really wanted to, you could take a bunch of liquid cooled Tesla modules and fit them in a old leaf, when it’s batteries degrade.
It's not ridiculous but it does lead to this conclusion: That there is probably a single optimal "skateboard" type design. Manufacturers are converging on this anyway.

https://en.wikipedia.org/wiki/Skateboard_(automotive_platfor...

Then the real truth that no one wants to hear, is that there is no need for so many different types of passenger vehicles.

It's astonishing how similar all cars are becoming, outside of the superficial stuff. For a reason. Convergence on the optimal solution.
They are also optimizing for superficial appeal to customer, not 100% functionality.
If that’s actually true there will be one player to rule them all. Tesla essentially been selling same car for a decade.
This is nonsense. That's like saying 'everybody converging on a 4 wheel design'. The details of how the platform is built and integrated is still very different.

> Then the real truth that no one wants to hear, is that there is no need for so many different types of passenger vehicles.

Good that we have you to tell us what vehicles people should drive.

> Asking for interoperable batteries between vehicle brands is like asking for interoperable engines between vehicle brands, it's ridiculous

No, it is like asking batteries for radios follow the norm (so that they can be swapped), or petrol to be the same for all cars, with the same pouring mechanism (a round hole), or USB connectors to be the same between devices.

Swappable batteries preclude the large weight savings of a structural battery that is an integral part of the car's chassis.
I'd say it should be possible to have a certain degree of compatibility just like I expect my GPU to work on any of the MB on the market supporting PCI-e(Space and cooling issues asside). The battery should be just a battery with a protection system. Charging and BMS should be separate and replaceable along with the battery. The OS of the car should support x number of BMS systems with a unified protocol.

You want a Li-Ion pack? This BMS and this shielding are required. You want Lead batteries? This BMS needs to charge it. Do you want that new battery tech? You need the new BMS for it and an OS driver for the BMS. Plug it in and it's done.

The battery is the most expensive part... which susceptible to wear, abuse and fraud. (and fire). Making them easily swappable with minimal tools means they'll be easily stolen.
The negative replies to this are largely valid, but ignore the fact that we can have multiple strategies at the same time.

I for one would love to be able to take my battery out, swap it, have a few so some are always on solar charge. Be able to take one onto the boat. Use a new one for higher performance special occasion driving, and an old one for bumper to bumper slow commute which I can wrong every last mile out of. Have a few which default to home backup and can support the grid.

The highly optimised around the battery argument is very good. But it doesn't mean that we can't also have a company focusing on interchangeable batteries with a third party market.

Let the buyer decide. Neither solution will get 100% of the market, or zero. Why write one off right now with confidence when both could have a healthy market for the hugely diverse users (ie not just SV nerds)

The fed enforcing interop wouldn't be black and white. Just like Mercedes are allowed to sell super low MPG AMG Sports cars because they also sell lots of small city cars. The fed could encourage things in this direction without being black and white and mandating 100% of batteries be replaceable.

I think the downvotes of the parent comment here aren't what downvotes should be for. He or she brings up a very interesting point

Literally nobody is arguing against some companies being allowed to do that.

And Nio does in China.

Easily, yes, minimal tools, no.

But a 93 second full battery replacement for the Model S:

https://yewtu.be/watch?v=H5V0vL3nnHY

The people I know with electric cars charge at home the vast majority of the time. Swapping adds a lot of complexity and cost in the pack design. NIO is doing swaps as another commenter pointed out, but more customers in China live in apartments. In the end, I think charging stations will be ubiquitous, even for apartment dwellers, and swaps will be rare.
They are too heavy to swap them by hand
The comment you replied to said minimal tools.

In practice this would probably mean something like: a vehicle hoist, a suitable lifter, a couple of sockets and a racket handle, and a screwdriver or two.

Tesla actually had a battery swapping machine at one point, I think it was at the Kettleman City location. There's no reason the battery and charging infrastructure can't be leveraged for all passenger vehicles. The battery on your electric vehicle should be like the propane tank on your gas grill - just swap it out when you need to. (This doesn't preclude recharging it)

Not to mention, when the battery reaches the end of its normal life cycle. Just goes straight into the recycling pipeline.

Tesla's swapping machine was just for tax credits. It was never built to be something they'd sell to customers. They built that one demo location and then abandoned battery swapping completely as not viable.
If they legitimately cared instead of faking it they could probably do better. Especially if the target is a maintenance-style machine setup rather than a 30 second gas station rival setup.
Nio's cars support battery swaps:

https://electrek.co/2021/07/09/nio-plan-open-battery-swap-st...

You can rent the batteries and upgrade and downgrade the capacity as you like. It also makes it easy to replace the battery in the future if you buy it outright.

They offer 70 and 100 kWh packs now. They're aiming for 150 kWh packs next year:

https://insideevs.com/news/465188/nio-150-kwh-solid-state-ba...