[0]: https://www.google.com/amp/s/electrek.co/2018/04/14/tesla-ba...

[1]: https://news.ycombinator.com/item?id=15144315

Basically, the point is a standardized, easily removable, hotswappable widget makes a lot of sense when it's a minor item. But when it's half the weight, value, volume, and engineering challenge, you have to sacrifice all of those attributes to use the hotswap widget.

This originally made me think, if only we could pump fluid electrolyte in & out of the car, and let the battery take whatever form it likes. Then I realized that's called a flow battery, and then I realized that a hydrogen fuel cell is a type of flow battery.

A big reason for Tesla's design was that Panasonic's Cells (a partner of Tesla) was superior on power-density / weight.

I expect that power-density / weight will continue to be an important factor in electric cars, especially when you need thousands of these cells to power an electric vehicle. As such, I don't expect lithium ion cells (even "standard" 18650 ones) to become a commodity.

I get that you said that they each have their own chemistry, but are the owners of all of the chemistries choosing their customers?

Spoiler alert: Chevy Bolt is over 230 miles. Nissan Leaf is 150 miles. If you want 300 mile electric range, your only option is Panasonic (and by transitivity, Tesla, who is their only customer at the moment).

That feat alone means that Panasonic cells give you a real technological advantage. A Panasonic Li-Ion cell is worth far more than a typical Li-Ion cell. It is NOT a commodity, but a fundamental business advantage these days. If the other LiIon cell manufacturers manage to get real improvements to Power Density / weight, then maybe we can talk about Commodifying LiIon cells.

But for now, its Panasonic only.

* better sustained performance compared to NCA

* better peak/burst performance compared to NCA

* easy to adjust formula to show preference for capacity, performance, or longevity (most automakers skip on the 3rd)

Tesla prefers NCA battery because:

* it has more capacity / weight

* it's much cheaper due to lower cobalt requirements (main reason)

The main reason traditional automakers tend to shy away from NCA because:

* it is far less stable than NCM, which puts it a much higher risk for fire or explosions

* it costs more than NCM to make stable (I suspect Tesla chem is cutting corners in this area)

You can decide whether the trade off of safety for range is worthwhile for you; traditional automakers err on the side of safety.

Here's a scientific paper about the safety of various battery chemistries (warning, PDF): https://res.mdpi.com/wevj/wevj-06-00572/article_deploy/wevj-...

Summery is that, compared to NCM, NCA inhibits thermal runaway at much lower temperatures, offers significantly less short circuit resistance, slightly less overcharge buffer before thermal runaway, significantly faster thermal runaway after overcharge buffer is exhausted.

The result of thermal runaway is fire or explosion.

But in any case, technically even the Tesla just uses hordes of 18650. But that doesn't make it any easier to remove the pack itself.

In principle they should be good (you just need some small battery to start them), but the problems of making them safe may overwhelm any gain you get from using fuel cells.

What? Most laptops on the market today have removable batteries, to the point where you can buy spare battery packs for them for long trips. Same with a lot of smartphones.

Unless of course you mean Apple products.

There are few consumer laptops with removable (pops right out) batteries these days. I haven't seen one since the early 00s.

At best there are still those which are user-serviceable (requires a screwdriver and some finesse at the least). And Apple products where you cannot do even that.

Some business models still have the external removable battery, but have lower specs and higher price than consumer counterparts.

Almost no modern smartphones have removable batteries. It's rare enough that the LG G5 tried to make it a selling point a few years ago.

* Nexus 4

* Nexus 5

* Nexus 6P

* Samsung Galaxy S7

* Samsung Galaxy S9

The only recent phones I was aware of with removable batteries were the LG G*, but I don't think the latest model of those has a removable battery anymore either.

Even though swappable batteries do continue on the low-end, I think maybe [in the US] only Motorola actually makes spare batteries available for consumers to purchase, even for devices that don't have readily swappable batteries. I think this year Motorola will ditch swappable batteries, pretty sure, but will supposedly continue to make spares available for select models via Ifixit.

https://www.renault.co.uk/renault-finance/battery-hire.html

https://www.greencarcongress.com/2017/12/20171218-nio.html

For a quick recharge the current-generation 120 kW SuperChargers were deemed good enough. Now that they're growing their fleet of mobile service trucks, battery replacement might get re-introduced as an at-home service.

Cars overall depreciate on the scale of 10/15 years or so. Right now the depreciation rates are the same. It's an interesting artifact of ICE vehicles that people don't expect any car's drive train to last more than ~15 years. It shouldn't be a surprise that car manufacturers were skeptical of battery tech right up until they hit about that same magic lifetime number. It also shouldn't be a surprise that car manufacturers won't have a lot of incentive to push it past that magic lifetime number, because that would affect car replacement rates and that would affect the bottom line.

Mostly battery replacements look to keep the used car / secondary markets busy, and if we expect to see interesting innovations in battery replacements that's probably where we'll see them as more electric cars get resold to secondary and tertiary owners. (Just as we see the used car markets pushing innovations in after-market parts in general.)

You are right in that maybe leaves room for a disruption that someone might be interested in a "car for life" model where electric cars might be built far more sustainably with an ideal towards larger reuse of the vehicle over the owner's lifetime and a larger depreciation scale on the order of multiple decades.

That would be a huge game changer, and would potentially be great for the planet (less steel production alone, conceivably). It would also take a lot of work to make it happen, given the existing industry reliance on the current business model for most of a century now. It would also be one extremely huge disruption in so many downstream industries, secondary markets, etc, to the point where it may be more of a revolution (in the industrial sense, but also perhaps the war sense) than a "simple" disruption.

https://nhts.ornl.gov/vehicles

Of course, we can view the car as a single purpose ‘move from A to B’ device, but there’s other viewpoints where the car has a performance (speed, efficiency, handling assist, smoother engine management) profile, quality of life and security features (auto cruising, speed limit detection, tire pressure detection, auto braking etc)

All of these should have evolved enough in 10 years that you get significant benefits from an upgrade.

Obviously new cars are in general an upgrade. Most people driving old cars do not do so because they want to, they do it because new cars are too expensive.

There's certainly degradation that occurs to a car with lots of mileage, but it's generally with the interior and the electrical systems. There may be some exterior corrosion, too, if you live in the appropriate climate though corrosion resistance has improved a lot on cars lately. And, of course, all bets are off if you don't properly maintain the car.

I should get around to replacing some suspension components to make it more comfortable to drive, probably will later this year when I get some time. They're not terribly expensive.

The other side of it too is accidents are likely to always be an issue with cars (even in a world of self-driving cars, I suspect), and there are plenty of other safety reasons cars shouldn't be expected to be driven for too many years without replacement.

Electric or ICE, as you hit 150-200k, you'll have long since worn through all kinds of rubber bushings on virtually every part, you'll be approaching the point where you've had to replace starter or alternator, you'll have (hopefully) replaced your suspension a couple of times, and so on. As the value continues to decline, the 'value' of spending parts+labor money on any of these replacements starts to decline.

It helped that I was able to get a brand new loaded Volt for $9,000 off plus an additional $7500 tax rebate. which in the end means that I have the best of both worlds for $25,000 with a full warranty etc..

Having said that, I remember that 20 years ago, those kinds of parts would need to be replaced after maybe 60-80k miles. However, everything seems to be better built now - my van has 214k on it, and no sign of needing replacements - same goes for all the cars I've bought and sold (20 or so) in the last 3 years.

Further on down the mileage line, virtually everything is wearing out. Rubber bushings in control arms. Rubber bushings in all your rear suspension points. There's rubber EVERYWHERE on cars, and it all wears out. Motor mounts, transmission mounts, shock mounts. Again, it'll still go down the road, but it won't drive like new. You might have a lot of slop in steering, slop in cornering. The car might pull one way or another under acceleration. It'll be louder, harsher, clunkier.

Going a little deeper, probably your tie rods and sway bar links are shot, as are the control arm bushings (which may or may not be integral with the entire control arm).

Not necessarily by design, it's simply inherent to electric drive, you have much less "metal grinding against lubricated metal with a controlled wear rate" situations that seems to doom ICEs and the associated transmissions on the long run.

There was an Israeli battery/car tech startup that was built on that model, but it sadly went kaput. I still firmly believe that the hot-swap model is the winner long term.

Batteries aren't as "fungible" as something like gas or electricity is. (is that the right word to use for this?)

You can be reasonably sure that gas is gas, and electricity is electricity, but not all batteries are going to be the same. Some will be higher capacity, some will have greater life, some will be heavier or lighter, some cars or trucks will need bigger or smaller batteries, etc... Some could even be damaged and unsafe, which is something you don't really get with gas (sure bad gas could ruin your car, but gas from a pump isn't going to catch fire randomly and hurt you because it's bad)

At the end of the day, it's just too much variation for me to trust a random battery pulled from a depot of them.

The thought experiment that put me over the edge was imagine instead of batteries, we were talking about tires. What if it was deemed better to swap out your tire when you had to put air in it. So you brought it to your shop and they pulled the wheel and tire off, then put one that they had in-store on full of air, and they refilled the one you gave them and handed it to someone else.

Would you trust that? How would it change how we handle tires on cars? Would the public go for it? I really don't think they would, I know I wouldn't, and it's the same kind of feeling with car batteries.

Would you take the marginal risk of a battery swap (plus the costs of all those batteries in inventory throughout the country) in order to save 60 minutes on a 600 mile trip?

And is the infrastructure to do all of that really less expensive than putting a charger at enough restaurants for you to just stop and eat while the car charges?

I honestly think I would for a trip like that. But at the same time I wouldn't want to be stuck with a potentially "sub-par" battery after that trip either. And those 2 things seem pretty at odds with one another.

Like in a perfect world I'd love to be able to have it as an option, but I'm not willing to give up "my" battery in my current (hypothetical!) car for one of unknown quality.

Perhaps a "booster" system would work then in this case? You don't swap out the actual battery in the car, but instead swap out batteries that go in your trunk or something and charge your actual battery while driving! Then you don't care about how shitty that pack is, you are only using it to "recharge" on the go!

Another thing to consider, it is much more likely that we will see 400mi range electrics sooner and/or easier than we could roll out a large network of battery swap stations. I think the concept of battery swap is solid. Most batteries have BMS systems that prevent over/under charging and any physical damage can be detected. Plus the battery would likely be owned by a third party, so it would not be a big deal if you got stuck with a bad battery, you would just swap it out for another one. The main issue is logistics. Much harder to build a network of battery swap stations then it is to put some fast chargers in parking lots. And if we can get the range up to 400 miles, then a 600 mile trip could be done with 20-30 minutes of super charging.

[1] Assuming a ~300 mile ICE range, you need to fill up at the start, at the middle, and at the end (although one could wait till the next morning, but I think it should still count). Whereas electric, you start full (because you plugged in overnight prior to departure), and you end full (assuming you can plug in overnight at your destination). These are rough assumptions, but I think 2 fuel stops is a good approximation, but many would need 3 (including my own ICE car, which gets ~250 miles per tank).

This was really the point that I was getting at. It is ultimately cheaper to put good chargers at every restaurant and bathroom stop than it is to put battery swap stations (with inventory) at enough locations.

The major one being your first paragraph. Charging tech is getting better every day, and we are already at the point where 15 minutes can get you a good chunk of charge with current batteries in some cars. I think things will change (no more "gas stations" (or places solely devoted to providing fuel), and a lot more charging spots at destinations that are worth spending 30 minutes at), the bit of charging time every few hours will begin to become a non-issue, and the whole complicated logistics nightmare of managing a fleet of physical battery packs at various depots across the country will go the way of the flying car. That thing that everyone thinks they wanted but it was just never practical to have.

There was a time where gasoline brands mattered, and the gas you got from Esso might have been slightly different than the gas from Sinclair or Hess or whatever, but that was intentionally regulated into irrelevance because the benefit to consumers of a consistent product was more valuable than allowing brands to differentiate themselves.

It took several decades after the widespread adoption of the car (and gasoline as the dominant fuel) for this to happen, and during this time users had to struggle with varying fuels produced from different feedstocks, gumming of engines, dilution of crankcase oil, etc. (This is referred to in period motoring publications as "the Fuel Problem".) The automobile industries and the oil industries basically pointed fingers at each other until the US Bureau of Standards had to step in and mediate on a consistent set of standards that the refiners would produce fuel to, and the engine manufacturers would design their engines against. If this hadn't happened, and the market had been left to sort it out, I suspect you would have ended up having to fill your Ford car only at Sinclair stations, based on some Ford/Sinclair arrangement or whatever.

Unfortunately the 'solution' to the Fuel Problem in the 30s involved the introduction of tetraethyl lead, which mars what would have been a good example of mutually beneficial public-private cooperation and useful regulation.

Sure, I put 87 octane in our SUV and 89 in the audi, but yeah for the most part gas is gas, unless something is wrong.

Getting 2 random batteries to have less than 10% variance among them is going to be a lot harder (especially when battery health changes a lot depending on how they are treated. Users running them to near empty in cold weather is going to shorten the life a lot faster than the guy who only runs it to 50% before swapping out every time), and I have a feeling that the amount of "problem" batteries will be much higher than the amount of times i've ever had "problem gas" (which is basically 0 times so far).

IIRC ours is the average of 2 different methods (it normally shows as R+M/2 on our pumps), whereas I believe the EU uses only the one that tends to be a lot higher (RON I think)

Most gas stations around me have 87, 89, and 91, and sometimes 93. I believe your octane numbers will be on average like 6 higher than ours for the same gas.

For starters, with gas and other things like it, there is a supply chain which is pretty tight. If one supplier starts giving bad gas, you can blacklist them and move on. With batteries you'd have to test them after every single use, because you don't know what the last user has done to them.

Then they need to be physically inspected, because a puncture or deep gouge or crack or something could cause big issues.

Then their actual capacity needs to be tested some how. I'll admit that I've been outside of the battery ecosystem for a while, but years ago it wasn't all that easy to "spot check" a random battery to find it's health and capacity. Most devices would gain that knowledge about a battery over a few charge cycles. You need to make sure you know it off the bat, and you can't be wrong or I'll lose faith in the company and the tech.

Then throw in the complication of different sizes, different shapes, different technologies, different requirements (this isn't just the "gas" part of fuel, it's also the "gas tank", so one size will never fit all), and now you are maintaining huge banks of different kinds of batteries all while charging them to the right amount without degrading them, and checking that users didn't run them in below freezing to almost 0 percent which would reduce the lifetime and aren't damaging them in any way which would be very unsafe.

In an ideal world, it sounds like a great idea, but I just don't trust that it can be pulled off without making the user's life worse than if they didn't do it.

Bringing it back around to the topic, I don't know that we can wait on solid state and this represents a work around while smarter people that I can figure that out.

There's no reason that batteries can't be checked that the meet spec before they are replaced in your car in the same way.

For EV fleets we used to only be able to do this per vehicle (not by battery) while the vehicle was home and charging, by shipping the fleet battery metrics to a central aggregator. But that is overkill for consumer vehicles.

The last time I tried to do that (6 months ago) they rejected it. I had to go to the original supplier, as they were no longer refilling each others branded containers. In spite of standard fittings/sizes/etc. They've become significantly less fungible, and I can no longer just go to the nearest or cheapest supplier based on my needs.

Even worse, I've just realised that I now effectively have two un-refillable tanks, as Homebase just shut down all my local branches. Which I paid deposits on when I bought them, obviously. Nice ...

Automated battery swapping takes tight standardization. It's in use for some industrial robotic AGVs.[1][2][3] Mechanically, it works, but all the vehicles have to be very similar. A taxi fleet, maybe.

[1] https://www.youtube.com/watch?v=lvqbw0UT5pM [2] https://www.youtube.com/watch?v=uKg8Zxmmpv4 [3] https://www.youtube.com/watch?v=vJwtwdeAXhM

I wouldn't. Not to be a naysayer, just giving some context. The cars I buy are already in that price range, and I already pay "comparable gasoline prices" (via gas, hah), so in my area all I'd gain is a different type of gasoline when in city, and when I go camping/etc I'd be at risk due to being in slow adoption areas.

Don't get me wrong, I want electric, but due to the areas I drive I've already settled to likely being stuck with Hybrids for the next X years.

With that said, if the model you laid out still ended up cheaper than gasoline it would at least tip the balance towards its favor. I just can't handle a zero sum trade - and I imagine the same would be true for many around this large, slow to adapt land masses.

Cities will become much, much nicer though. Air quality alone. I look forward to it

Today's batteries aren't really good enough for driving on vacation, but private practice is moving too. The number of 25-year-olds with driving licenses is dropping, I've seen surveys that list the iphone above a BMW as a status symbol, and then there are phenomena like this: https://www.citymetric.com/transport/driving-london-has-been...

Picked up our Model 3 in September. We've already driven it to West Texas, North Carolina, South Carolina, as well as multiple trips inside of Florida. Hit 10,000 miles yesterday.

I would never argue that road tripping in a Tesla is as worry-free as it is in a ICE-powered vehicle, but with a modicum of forethought I'd say the Model 3 is plenty good enough for driving on vacation today. The hardest part is limiting yourself to hotels that offer Level 2 chargers, but pretty much all new construction hotels have EV charging of some sort.

https://www.fastcompany.com/3028159/a-broken-place-better-pl...

https://www.wired.com/2008/08/ff-agassi/

Off road driving

Unless we get a battery that has - at a minimum - the same capabilities as a tank of gas, off roading is going to be relegated to strictly internal combustion, or maybe hybrid powered only.

Right now, even the tech in a Tesla would not get a decent off road rig (up-armored, lifted, accessories, large MT tires, etc) out to the trail, over/thru it, and back to civilization. What we have right now will generally get you over a decent trail - but you'll probably have to trailer the rig in and out of the area.

Even if solid-state batteries prove to be practical, it still might not be enough for any kind of serious off roading capability; that is, anything more than a "day trip", or where unplanned excursions off the route are involved.

Because you won't have any easy ways to recharge the battery while on the trail.

Right now, you can carry your own fuel with you, and refuel as needed or if there's an emergency or whatever. Other people when out with a group may also have fuel others can use - so the concept of universality comes into play.

It's not possible to bring a generator (and fuel) or a solar panel with you to charge the battery of a hypothetical off road vehicle, unless this new battery tech allows for short charge times at lower voltages/currents as well (which probably isn't on the table). And you likely can't bring along spare battery packs - the size and weight would be prohibitive (then again, some people bring along replacement axles); unless the weight and size of such solid-state batteries are less.

Yes, off roading is a small segment of the population, and so maybe it would be ok if that portion remained to use IC engines; presumably a battery swap system, or a solid-state battery pack might be the solution for the majority.

So then, the business challenge becomes to extract more value from a pack than a car owner can. There are many opportunities to do that:

- much improved inventory management, buy in bulk, monitor and adjust charging parameters etc. to prolong battery life based on observed performance of the specific battery type.

- this allows you to optimize battery purchase and chemistry to minimize overall cost even at the risk of a reduced lifetime

- centralized recharge gives you economies of scale to purchase cheap electricity during the night and enter all sorts of contracts where you can earn money by acting as a last-resort supplier at peak times, without actually using significant charge cycles on your inventory.

- the same economy of scale is passed down to customers who won't be bothered to install high power fast charge circuits, or are simply unable to.

- differential pricing based on customer needs, for example, new, top performance packs for home users and degraded packs with lower peak current for less sensitive customers, taxis and commercial fleets etc.

- batteries depreciate even when they sit charged; a modular system will allow customers to keep (and pay rent for) a single or low capacity unit on a regular basis for city use and opt for full capacity when they really need the extra range.

Just the labor cost for the procedure is likely to greatly exceed the $20-40 estimate.

They replaced the battery in 90 seconds and it was completely automated.

I really like the idea of having gas stations replaced with battery stations. Drive in to the "pump", put in your credit card, and let the automated system replace your battery.

The space that was used for tanks could be devoted to charging infrastructure.

I am very skeptical of that claim. This seems to hold true for laptop batteries. Not cars.

I bet in ten years having your battery swapped is about the same as having your belt and pump swapped. A couple hours mechanic fees plus the cost of a new battery minus the resale value of the old one. Done within a day, simple once every 250k kilometers to keep your car run as new.

A model 3 without a battery would still be 30k, and that's with Tesla on razor margins and with the much cheaper supercharger infrastructure.

All that to say there is a lot to be gained by not doing a battery swap system.

It could be that a quick hot swap is technologically complicated, especially given Tesla manufacturing tolerances. Also if the hot swap fails, you're stranded without a car.