Just a few weeks ago, there was a little flurry of stories about how a Department of Energy official proposed a competition for American carmakers to come up with an EV that would sell for just $16,000. Keep in mind that the current cheapest EV in America right now, the 149-mile-range Nissan Leaf, costs nearly twice that amount at just under $30,000. Selling an EV for $16,000 is a big challenge, and will take some really innovative and out-of-the-box thinking to pull off.
The good news is that I’ve barely even seen the inside of that box, and neither has our own Bishop. And between the two of us, I think we have some pretty good ideas for making an incredibly affordable yet still useful and desirable EV. I don’t think this competition has officially begun or even if it will be A Thing at all, but that didn’t stop me from making a bunch of inarticulate demands of the Bishop and him processing those into some wonderful drawings of what I think are pretty damn good design ideas. So, with that in mind, let’s take a look at what sort of cheap EV the Bishop (let’s be honest, mostly Bishop) and I came up with.
I think there’s a crucial rule behind developing a successful inexpensive car, and that rule is honesty. All of the inexpensive, mass-market “people’s cars” that have become icons had this in common. Cars like the Volkswagen Beetle or Citroën 2CV or Renaut 4 or Fiat 500 never tried to be something they weren’t. They all had ideas that made sense for their own particular goals, even if those ideas and solutions may seem peculiar when compared to more mainstream cars. But they worked.
There’s a reason nobody is going to be collecting Nissan Versas in the future, but people still will be collecting vintage Minis, and that’s because the Versa was never comfortable with what it was, and always strove to be seen as a somehow-shrunken premium car that it very much wasn’t. The cheap cars that somehow captured our hearts and minds, the ones we all know and love, they never tried to be anything other than what they were, no matter how weird and cheap that was.
So, in designing our $16,000 EV, that is the philosophy I want to keep in mind. We don’t want to make something that’s trying to be an ensmallenated Tesla Model S or Lucid Air; whatever we make will openly and proudly have affordability as a key core principle. And, to go a little further, I’d like this car to be affordable as you own it, too! I want a car that’s forgiving of human failings and circumstance and chance and fate! If something bad happens, I want this car to work with you, and be cheap and easy to repair.
Modern cars are nightmarishly expensive to repair, for both mechanical problems and fixing damage from a wreck or other outside circumstances. This car will understand that, and try to be better. This car will also be easy to upgrade, on a timeline you as owner choose. It’ll be as modular as possible, with standardized parts that can be easily replaced and upgraded. It’ll be hackable and customizable and a rich aftermarket to support the car should be allowed to form and thrive; this is an idea I’ve dreamt of for quite a while, and I think now is the ideal time for it to happen.
With these ideas in mind, I made this rough sketch to explain what I was thinking to the Bishop:
There are a few key features to note here:
- All lighting equipment is cheap, off-the-shelf stuff. The taillights, indicators, marker lights, and sealed-beam headlights are all out of parts bins. If you break one, you can order one online for dirt cheap or likely find a replacement at an auto parts store nearby. No more headlights that cost thousands of dollars to replace.
- The bumpers are big, unpainted things designed to take whacks and shrug them off.
- The need for unique body panels has been reduced as much as possible. The front and rear door stampings are identical parts, just reversed. The same goes for front and rear clips.
If the reversible panels idea sounds familiar, that’s because I cribbed it from those old masters of penny-pinching, AMC:
The AMC Cavalier concept car was all about reducing the number of stampings and dies; AMC only really implemented this in one stamping for front and rear bumpers on some of their cars, and the Cavalier eventually grew into the car that would be the Hornet, but the clever ideas in the original concept are certainly worth stealing.
From that crude sketch and a lot of back-and-forths with the Bishop, the car that I just now decided I want to call the Modulo took shape:
Here’s a view from the rear quarter:
Sure, it looks a little funny, but I think for a car like this, that’s a good thing! All truly successful cheap, every-person’s cars have had a quirky look about them. Let’s get some callouts on here so you can see all the crucial features that a car like this I think needs to have:
Aside from the cheap, off-the-shelf lights, we have ports for sensors and cameras that can be added whenever the owner decides they want, say, dynamic cruise control or some sort of Level 2 driving assistance. You can buy the car without those features, then add them in later when you’re ready. That concept will carry throughout the car.
Here’s what the Bishop has to say about the panel-reuse concept and overall design theory:
“Here’s what you need to know about manufacturing: the tooling needed to make things like stamping out metal and molding plastic is expensive. Millions of dollars need to be spent up front before you’ll see any parts. At the same time, it goes without saying that making 50,000 parts will likely give you a cheaper per-piece price than making 25,000 pieces; you’ll pay for that tool much faster.
What that means is the fewer parts, and the more parts we can reuse, will make a product cheaper. The key to the Autopian car design is symmetry. Most cars are symmetrical left to right, but few are mirrored front-to-back; ours will do just that. Starting with the unibody frame underneath, we’ll use gigapress-formed parts where hopefully we could get the left and right to be exactly the same pieces welded together. All frames would be painted the same white or grey regardless of the colors of the other body panels we add to keep costs down. The fenders and hood that bolt on can, of course, be in different colors, preferably bright and fun ones.
There’s been talk about cars like the Rivian requiring expensive body repairs for even minor shunts. The Autopian car’s visible parts are all easily replaceable; just unbolt them and ad the new one in minutes.
The front and rear doors are mirror images of each other, so left front would be the same pressing as the right rear. Lastly, the bumpers are black plastic pieces that are used on both the front and rear of the car, with apertures that can accept different lights. That was a key requirement from Jason; he’s sick of hearing about $1000 lamp assemblies on cars and wanted the Autopian car to have lights you could buy almost anywhere inexpensively. Headlamps are old-school sealed beams, and taillamps are the “orange stripe” stock ones made by Hella and others (a reflector ring fills the space in the aperture).”
We have a main unibody frame that’s always one basic color, and the body panels provide the actual color of the car. And those panels are kept to a minimum! Same doors front and rear, same unpainted, rugged front and rear fascia panels, same fender/wheel arch panels front and rear. Here, look at the basic components:
The motor and gearbox assembly will be mounted on the rear axle, close to other crucial drivetrain and charging components, which will be effectively under the rear trunk floor, and accessible via a removable floor panel. Up front, we have a mirror of this compartment but as a frunk, and below that floor will be more power and electronic hardware, along with the 12V battery.
I especially like how the doors turned out; there are just two stampings, and both doors open at the B-pillar, meaning the rear doors are suicide-type doors. The windows open in circular arcs, and operate manually with simple catch-and-slide mechanics – unless an optional power window module is installed, which can be added on a door-by-door basis. Speakers can be added to any door panel, along with anything else the aftermarket decides to come up with.
The rear “wagonback” cover can be removed easily as well, transforming the rear cargo area into an open bed, like a small truck. Here’s more from the Bishop:
“Naturally, the Autopian car isn’t exactly the same front and back; the front would have a typical hood and windshield while the rear is covered by a “wagon back” style hatch. My guess is that we could offer a standard “trunk lid” rear or a “fastback” style roof for those that prefer something different. The roof is a wide open hole; a collapsible, removable canvas top would be standard. You could upgrade to a multi-piece rigid unit or even a solar panel one to charge batteries and power your bluetooth speaker at a picnic or beach. Speaking of beach, after removing the roof, you could even unbolt the rear “wagon back” hatch for a full open style similar to what you could do with a 1986 Nissan Pulsar NX (and like that car a canvas cover could be offered in case it starts to rain while the “wagon back” is back at home).”
Another thing I demanded of the Autopian Modulo is that it should have removable, swappable batteries. I’ve thought that EV batteries should be easy to remove and swap out as an alternative to charging or for maintenance purposes for years. The EV industry has not agreed with me, and has been integrating battery packs ever more tightly into the structure of its electric cars. This has some engineering advantages, but it also means that replacing a bad battery pack is extremely costly major surgery.
While EV battery life is generally better than was expected in the early modern EV era, it’s still not a question of if you’ll eventually have to replace an EV battery, it’s when. I can’t abide the idea that a car should have such a financially crippling time bomb embedded in it, so the Autopian Modulo will have modular, standardized battery packs.
There will be slots for six batteries, three per side, and they slide out from under the rocker panels. The car can run with as few as one pack if needed and the base model will likely be sold with two by default, but buyers can add more if desired. I would hope each module is capable of delivering at least, say, 40 miles, for a total range of 240 miles with a full six-pack on board. Even if the individual batteries only get 30 miles per pack, the total would still be 180 miles, which is still decent enough for an entry-level EV like this.
While you can use conventional EV chargers and fast chargers to charge the Modulo, there’s also the option of battery swapping. They’re already doing something much like what we’re imagining with scooters in China:
We’re imagining something similar for the Modulo, ideally done in an open-source way, so any number of companies could operate Modulo-compatible battery swap stations that meet dimensional, voltage, and connector standards. A typical station may look like this:
They would be sort of like propane tank stations are at countless gas stations across the country: you’d pay for the “rights” to however many batteries you want, and you just swap them out when they’re depleted, and back into the system they go. When they get too degraded, they’re pulled from circulation and go on to become wind farm load levelers or some similar use. This way, you may own, say, four “batteries” but you never actually have to physically own a battery that’s constantly and slowly degrading.
Plus, you can upgrade your range as needed. If your daily life works just fine with three batteries, then that’s all the battery rights you need to own, except for the three times a year you take big road trips, and in those times you temporarily pay for access to three more batteries. It’s flexible and easy and you’re never going to be dragging around battery weight you don’t need and never will be stuck having to shell out $20,000 for a whole new battery pack.
The inside of the Modulo is based on the same principles of modularity, ease of use and repair, and ready upgradability. The HVAC unit is optional and mounted in the center console; this can be a basic unit with just a heater or a full HVAC unit with A/C. The HVAC distribution pipes form the structural basis of the whole dashboard, with other components mounted to them.
The instrument cluster and center-stack infotainment displays and touchscreens are both cheap, off-the-shelf Android (or iOS, why not?) tablets running an app for their specific functions. They connect to the car for power and data with standardized connectors and can be replaced or upgraded at will. They’re not necessarily automotive grade, but if they fail after a few years, they’re cheap to replace. Plus, as technology improves, better tablets capable of running updated infotainment or instrument cluster software can be used. Systems like Apple CarPlay or Android Auto can be used in lieu of the center stack infotainment tablet if desired, too.
Like the rest of the car, the modular interior would be designed so that aftermarket options would be available, so if you wanted, for example, nicer leather seats, you could swap out the more basic seats the car came with for fanicer ones whenever you wanted.
The seats, like everything else the owner has access to on the car, would be designed to be easy to remove and install with only basic skills and tools. This would go for as many elements of the car as possible. Flooring may start as rubber matting, but that could be removed and replaced with carpet, using the same basic fasteners. The same goes for interior door panels or headliners or whatever.
This is the way we can get to a $16,000 useful, desirable EV. Maybe that $16,000 edition is the most basic one, with rubber flooring and manual windows and just two battery packs and cheap tablets and no extra cameras and sensors and so on. But what makes the Modulo different from most cars is that you can buy the $16,000 basic one and over the years, as your fortunes improve, you can easily add and modify your car until you’ve doubled the range, got better seats, full carpet, A/C, adaptive cruise, whatever. And you did it on your pace, and installed every upgrade on your own, with ease.
If you get in a small wreck, it’s not a devastating thing; it’s a little bit of money and a weekend’s worth of time to make things right again. As the car ages, you’re not stuck with a degrading battery and vanishing range; you just go to your nearest battery swap station and get yourself some new modules.
I think making a good, cheap EV needs to be a pretty radical departure from the sorts of EVs we see now, and needs to be something that looks at the entirety of the ownership experience, not just something that looks sleek and crows about range and 0-60 times. It needs to be something that’s not just affordable at the point of sale but stays affordable throughout its long and devoted life with you.
It’s time for the Autopian Modulo. Are you listening, Advanced Research Projects Agency – Energy (ARPA-E)? Go ahead and officially issue that $16,000 challenge! We’re ready! Well, unless we have to actually build these, in which case, um, we don’t have those resources. But we’re open to talk!
Modularity for anything more complicated than the bumper makes it more expensive, not cheaper. It has to be designed with accessible bolts instead of being welded or riveted. Anything that requires wiring needs the wiring installed even if you didn’t get the option.
CANbus All the Things, and the wiring becomes easier to deal with; a twisted pair plus a power lead going around the part, and if the option was selected, it’s plugged into the bus, and if not, a terminating resistor is plugged into the bus instead. Which doesn’t invalidate your argument, but might mitigate it somewhat.
This is so awesome and smart that it will never happen because there is no justice in the universe.
That said, one more detail I’d love to see is a service chase for all the wiring. Instead of a spider web of harnesses that are arbitrarily jammed in, run all the interior electrical in a single conduit, maybe along a rocker or in a flat spine along the roof. Use the same gauge wire, same connectors everywhere and make the door wiring dead simple and easily detached.
Also use the same fasteners averywhere. No more Torx + Allen + Philips x 4 different diameters
Man, I LOVE this idea. Imagine a young adult with their first decent job, just getting started in life, and this car could be a great first new vehicle and fully adapt to the life changes over the next 10 – 15 years. This will reliably get you to and from all the work and errands and kid activities you need for a long time.
As a high school kid I would have loved to be able to customize my ride to this level. I used to pour over that JC Witless catalogue for hours just imagining and to have a healthy aftermarket for all the bling and practical upgrades would be amazing.
The swappable/rechargeable batteries are a good idea too. I don’t want a car that will force me to swap only, but to run the batteries for years and easily replace them solves the problem we have with our ’17 Leaf – mechanically awesome car, range down to about 70 miles now. Will be worthless in a few years and end up as the nicest car in the scrap yard.
I really do hope that someone with the resources to build this listens and finds a way to make it real.
Absolute fucking genius. I love the Pontiac Fiero-like substructure with bolt-on steel or plastic body panels. I love the simplified interior. And cribbing AMC cost-cutting techniques is exactly what this project needs.
Somebody with money needs to back this.
You guys sound like the commenters who say cars could be cheap again if we just got rid of power windows and locks.
A new Nissan Versa IS $16,680 (sometimes I’m tempted to buy the 5MT version because maybe I will be the asshole that’s collecting Versas in the future). A new Nissan Leaf (a Versa minus a very complex ICE and transmission but plus a battery pack) is $28,140.
You’ve addressed the cost of every trivial thing except what actually makes a Leaf expensive.
Also, in Europe, the battery leasing alluded to here has been tried, including with the Leaf, but apparently was not too popular:
https://www.electrifying.com/blog/article/what-is-battery-lease
See also, addressing the obvious advantages but also challenges:
https://www.technologyreview.com/2023/05/18/1073291/rent-ev-battery-swap/
True, the battery/drivetrain is a huge expense. Makes me wonder if a gas version could be close to $10k.
Reminds me of the Fiat Panda Mk1. And the new Citroën Ami. Love it!
Also a lot of Fiat Ritmo/Strada Mk1 with the googly eyes headlights in the bumper 🙂
I do like it a lot, when cost cutting makes design simpler.
And with AMC out of the picture, who makes really cheap and simple cars? I’m thinking Dacia.
I’m not convinced that suicide doors are necessary or even desirable here.
The real win is when the doors are fully symmetrical on both sides of the car and capable of being hinged or latched on either edge, so that any one door can be inserted in any of the four portals and fit perfectly just by moving the hardware to the other end. Then, instead of having two sets of dies for stamping the doors, you only need one.
And with fully symmetrical doors, you don’t need to worry about the safety drawbacks that go with suicide doors.
What safety drawbacks?
They kill themselves, hence “suicide doors”, they become unhinged more easily. It’s a metal health issue that drives them mad.
That would be the refrigerator door solution, where you can reverse the swing and use trim (a long handle or plastichrome, or these days just a small cover in the same finish as the rest of the appliance) to cover unused mounting holes.
I would buy this,in fact I will be crossed if you don’t build it now.
God damn, that looks like Socialists ideology and we all know that means Communism, dang it (sarcasm alert before I trigger the internets, lol).
TBH, this sounds way too easy and far too awesome for mainstream manufacturers to make, unless it’s a very unique sub-brand – this is one of those ideas that the Swatch peeps would have come up with if they were thinking about making cars in the 2020’s.
Key points:
LFP chemistry with pseudo 800 volt architecture. (two 400 volt batteries in parallel). This will enable longevity, full usage of capacity + somewhat quick charching, balancing out the next point, which is: small capacity battery.
Sparse interior, think Dacia Sandero and not a unique design as this article points to.
Off the shelf components are always cheaper than unique bespoke components.
RWD only.
Drum brakes in rear.
Does that $16k include TruCoat installed right from the factory? Cuz I don’t want any oxidation problems.
Love it and it reminds me a bit of Jean-Marie Massaud’s We.Me concept for Toyota from 2013
A pushme pullyou version of the Ford O21C Concept Car would have me cutting a check.
Do people still use checks?
Swiping my iPhone — whatever.
Speaking of checks, the Checker cab only had one door repeated four times. Having two different designs of doors complete decadence.
I’m not cracking on the overall idea, I love all the thought that went into it, but my MBA side has to point out that you’re not avoiding the super-expensive battery replacement cost, you’re just amortizing it so the user doesn’t see it all at once. Those swappable battery packs have to be replaced eventually so they have to be priced to anticipate that (though the labor side is way lower since they’re not integrated into the frame).
I had similar thoughts, but then I applied Moore’s Law. The batteries will get smaller and cheaper, just like integrated circuits. And if enough people are using swappable batteries, the retail scale of it all would bring the costs down even more. It’s just a matter of time.
Plus, I can better afford small costs over time than big cost up front. Maybe not the best financial move, but for someone who needs cheap reliable transport NOW this might be the way that happens. I can see this aimed at recent school grads with their first decent job. Future salary increases combined with costs going down over time would make this very doable for the Average Consumer, I think.
If only Moore’s law actually applied to batteries! We’d all be driving vehicles with 1000 miles of range that ran on a handful of AA cells. I get your point, battery technology improves over time, but not at anywhere near the rate of Moore’s law.
You’re right, but Moore’s Law does apply to all technology- with the “regulatory” caveat. Just look at how much better EVs are today than just 15 years ago, when they were a joke, and not even a funny one.
I like it but the reality is, this is like a brown manual transmission long roof. Everyone talks about wanting one, but the potential market is low to non-existant. No manufacturer could make money on it except as a lost leader used to upsell the fully optioned moly muffler bearing units priced at a base of 45k.
I’d buy it.
I think the modular swappable batteries would struggle with fast charging at a good clip due to cooling requirements for decent charging speeds. Not sure how to overcome that
Batteries that need cooling systems are a transitional technology. They’ll be gone as the technology develops.
Build the car in three modules. A passenger compartment, a front compartment, and a rear compartment. Make the front and rear as identical as possible and you could even make it have an AWD upgrade option.
And if they’re complete modules, crashing one of them doesn’t total the car. Just unbolt it and replace it.
This is an excellent refinement of an idea we’ve seen Torch mess with for years here and on The Other Site. The belief that A Better World Is Possible is right there in The Autopian’s name, after all. It might mean an end to major model cycles, but if we look back at the long production lifespan of cars like the Beetle, the Hindustan Ambassador, or the Peugeot 405, I’m not at all convinced that’s a bad thing.
Look: Fiat Ritmo, 80’s.
Seats: Fiat Panda, 80’s.
Huge real estate of unpainted plastic becoming part of the design: Fiat Ritmo, Fiat Panda, and every European small car made from the 80’s to the early 2000’s and sold on a local market (Fiat in Italy, PeugRenauCit in France, Seat in Spain, etc).
Ha. Fiat would have been right about something ?
Ok, so lets run some rough numbers… On a car optimized for low cost and re-using body panels, stampings etc, you’re going to be pretty far from an ideal aerodynamic shape. Even assuming optimized underbody parts, a mid-sized crossover type vehicle with the general shape you’re showing would be lucky to hit 3 miles/kW-hr, which means that for your 40 miles per module design, each module needs 13.3 kW-hr.
If cheap & safe is the name of the game, LFP is the answer. Less $$ in Ni, Co etc, less need for TRP content, less need for structure to provide compression to the electrochemical stack (especially for prismatic can or cylindrical cells). The next generation of LFP is looking like it’ll be about 200 W-hr/kg, and on a module level lets be generous and say $120/kW-hr even with all the module interfaces necessary for swapping. So each module will cost $1600 and weigh 66.5 kg (probably more like 80 kg factoring in packaging, bussing, interfaces for swapping and cooling etc).
The cost for an initial 2 modules to go with the vehicle seems within the realm of plausibility, leaving $12800 for the rest of the car including motor, inverter etc. If Toyota’s IMV 0 truck is $10,000 and the electric motor, invertor & 1-speed is about $500 less than the basic 2.0L engine & transmission (which are already going to be amortized and as de-contented as possible), then we’re left with $3300 to add a second row of seats, roof, air bags & other safety improvements required for a USDM vehicle, and the added costs of all this modularity (which won’t be as cheap as you’d think). Ok… not bad so far.
That 80 kg module though… that’s 176 lbs – that won’t be that easy to swap out, and you won’t be doing it though the rocker panel – that’s an incredibly important piece of the vehicle’s structure required to pass side-pole crash testing – it needs to be thoroughly tied into the rest of the vehicle in many places to be effective, making it very difficult to just swing up and out of the way. Also, if you aren’t using the battery structure for absorbing and dissipating any side-pole crash loads, you’re carrying a lot more weight around than your competitors are, making energy efficiency and cost much worse.
Electrically, lets say you’re aiming for 300V when fully charged for a 6-module pack, so each module is at most 50V. This is chosen because 50V DC is the dividing line for low-voltage vs high-voltage based on the ability of DC to injure you, and at which point the protective gear and level of electrical isolation required increases dramatically, increasing costs with it. If each module can be 50V at most, and LFP is 3.5V at 100% SOC (“nominally” 3.25-3.3V), then you have 14 cells in series, with each parallel group at 290 A-hr. However, if you only have 2 modules installed, you’re at only 100V, which is going to force those modules to work a lot harder and push a lot more current through the motor, invertor etc for the same power and performance, forcing the motor and invertor to be either incredibly oversized and more expensive or forcing a serious performance de-rating when operating with 2 modules vs 6 (and degrading those two modules much more quickly even still). The modules could all be wired together in parallel, but this forces each module to have a much higher voltage, safety requirements, electrical isolation and cost because of it, and forces each module to have 84 cells at minimum each instead of 14, increasing manufacturing cost, complexity, module mass and volume (lots more interfaces between cells to connect, lots more waste in additional cell can structure and bussing, cell balancing becomes significantly more difficult and wasteful with a string of 84 cells vs a string of 14… it all adds up).
You won’t be able to charge those two modules much faster than you could for 6 either – the cells are still C-rate limited for charging without Li plating or dendrite growth, the thermal interface between the cells and the coolant is the same, and while you have more mass in the coolant system if its sized for 6 modules (oversized for 2), this will only buy you small gains.
If swapping is going to work at this price point, I think it’d need to be a rare occurrence, done at a dealer or mechanic with more robust connections to the vehicle for structural, electrical and cooling needs. You’d also have to have an empty module “box” for reacting side-pole crash loads that would be a lot less expensive but bolt in place – maybe also plugging cooling connections and bridging electrical connections. There are lots of reasons why adding and removing modules on-the-fly isn’t something that has been offered yet – it adds a lot of cost and complexity for a benefit that people are highly unlikely to use. Even in packs that are designed to be somewhat modular and offer different capacities within the same basic pack (Like Lucid or GM’s Ultium), there is a reason that the variants don’t stray too far from the maximum. In Ultium’s case, GM can go from ~100kW-hr from the Lyriq/Blazer to 200kW-hr for the trucks because the large pack is essentially two smaller packs stacked on top of each other and generally run in parallel unless utilizing an 800V charging source, in which they’re connected in series.
Brillant
You really hit all the high points.
Before you plug batteries together in parallel, you have to balance the modules. If you don’t, the result is similar to pluggin them in backwards. Balancing is not a swift process. Connecting in series is safer, but if one module is empty, the whole pack is still dead. So once again, you’re back to the painstakingly slow balancing process.
It’s an interesting concept, but… not buildable at that price point. Yet. You’ll want gigacastings, 3d printable parts, and almost no human touch to pull it off. Let’s assume that we have good batteries (200wh/kg) and get decent efficiency of 4 mile/kwh (doable with that shape if you make it slow.) Those 40 mile packs weigh 10kwh / 200wh/kg = 50kg + whatever the structure and cooling require, so 60kg or 123lbs? You aren’t lifting that into a charging bank alone. So now we need robotic battery swap stations, which has been done. This isn’t looking inexpensive at all.
3D printing is not a mass production technology, moulding or casting is cheaper.
I like it. It’s cute.
I’ve thought about this idea a lot, so here goes, randomly-
I really like this idea. But I suspect its lack of streamlining will work against it.
Low drag is how you get more range per dollar spent on batteries.
In the interest of both streamlining and symmetry, it may be worth studying symmetrical boats for inspiration on how to refine this shape into something decently slippery while being able to keep the amount of stamps needed down.
If your Changli were a dog, it’s be growling with jealousy right now.
If canine Changli could tolerate being clipped and groomed with a chainsaw, it’s relaxed enough to go along with this.