Why not swappable batteries?
Posted by reidpants@reddit | cars | View on Reddit | 70 comments
With the unveiling of the Robotaxi's wireless charging, it made me think that a (maybe subscription-based) swappable battery system would be a model worth exploring for EV manufacturers. It would address the issue of battery degradation and replacement costs. It would also provide new avenues for emergency responders to contain fires. Am I just dumb to think this model isn't possible?
stav_and_nick@reddit
Because it's a massive pain in the ass
There's only one company that has done it at scale, Nio, and let me tell you about the difficulties they're currently having;
So, for one, they're not profitable. They've been in the space for a long time and yet unlike even newer entrants like Li Auto, they just can't stop being a money put
Secondly; they just released a new subbrand which is attempting to be like... the Volkswagen to their Audi. This new car is causing a massive issue because it has a slightly different sized battery. They designed it! It's their own battery in their own swapping stations, and they're having problems with it!
So if that's the issue that one brand has, imagine how difficult it'll be with 10 brands with multiple models and potentially many differences in model year
Mediocre-Cat-Food@reddit
People never consider what a huge hurdle swapping the cooling is as well. You either need to swap everything with the battery, or add quick disconnects that add another unnecessary point of failure.
Additionally, I can’t see a battery swap being faster than DC fast charging. If everything was on 120v and 240v sure, but DC fast charging has gotten pretty good
dont-YOLO-ragequit@reddit
On the other hand, if batteries are meant to be swapped, there shouldn't be a need for cooling.
Wether it's short term or longer term, the batterie's life expectancy becomes a bigger cost for the battery swapping company that can charge more for the service.
What I mean is if we are thinking about it, a battery can be swapped and charged at the perfect charge rate and tested at charge station where they can be discarded and recycled if they fail tests. Why go through the hassle of battery heat management if the Nissan Leaf can do without and last at least 4 years instead of being cooled and lasting 8 years? Keeping the batteries as good as posible with heat conducting materials and recycling them more often could be a better play if swaping and bleeding coolant is that big of an issue.
JawKeepsLawking@reddit
Its not just for longevity. Your battery isnt gonna shrug off 110 degree weather with the ac on going 90 mph.
SirLoremIpsum@reddit
Because even if you are swapping batteries you want them to last a long time no...?
dont-YOLO-ragequit@reddit
What I'm trying to say is if Nio's plan is to make cars you don't have to charge and just need to swap batteries with, they could benefit from selling the car but renting the battery. The problem is the battery tech, charging and management is changing too fast to make a long term solution.
It could be a better idea to say, have a battery pack outsourced and a compact sedan, a crossover and a compact Hatch/wagon, have uncooled lithium ion batteries that they control the charge(since they are the only ones charging it) and then later on, be able to switch to lithium phosphate batteries and faster charging when it is affordable( which would be in 5-8 years anyway) or move to denser lithium phosphate with enough spare space to integrate a small but complete heat pump.in the same module.
This way, as long as the price of upgrading batteries can be spread with each swap, they could keep making less complex batteries if it matches the expected remodel and latest battery tech.
Like maybe it would be a bit more expensive but it might be convenient to have a car with 200mile of range that later gets upgraded to a 300 mile one to stay competitive among new cars but it also means older cars get a similar 300(minus the wear on their powertrain) miles.
visceralintricacy@reddit
Because the leaf sucked SO much ass for it, and when the companies already a money pit, replacing batteries twice as often is going to completely kill them. Nobody wants to pay for that.
dont-YOLO-ragequit@reddit
The leaf was also very small and IIRC with battery tech that is older.
I'm mostly brainstorming here but I see 2 different usages of this battery and this would determine how cooling really matters.
If cars are meant to almost not be charged outside of the charging stations( or charged just as a backup/range extender) and the goal is to "not have to charge it but always swap it in under 30 secs like the videos have shown(Basically like swapping propane tanks). Longevity can be managed by having a fleet of batteries to be charged, tested, swapped and upgraded. (Basically someone buys the car and a core on a battery and a battery company makes sure all batteries with under 70% health are recycled or upgraded with new materials. In this sense, frequent to daily swaps would have their own short term fees and longevity on a commuter/ city cars matters less.
On the other hand, if the goal is to have the car and battery last years (like a current car battery) and just have the battery be swapped on long road trips, then it's different and it will need more expensive batteries with cooling.
stav_and_nick@reddit
I believe Nio does the quick disconnect thing. Which has lead to several videos of poorly secured batteries falling out as a car drives away. Not particularly good PR given you're supposed to be a luxury brand!
hi_im_bored13@reddit
To add onto this, at one point tesla did have a battery-swap station, and eventually closed it down in favor of improved supercharging.
stav_and_nick@reddit
Yeah; China is now start to see mass rollouts of 500 kw chargers, plus batteries that can handle them. 0-80% in less than 10 minutes. Which btw means it's FASTER than swapping the entire battery. Iirc the Nio swap station takes around 10 minutes.
It's in upmarket stuff now, but when you look at where we were 10 years ago with EVs, and extrapolate to 10 years in the future... I just don't see a place where the math works out
MAYBE in big rigs, where you'd start to need 1 MW charging if you wanted super duper 0-80 in \~10 minutes like passanger cars. But even that's dubious imo, if only because public safety regulations would probably mean it's just more efficient to use 500 kw chargers while truckers rest
burlyginger@reddit
Faster, but has to come with the side effect of degrading the battery?
That's an insane rate.
stav_and_nick@reddit
Allegedly, no. It's just able to tolerate significantly higher heat while charging compared to previous generations. Similar to how 4C rated batteries had less degradation and better charging compared to 3C, and apparently 6C batteries are in the works already to replace 5C
Seems like so far we're still in the "no downsides" area that computer chips used to be in. Each generation gets more performance for cheaper with less degradation/power useage
Now of course, these are still basically brand new. Maybe it'll die in 5 years? But so far CATL batteries have worked to spec, so who knows
burlyginger@reddit
That's fucking wild!
stav_and_nick@reddit
Look up the Li Auto Mega if you want to be amazed. Someone got a video of one peaking at 552 kws (!) while it was charging
burlyginger@reddit
That's over 2000A at 240V.
I'd be surprised if the cables don't get significantly magnetic at those levels 🤣
JtheNinja@reddit
DC fast chargers are higher voltages than 240. It’s usually industrial style input from the grid (480V 3-phase in the US, idk what China uses), then DC output to the vehicle at the battery pack voltage, typically 350-900V depending on the car’s design.
stav_and_nick@reddit
Yeah, call me a coward, but I think i'll still to max 350 kw stations, at least for a few years. I don't fuck with bigly electricity
Allegedly it's a 200-1000V system, 600 amps, max theoretical output of \~600 kw on the newest and best one by huawei.
lee1026@reddit
The state of CA offered them money if they do battery swapping. So they opened literally a single station open for not a lot of hours a week. This fit with the literal text of the requirement, but they didn’t go beyond that.
reidpants@reddit (OP)
While I agree with this take as a serious hurdle, it is still likened to Tesla's proprietary supercharger network. I would even argue that going the battery replacement route would push Tesla to standardize their batteries, which would save costs from an R&D perspective. Given that even Tesla struggles to make money strictly off of their EVs while every other manufacturer is losing money, this would be a viable option.
Your point about brands is a good one, too. If I were in Tesla's, GM's, Ford's, etc. shoes, I'd worry that what happened with the lightning cable and the EU would happen with EV's, and there'd be an eventual regulatory push to standardize batteries, connections, etc.
biggsteve81@reddit
Tesla has gone a different direction. On the Model Y the seats are attached directly to the battery as a unit and then the battery is shoved under the car. So if you pull the battery out of a Model Y the seats come with it.
They did this to save cost on assembling the vehicle, since it is easier to fit seats to a battery than bolt them down inside of a car.
activedusk@reddit
Battery swapping already exists for electric cars, NIO offers it in China and it is automated and relatively quick compared to charging.
Why is it not a widespread solution across the industry and not pushed forward by more EV makers including Tesla despite the technogy for it existing already is the question.
The main reason I would guess is the need for higher investment for the battery stations, they need to have more battery packs than there are cars sold and those stations still need to charge those extra packs as they are being exchanged so you still need to set it up as a charging station but with extra steps and more money.
Second problem has to do with platforms for automotive use. I m sure you heard various car makers name their latest platform that allows them to make all sorts of vehicles on those platforms and cut costs by not having to design a new one for every vehicle model. So what problem does it create for EVs then? A small vehicle will have a smaller capacity battery than a larger one, you can guess the dimensions and weight problems in sharing the same battery pack for all models. Then there is the platform and compatibility between the connectors for power and thermal management or even dimensions and fastners. For how many platform generations will you keep backwards compatibility? As you develop more cars you find problems you were not aware of in the past and make new revisions but the need to maintain legacy support will block or impede implementing new and batter pack designs forcing you towards lagging behind others. Think stuff like pack voltage for example.
Last but not least is how people take care of their cars, battery health is not universal, why would you agree to let go a pack with little degradation you treated well for months if not over a year to get one that was trashed by the previous user? Nobody ought to make this kind of exchange in their right mind and if the manufacturer has to take these losses and take those packs out of circulation sooner, their operating costs increase as well.
ShadowtheHedgehog_@reddit
We just install a shit-load of D-cell batteries. Like with an 80s boombox, but bigger.
cheeseshcripes@reddit
I actually don't hate this idea, you get a lighter car with less range if you don't feel like you need all of them. And you can charge and drive at the same time.
reidpants@reddit (OP)
I actually think we could just daisy-chain 9V batteries instead.
Buckus93@reddit
Wait till you see what's inside a 9V batter...
reidpants@reddit (OP)
… more batteries??
Buckus93@reddit
It's batteries all the way down...
But seriously, it's six roughly AAAA-size batteries wired together.
Buckus93@reddit
It's time for the Mercedes AA-Class.
https://youtu.be/0k1tbf8muMc?feature=shared
reidpants@reddit (OP)
Finally, someone that GETS IT!
Buckus93@reddit
No one has mentioned this at the top level, but a big reason is that the batteries are typically an integral piece of the car. In the case of the Tesla Model Y, the battery pack is literally the floor of the vehicle (as opposed to being under the floor).
In order accommodate a swappable battery, vehicles would have to be structurally sound without the battery installed, which means more structure around a battery compartment, which means more material and more weight.
Plus all the reasons others have replied with.
WeAreAllFooked@reddit
All EV manufacturers would have to agree on a standardized battery to make this even remotely possible, which won't work for a litany of reasons.
Swapping batteries would have to be profitable, which will never happen if you take Nio as the case example. There's also the issue of what happens, and who is responsible, when a battery pack gets damaged. Are they going to foot the bill to replace it (doubtful), or are they going to try and lay blame on the previous user(s) of that dead battery? What happens if a battery dies? Are they going to offer roadside assistance when a cell pops in one of these swappable batteries? What happens when you go to swap a battery in before a road trip and you find out it's old and only has 70% of it's rated capacity available?
At the end of the day it's a lovely idea that turns in to Swiss cheese when you start looking for the holes in the reasoning.
reidpants@reddit (OP)
I'm thinking of this more from a perspective of one manufacturer decides to do this, like a Tesla or GM. Battery swapping gets to a point where it's fully automated and only human element is an inspector, sub 3 minutes. Batteries with under 80% life are at the end of their life-cycle, and recycled for the manufacturing of new batteries. Damaged battery back protection would be based on tier of subscription, like accidental damage protection warranties for laptops, cell phones, etc.
lowstrife@reddit
Those are all great concepts and would need to be part of a system, but how you actually make that actually fair to everyone in the real world isn't really possible. This is impossible to achieve for vehicles and batteries owned by the end user, you could never financially compensate someone $3000 for a shittier battery when they paid you $33 to charge it. The economics just don't make sense.
reidpants@reddit (OP)
No one would own their own battery in this model. They'd be able to hold onto it for as long as they needed, and part of the Master Services Agreement / Terms and Conditions would be an agreement that the range now becomes with a guaranteed minimum (80% battery range).
xarune@reddit
You have two scenarios for needing a swap.
Long distance travel or apartment dwellers where charge time matters. For those people DC fast charging is quickly closing the gap well enough it's north worth the investment. Electricity is electricity and is everywhere; a battery pack swap center requires infrastructure, logistics, employees, real estate, inventory, etc. Any subscription service would be so stupid expensive to cover logistics overhead, service, and depreciation it has zero chance of competing with higher speed DC charging.
And then worn out batteries. Outside of Gen 1 Leafs which were passively cooled and had bad temperature management software, battery degradation really hasn't shown to be a major problem. Yes there are some lemons from the factory that get a lot of attention, but otherwise batteries are showing the typical bathtub curve and them losing capacity is not really a major issue. I would wager that going forward, you will rarely see EV batteries replaced before the 60% degradation mark in general: a 300mi original battery with 200mi now still has a lot of value as a used car and it's easier to sell the whole car than tearing it up to recycle before then. Batteries going bad has been a fear that has yet to really materialize in a significant way.
OEMs don't want to be responsible for owning that much equipment, particularly stuff they are "loaning" out to individuals to use and abuse. They offload their cars to the dealers nearly immediately in terms of ownership. They also don't want a bunch of batteries with 80% capacity that still have life left they have to offload at a deep discount.
Your only real path here is commercial vehicles that already enjoy major dedicated infrastructure, have more modular vehicle frames, and are tools to make money and downtime matters: so basically OTR trucking on fixed routes.
WeAreAllFooked@reddit
Tesla went the other direction and bolstered their charging infrastructure and tech because swapping batteries out isn't worth the extra headaches and engineering challenges. If Tesla couldn't make swappable batteries work with their EV market share I highly doubt someone like GM would even consider this, especially if the swap station only deals with one specific manufacturer.
reidpants@reddit (OP)
See, I just watched a video of their van, and have looked into their nightmare of a process for mass-manufacturing the cybertruck, so it's just hard to imagine that even they would shy away from swappable batteries. The only logical reason I can come to as to why they wouldn't do it is that they'd end up with a 2001 Toyota Camry crisis where no one wanted to upgrade their cars and as such, forcing people to upgrade every 5 years becomes the much more profitable model.
WeAreAllFooked@reddit
They don't make it removable so they can use it as part of the chassis to save weight, manufacturing time, and engineering costs. It's hard to make the battery removable when the entire chassis is using it a structural component.
NFA_Cessna_LS3@reddit
One of many reasons is.....people don't want someone else's junk. Lets no kid ourselves li-ion batteries should be treated a certain way and only last so long. If I take care of my battery why would I want it swapped with another from a ghetto blaster?
How do you handle warranty issues "I took care of my battery its not my fault I received a trashed unit and it croaked"
Infrastructure where you can get the batteries swapped
All of these issues can probably be fixed without much effort but li-ion batteries; imo are a transitional holdover until solid state batteries come around and solve range anxiety, weight concerns and charge times. Effort vs reward just isn't there.
t-poke@reddit
This is exactly it.
One might say "Oh, so just swap it again and you might get a better one" and that may be true, but I don't want to find out I got a trashed battery in the middle of a road trip when there's 75 miles of range left and 100 miles to the next swap station.
I'm not playing Russian Roulette with EV batteries.
reidpants@reddit (OP)
I like this take but since it's me and D-cell battery guy in the pro corner, doing TL;DR replies:
No warranty issues, manufacturer owns the battery and regional cost of battery QA/QC is baked into subscription costs.
Infrastructure would require 5 to 7 yr R&D, dev. and implementation. Could develop a partnership with Gas Stations like Flying J, Love's, etc. to build out automated swapping mechanism. I figured it would be something similar to a car wash with a human attendant on hand for final inspection.
NFA_Cessna_LS3@reddit
so I need to buy the car and then pay a subscription???
so they want to sell a product, bill me with a subscription then promise me it'll get better down the road because they have a dream? do you understand what it'd take in legal, financial, infrastructure and logistical resources? all for what, so people can dump li-ion batteries like a stage 4 clinger girlfriend when solid state batteries come around?
tonytwocans@reddit
Think about how expensive a BEV's battery is. Now double that price for the extra battery to be swapped. It would increase the cost of each car by 50%.
zzzzbear@reddit
it's existed for some years and does not increase the price
https://youtu.be/hNZy603as5w
tonytwocans@reddit
Can I get a tldw
zzzzbear@reddit
Nio does it in China at dedicated swap stations, it works fine
people will point out Nio's other flaws as a business but this is a very intriguing model they have going
t-poke@reddit
Another video had some guy timing it, and it took 6 minutes to swap the battery. Great. That's faster than DC fast charging.
But here's the problem with that though. That swap station can do one car at a time. If it's busy and there are 5 cars ahead of me, I've got to wait 30 minutes until it's my turn.
In that same building footprint, you could have 6 DC fast chargers and all 6 of us could charge at the same time and be on our way in 20 minutes.
And I can take a piss, stretch my legs and grab a snack while my car is charging. Looks like you have to be in your car during the swap, so now after that's done I have to go park my car elsewhere and take care of that other stuff.
Some of the largest Supercharging stations have dozens of chargers, and plans for even bigger ones are in the works. I don't see how you can scale up battery swapping to match that kind of capacity. It wouldn't be practical or cost effective.
zzzzbear@reddit
its surprisingly analog and could be automated around a standard easily
I doubt it becomes default but it's good to see companies testing infrastructure options
reidpants@reddit (OP)
I think that you'd drop the initial cost of the car substantially, but the manufacturer would then make their money back substantially with subscription models and depending on usage, it would indeed be more expensive in the long run to stay with the same car.
Head_Crash@reddit
Battery swapping would work for commercial vehicles but it's difficult for passenger cars because EV and battery technology are quickly evolving and hyper-competitive. Limiting vehicles to a universal battery type creates more problems than it solves, especially when charging speeds are already sufficient for long trips.
simon2517@reddit
Look up Nio, they've done it. So it is possible.
There are definitely problems though. One of the biggest is that it's really hard to imagine a battery swap system that's compatible with you owning the car's battery. So Nio customers who want to swap need to lease their batteries even if they own the car.
Other problems include having to design all your vehicles around a single battery size, and the fact that it's inherently more expensive to build battery swap stations than fast chargers (since battery swap stations have to include fast chargers to prevent dead batteries piling up...).
Overall I don't really see it as a concept with much of a future.
JtheNinja@reddit
This is really the main catch. This seems like a great idea in abstract where you assume the station magically has fresh batteries for everyone who pulls up. But to actually make this happen in the real world, you need to charge the batteries just as fast as a conventional DC fast charging station. If you can swap a pack in less time than it takes to fast charge it, the station will eventually run out of charged packs unless each swapper can recharge several packs at once. But if you can do that, you can also just build a DC fast charging station with more stalls!
I’m actually really curious how Nio handles the situation of a user arriving at the swap bay, but it doesn’t have any batteries that are charged higher than the one in the user’s car. Do you just wait until it finishes charging one of its spares?
Quatro_Leches@reddit
They’re heavy as hell
MechMeister@reddit
Imagine how many times your USB cable comes loose for charging your phone so you have to buy a new cable. Now imagine that it's 480v, the size of your fist, and makes your car work lol. You cant unplug a battery every 200 miles and have it last.
TSLAog@reddit
I’ve been in EV repair for 14 years (early Tesla tech, now at Lucid)
I even helped at the only Tesla swap station at Harris Ranch for a week, it was a train wreck.
There are a TON of technical, software, homogenization, and physical reasons it’s not done… I could literally write a 15 page document on why it’s a bad idea.
Shienvien@reddit
Battery degradation is actually a big reason why swapping station would be even less viable than you'd think - if I live in a less densely populated area, and the next stop is in 250km. Brand new battery for my car would be 320km range going 90kph/55mph. How do you guarantee I'll get the minimum 80% life left battery I need, not a 75% life left one? Do you only keep nearly brand new batteries in your station because you don't know if the next person will need a new battery or old ones would be OK?
DrSpaceman575@reddit
Because they'd be swapped right out of my parking spot in my neighborhood
ItsAStillMe@reddit
No more subscription anything. There is a value in actually owning things and not just renting them.
ElChaz@reddit
There's also a value in renting. For example, with battery technology improving so rapidly, buying the car but leasing the battery pack and getting a new one that doubles your range every 5 years sounds pretty cool.
It's too simple to just say "owning > renting" and leave it at that.
ItsAStillMe@reddit
No. Owning is always better than renting/subscription. Once you physically own it, it cannot be taken away. When you just rent or subscribe to something it can be completely taken from you or disappear when money is stopped being paid or the actual owner decides you can't have access to it anymore.
reidpants@reddit (OP)
I agree here, this is more me trying to understand the why no one has gone this route. I don't own an EV and my personal opinion is that mild-hybrids with low-emission renewable fuel (like what VW is developing). I'm wanting to understand why this as a business model failed so quickly.
ItsAStillMe@reddit
Battery technology in any form isn't good enough for it. At all. EVs are a "solution" to a problem that doesn't exist. Yes we need to take care of the planet, but people driving ice cars doesn't contribute to the things as much as the powers that be want to claim.
cirebeye@reddit
I like where your head is at, but the way cars are now, the batteries would be very difficult to get to. In order to make then accessible while still safe, you'd probably add a good amount of weight at the sacrifice of miles per charge and/or lose cargo space.
Not only that, but these batteries are HEAVY! It wouldn't be a job you do in the driveway. You'd have to take a car to a service shop to perform the work.
And then there's the disposal fee that would be attached to this. Batteries are quite toxic, and in this size, there'd be a hefty fee that came along with it.
Best course of action - lease an EV. There are some extraordinary deals on EV leases right now, and you'll be out of the lease before there are issues with the battery holding a charge. Not only does this work out better for yoh, but it gets us closer to resolving the issue you're bringing up. When dealerships in five years are overburdened with used EVs off of leases that now hold less than 200 miles a charge, they'll want somethinf done. It's only at this point, when it has a negative financial impact on the dealerships, that they will lobby for better technology and serviceability.
Simon_787@reddit
Modern EV batteries have long cycle lives anyway, so degradation isn't really a strong argument anymore.
MuKaN7@reddit
I'll take a bite.
There are several cons to administering a swappable subscription service, so I'll only hit a few off the top of my head. Demand (profitability) and design (ease of the swap).
Design: space is limited and cars are heavy. It takes a lot of energy to move them, so the cars are designed around the battery. So your choices are 1 big unit (I know, misnomer) that has to be swapped or individual modular units. 1 big unit would require a forklift and a modular unit would require valuable real estate to support it, so you'd effectively have less range. Also, since they are heavy as sin, they need to be fully supported by the unibody/skateboard frame. So it won't be as simple as removing 4 bolts without a lot of machinery. Either way, grandma ain't swapping those batteries.
Demand: demand is constrained by design and the practicality. Outside of E-bikes and e motorcycles, most platforms aren't designed for easy swaps. So unlike switching out the batteries in my TV remote, It'll likely take 1 or more techs over 30 minutes (realistically hours) to do the swap. Way more than the time it takes to charge at a rest stop.
I can 100% see a standardized e-motorcycle battery/battery mount becoming a thing if the right things align. I don't forsee swappable batteries without a lot of R&D or technology changes. There isn't really enough juice worth the squeeze when charging infrastructure gives you better performance timewise. The best case I can think of is manufacturers adopting a booster pack/second fuel tank system. But like with gas, it requires precious real estate, hence why a lot of them get mounted in truckbeds. And in order to make the juice worth the squeeze, you have to carry a lot of it.
chlronald@reddit
I am waiting for a day to have a swappable and modular battery design, where I can slot in one battery for short commute drive and put more in for longer range / higher endurance.
Buckus93@reddit
Maybe when batteries are the size of a small briefcase, but otherwise it's not happening anytime soon.
Snazzy21@reddit
The problem is that not every vehicle is the same shape, if you want to maximize battery size you'd wouldn't want to be limited to however of the standard sizes you can fit.
The tradeoff of an easier battery replacement every 10+ years isn't worth it. Cars aren't made for the 2nd or 3rd owner anymore, so why would a first owner car what happens after they sell it. It would add cost.
This would be like using wet liners in a passenger vehicle.
ajrf92@reddit
Short answer: Time.