Mercedes‑Benz starts large‑scale production of electric axial flux motor

14 hours ago (media.mercedes-benz.com)

Mercedes acquired Yasa (UK) couple of years ago and now getting up to the speed in the production.

Here is a nice video that explains axial flux motors with a factory visit

https://youtu.be/B2Hl4c1iZK0?si=VfDYARyuaPVj1nKm

They are so, so, small.

  • I was curious how this thing works and asked Claude to visualize it -- mostly to see how good Fable is and I have to say, what it made was good enough for me to get a gist of it. Posted it here

    https://azimi.me/axial-flux-motor-explainer/

    • Every plausibly cool electric car innovation leads me to the same thought: “5-10 years from now, the restomod potential will be wild once these come down in cost.” For this, I am imagining retrofitting a Pontiac Fiero to reduce as much weight as possible and see if extended flight becomes possible.

      28 replies →

    • Can you elaborate a little on what you asked Claude to do here? This is a pretty impressive one-shot.

    • Thanks for sharing. I wish it was a bit more interactive especially when there are parameters, e.g. "Widen the disc and torque rises with diameter cubed" I wish there was a slider to see that effect and thus maybe why there might be a sweet spot.

      Also I have "The Way Things Work" on my desk right now and can't help but wonder, could you adapt some of the pages of the book this way? It seems like exactly the kind of content that would benefit from such 3D (interactive) visual explainers.

      11 replies →

    • I am not an expert, but I do know some physics and I know how to read, and I’m pretty sure this is full of BS. Also it’s a really crappy visualization.

  • Great video!

    So it looks like axial flux, the OG was introduced in 1820 something and it wasn't easy to manufacture. So radio flux came after that and has been around ever since. So axial flux is making its come back this year!

    The video is very interesting too about decompounding returns when the motor is less with the other things need to weigh less too.

    Especially the bit about potentially not needing brakes in the near future because the regen is so capable. Which would lead to less weight and less parts even again!

    • Also found it fascinating, although on the discussion about brakes I thought about how regen braking turns off in my EV when the battery is full, because there is no where to put the power. So you either keep some of the battery always available to soak up braking energy (and hope people never charge to full at the top of a mountain and exhaust the buffer) or you include a set of normal brakes for when regen is not possible, both options negating the weight savings. Right?

  • I wonder how developing electric motors compares to combustion engines. My hunch says that it’s the main reason the Chinese high-tech electronics industry was able to develop and iterate leading electric vehicles so fast. (Edit: My more clarified point is regarding the machinery required + place to accommodate them to work on electric motors vs. ICE metal parts and all the intermediary parts transfering power in the drivetrain. The shop in the video is smaller than many would imagine.)

    When these hopefully go to the next generation Formula E cars, we’ll see some crazy improvements in cornering. The newest generation already has active 4WD. I imagine this can bring even better torque adjustment improvements.

    • The Chinese EV industry is actually lead by development of batteries, especially CATL. Along with the pack engineering, which is good old Mech.E stuff about heat transfer and physical strength.

      Secondarily power electronics; at that scale, you can't just pick a bigger transistor and call it a day.

      By comparison the motors seem to be a mostly solved problem, although I'm sure there's still some scope for power-to-weight engineering there, it's not as critical as the battery pack.

      9 replies →

    • Standing on the shoulders of giants, mainly.

      All the industrial processes and machine tool development that happened in the ICE car industry over the last century (and the electronic hardware manufacturing, more recently) was available day one.

    • Low power and high power electronics don't actually have a lot in common other than the broad types of engineers you need to be educating.

      China has industrial policy. The country and companies are able to invest in BEV technology knowing that everyone agrees on the direction.

    • >I wonder how developing electric motors compares to combustion engines. My hunch says that it’s the main reason the Chinese high-tech electronics industry was able to develop and iterate leading electric vehicles so fast.

      The talent had very little impact to be honest. The primary factor was a government looking 50 years down the road seeing that:

      1. ICE engines have little to no long-term future in transportation.

      2. global warming is a thing whether the right wing in the US likes it or not.

      3. They were never going to overtake the West in ICE engines and had to attack from a different angle.

      The US' lack of breakthroughs in EVs has little to do with technology or expertise and everything to do with an administration that is openly hostile towards EVs and renewable energy in general. For the rest of the planet, EVs becoming the primary form of transportation is just an obvious and logical conclusion, even if it takes us another 25-50 years to get there.

      China saw it and decided to heavily incentivize and subsidize the rapid expansion of EVs both to fix the air quality issues in China and corner the market.

      5 replies →

    • AFAIK Chinese electric cars used western motor drivetrains and often still do?

    • I'm not saying there isn't room for innovation but... Electric motors have been around for a century now already in many applications.

      It's not like the Dawn of the steam engine

      2 replies →

    • > develop and iterate leading electric vehicles so fast.

      How do you know this for a fact? Chinese press releases? You've driven one? Some auto blogger drove one?

      After world war 2 Gorbachev or whoever visited the United States and during that trio visited a supermarket. He thought it was a facade, possibly, put on just for him, there's no way Americans are this prosperous (or whatever, this good at agriculture, farm equipment, etc)

      Also do the race cars have 4 wheel drive, or all-wheel drive? I'm wagering all-wheel with "torque vectoring" and "Yaw control", like a Mitsubishi Lancer Evolution X.

      10 replies →

  • A few years ago I built a few small ones in my home lab. They are extremely efficient. With the right setup, have insane torque, and don’t require a lot of energy to do it. The best part for me was that they are completely brushless by design and are easily to keep clean.

  • Came here for this, thank you for providing a link

    Edit.... Video doesn't seem to explain very well either

It would have been awesome if that article had, at any point, explained what an electric axial flux motor was, and why anyone might want one.

  • Click "More" and scroll down:

    "In contrast to conventional radial flux motors, the electromagnetic flux in an axial flux motor runs parallel to the axis of rotation. The key components are arranged in a disc‑shaped layout: two rotors sandwich the stator from the left and right. This design enables an especially compact motor architecture, high power and torque density, and new freedoms in drivetrain packaging. In the new Mercedes‑AMG GT 4‑Door Coupe, the motor at the front axle is just under nine centimetres wide; the two motors at the rear axle each measure around eight centimetres in width. The three axial flux motors are integrated per axle into so‑called High Performance Electric Drive Units (HP.EDU), where they are combined with a compact input planetary gearbox in a single housing."

  • Visited Astrall Dynamics, a Chinese startup that builds quadrupeds with axial flux motors here in Shenzhen. Super cool to see the robots in actions, carrying 60kg of weight up over 20 flights of stairs quite rapidly. The high torque at the compact form factor was super impressive. As far as I understood they are more complex to manufacture, especially at scale.

  • It's a bit buried, but it does:

    > In contrast to conventional radial flux motors, the electromagnetic flux in an axial flux motor runs parallel to the axis of rotation. The key components are arranged in a disc‑shaped layout: two rotors sandwich the stator from the left and right. This design enables an especially compact motor architecture, high power and torque density, and new freedoms in drivetrain packaging. In the new Mercedes‑AMG GT 4‑Door Coupe, the motor at the front axle is just under nine centimetres wide; the two motors at the rear axle each measure around eight centimetres in width. The three axial flux motors are integrated per axle into so‑called High Performance Electric Drive Units (HP.EDU), where they are combined with a compact input planetary gearbox in a single housing.

  • “What“ might be a long answer, but why anyone might want one is to have increased torque density for the given volume and diameter. So they are thin motors where the generated flux is parallel to the shaft. And they are like the standard PMSMs where you apply the same driving algorithm from the inverter side to use them.

  • Most motors have N-S axis of magnets aligned tangential to the axis of rotation. Axial flux motors have N-S poles parallel to rotation. This allows motors to be thinner and wider as well as anyhow more lighter and sometimes easier made. Whether they make sense depends, it seems.

  • https://en.wikipedia.org/wiki/Axial_flux_motor

    • The automotive part is interesting:

      https://en.wikipedia.org/wiki/Axial_flux_motor#Automotive

      > Mercedes-Benz subsidiary YASA (Yokeless and Segmented Armature) makes AFMs that have powered various concept (Jaguar C-X75), prototype, and racing vehicles. It was also used in the Koenigsegg Regera, the Ferrari SF90 Stradale and 296GTB, Lamborghini Revuelto, McLaren Artura and the Lola-Drayson.[9] The company is investigating the potential for placing motors inside wheels, given that AFM's low mass does not excessively increase a vehicle's unsprung mass.[10]

      > In July 2025, YASA announced a prototype 550 kW (738 hp) 13.1 kg (29 lb) motor, equating to power density of 42 kW/kg, which the company claimed to be the highest ever achieved.[11] By contrast, the state of the art EV motor from Lucid Motors offers a 500 kW, 31.4-kg motor, or 16 kW/kg.[12]

      > The first application of these motors will be in the High Performance Mercedes‑AMG GT 4-Touring Coupe.[14]

    • Thanks for posting this. Axial flux motors aren't some new sci-fi invention. We've had them in gadgets for a long time like in the floppy drive example. This is just one of the first industrial scale implementations of high-torque applications.

      1 reply →

  • As far as I understand it's so small and lightweight you can put one on each wheel and remove brakes and still save weight (something something unsprung weight bad).

Very cool. Good to see more axial flux motors in the wild - will be interesting to see if they become the new standard in future. With smaller material costs the cost to manufacture at scale could actually become lower than radial

I expect radial will still dominate for at least another decade or so outside of premium performance focused cars. Radial has been battle-tested and proven. Axial still has a few more years to prove it's reliability in the field. Higher loads and stresses, tighter tolerances could make the axial motors less reliable overall especially at mass market trims. Mercedes is probably over-engineering for reliability and performance on the premium car

Radial is also "good enough" for most applications. The efficiency, form factor and weight improvements of axial is nice, but they aren't the limiting factor. Radial is already highly efficient, reasonably light and small. The real level for weight is the battery

  • I'd like to see these if these can be scaled down into ebike size, or beyond that into micromobilty form factors that people carry around (like lightweight scooters and skateboards).

I know what a "motor" is and I know what "electric" means but I have no idea what an "electric axial flux motor" is. I'm pretty sure it's not what Doc Brown used to travel through time but it could be.

About once a week I think I'm too dumb to be on this site and today is that day. Off to duckduckgo to do some research.

Edit: it's actually pretty simple. Here is the wikipedia on them:

https://en.wikipedia.org/wiki/Axial_flux_motor

It could use some more diagrams, but it makes the gist of it pretty simple to understand.

I remember when YASA announced it and when MB bought them. Amazing technology and advancement in electric motor design. Good to see they somehow try to commercialize it.

I want to do an engine swap in my 1980s Toyota pickup (like on Back to the Future) from a 100 hp 22r to a 150-250 hp fuel injected inline 4 or turbodiesel to raise the thermodynamic efficiency from 20-25% to ~40% to nearly double fuel economy.

Unfortunately, most modern engines are transverse mounted. They can fit any transmission with an adapter plate, but then they're set too far back into the firewall to access stuff like the high-pressure fuel pump (which is often mounted on the transmission side for easy access on front wheel drive vehicles). I feel that's by design for planned obsolescence.

So I really wish that someone would offer a 4-6" thick 100-200 hp (100 kW) axial flux motor insert between the engine and transmission. Optionally with a simple battery management system (BMS) storing perhaps 5 kWh to provide up to 15-20 miles of electric range and hybrid fuel economy with regenerative braking.

If anyone knows of one, please let us know! If not, then those of you who won the internet lottery could make a killing investing in a novel product that everyone wants but doesn't know it yet.

  • You could swap the 2LTE engine, that’s the equivalent hilux diesel for 80s. You may have to import though, they are few and far between in the us.

    Totally with you, I want more options for my Tacoma also!

    I’ve been playing with idea of importing a lhd hilux from Mexico

  • something like a BMW 330e drivetrain has that (as do most German RWD hybrids), the biggest issue there is going to be the inverter and control software.

For years we've been seeing the hype about how axial flux electric motors will save 100 or even a couple of hundred pounds per car. Does this announcement mean we're finally starting to see the first increment of that actually happening?

(The motor is less important than the battery in terms of total weight, long term durability, etc. But nonetheless, any improvement helps!)

Will it be expensive, unreliable, and repairable only by MB dealers? Much like their gas cars.

An interesting part here is probably manufacturing and not the motor itself : going from a prototype to something you can mass produce reliably is often the hard part

Glad YASA's achievements are being realised but the UK really needs to get it act together so we can fully realise the next tech breakthough.

My highlights:

> In the Coupé, the engine on the front axis is 9 cm (3.5 inch) wide, the two engines on the rear axis are 8 cm wide each (<3.2 inch).

> The fully electric "Performance" model accelerates from 0 to 100 km/h in 2.1 seconds.

ETA: Images of the engine:

https://media.mercedes-benz.com/article/bebac2af-acdc-465a-9...

https://media.mercedes-benz.com/article/bebac2af-acdc-465a-9...

Tangential, how much regen can this system support?

For example, can a car with 200kW propulsion have a 400kW regen (Tesla has upto 65) and are cost effective like friction brakes?

  • All motors are generators. It's only a matter of how you are creating the magnetic field with the stator windings.

    In order to generate a higher regen, you'd have to somehow get more energy in the motor first... and since its only rated for 200kW, good ol' physics limits you, IF thats all the energy you put into the system.

    If you roll it down a hill, or do something exotic like inverting the magnetic fields .... you can exceed the motor rating. But thats usually not recommended because the motor driver itself isnt rated to handle that power.

    • In general the rating of a motor is about heat dissipation, which in turn is a function of efficiency. What this means is you can exceed the rating by "some amount" for "some amount of time". Many motors are rated for not what they can deliver, but what they can deliver continuously without overheating, but you can get a lot more power out of them for a short time.

    • You're also ultimately limited at the point where the tires lose grip and start sliding. Which can vary a lot based on road surface and conditions.

  • 400kW regen is going to be a fairly alarming level of stopping for a normal vehicle. It's OK to leave emergencies to the friction brakes.

    • The idea is deleting the friction brakes entirely, saving costs, maintenance and weight.

      And 400kW isn't really all that much for a sports car. I remember 911 ads from the '80s that boasted "brakes with more than 1000 horse power".

      2 replies →

  • > can a car with 200kW propulsion have a 400kW regen

    At the motor level it should be the same, in propulsion you’re converting current to torque and in regen you’re converting torque to current, with the same hardware. The high voltage wiring is the same and will set the same limit on current regardless of direction.

    I believe bidirectional inverters are generally symmetrical as well, so that should not be a factor.

    Which I reckon leaves two factors:

    1. Battery C rates, afaik pretty much all chemistries have a higher discharge rate than charge rate, especially when trying to maintain them for a long time, so by that account regen power would at most be the same as propulsion (if the entire power train is sized for the battery’s charging rate).

    2. Artificial limitations, obviously you could always artificially under-prop, though that seems unlikely outside of niche applications.

    tldr: I don’t think so, except on a technicality (that you can artificially hobble propulsion).

    • How far fetched is the idea to use Super-Capacitors to take up the energy generated by braking and then slowly feeding it to the battery at a rate that it supports?

      1 reply →

  • Am I reading you right that breaking power (that you want to regenerate in the system) >> speeding power? Obvious now I come to think of it, and still pretty nifty new thing learned if true!

For family cars we need 4x 30-50 hp units. If the motor can weigh around 7 kg it can be placed directly on wheel. Adding durable brake discs (rarely used) and 2 inverters front ad back and we have the EV platform of the next 100 years

  • Motors in each wheel eats tires.

    Personally, I'd rather see FWD with 1 100HP motor in a 2200-lb 4-seater under $20K US, but that will never happen as the supply is artificially constrained to create high-end cachet.

    • I assumed it was to target the most motivated, price-insensitive buyers while recouping fixed costs.

  • do we need disk brakes?

I wonder, is axial flux motor as cool sci-fi sounding in German as it is in english?

Only slightly related but does anyone know anything about motors with magnetic bearings? As in, no contact or friction. I'm looking for a hardware project

  • There are fluid dynamic bearings which were used in VCRs and probably hard drives and definitely laser printer mirror motors - two sets of precise herringbone patterns cut into the ID of a bearing column and a tiny bit of oil that gets entrained between rotor shaft and that bearing column. As the motor speeds up the oil forms pressurized donuts. Only works at speed and generally only useful when there aren't side loads so applications are limited. Rolling element bearings are a quite developed technology and hard to beat in most applications.

  • Closest I can think of is flywheel "battery" storage tech many of which do have magnetic bearings and also some way to get power in and out of the flywheel so basically a motor. It's not exactly what you're looking for but there's prior art out there.

  • maglev but those do linear acceleration instead of rotational. seems tricky though for cars with such uneven surfaces.

  • I wonder if that is even possible since the whole point of the motor is an imbalance of magnetic forces.

I would guess that hydro and other generator forms would benefit from this design as well?

Personally I’d love to see this make it’s way into power tools and CNC motors.

  • Probably not. A huge disadvantage of axial flux motors is they have a large number of poles, which means that they get less efficient at high speed because they require say 5x as much switching.

    This makes them kind of unsuitable for power generation and really high power motors (despite their power density) where the main way you get more power is just to spin really fast.

    The other disadvantage is they have such a low amount of material in them, that the stator overheats really easily. And the topology of the motor makes it really difficult to get the heat out efficiently, which again limits their maximum power.

This headline claims large-scale production, but the article never indicates what that would be in terms of quantity per year.

Their motor is pretty cool. So are lots of other ideas and concepts. This is supposed to be about production. Arguably, the coolest thing about Yasa is the machines and process they have created to produce their motor in production quantities.

It disappoints me when an article promises to be about production but seems more to be a press release about the product.

I wish them well and would be excited to learn more about their actual production capacity.

For a century Germany's comparative advantage has been [mechanical] engineering. As a European I want (need?) Germany to succeed. Ergo: more of this, please.

  • Their advantage was mechanical engineering. They have historically not done well with electrical systems. So this is a change - hopefully for the better.

    • That's not quite right.

      For example Siemens and Bosch are large enterprises specialised in industrial scale electrical machines and parts (among other fields).

      Infineon was spun off from Siemens 25 years ago an plays an important role in chip manufacturing for automative systems.

    • Electrical auxiliary stuff used to be and maybe still is a source of trouble. Power electronics should be fine, there's plenty of know-how available in the country, and with any luck, car companies won't have time to home-grow a mismanagement structure to fuck it up like they did with software.

      Software and battery cells are the main challenges.

There isn't going to be a lot of improvement in overall EV performance/capability from better motors. Existing, boring motors are already close to 100% efficient, already small, and already powerful.

Advancements here chip away at margins, its nice but nothing to get super excited about. Whereas a modest ~20% increase in energy density from batteries would be amazing. Every little bit we improve there unlocks new capabilities. Towing long distances, smaller affordable economy cars and sports cars, airplanes, etc.

I am speculating but here might be reasons axial flux motors have advantage over radial flux motors:

1) torque: torque = applied force x length of the lever. Because the radial flux rotor must fit inside the stator, therefore radius << motor outside diameter. With the axial flux motor, the rotor is adjacent to the stator, therefore radius < motor outside diameter. Axial rotor radius > radial rotor radius.

2) space efficiency: in a radial flux motor you have 1 rotor, the coils arranged so that one end of the coil's magnetic field is useful to work on the rotor, the other end is not used. In an axial flux motor, (1) pancake rotor at each end of the coils, total (2) rotors, the coils can act on a rotor at each end. There is no free lunch here, to do useful work you still must provide more energy to the coil, but you can get the most from the space.

There must be someone here with a better handle on the electromagnetism, please correct me where I err.

Ah, another fantastic British innovator (YASA) having to realize its potential (and ultimately the downstream economic benefits of commercialisation) abroad.

Brought to you by the only country to have a space programme and abandon it.

  • Britain didn't abandon it's space programme. It abandoned a launch rocket programme though. That was over 50 years ago and the rocket was less capable and more expensive than alternatives at the time.

    • What alternatives? By your logic only one country should have a launch rocket. Thankfully that’s not a world we live in because that makes no sense. But I’m happy for you if you can be content with a space programme without a rocket, that’s a nice low bar to live with, you can basically never miss.

  • > Brought to you by the only country to have a space programme and abandon it.

    I mean, so did France; they both essentially folded theirs into ESA.

  • Did they have to? My impression is British companies sell out as soon as they can these days. Is this something that could be changed with policy? Does Germany incentivise running companies more? Or is this cultural, e.g. British people are more risk averse?

    • It's very difficult to raise late stage capital in the UK, especially for deep tech. We invent so much but our capital ecosystem is all tied up in land and our pensions providers don't want to know.

      5 replies →

    • I think Germany has tax rules that make exits harder, whereas it's very easy in the UK to sell. If you have a more free market next to protective ones it makes sense that your IP is going to flow in that direction.

    • It’s cultural. It is not difficult to raise a lot of money in the UK. The problem is that the UK (government, investors, employees and employers) got so high on the margins of services and finance in the 90s, that it has never recovered from this all-consuming addiction. Everything else simply attracts no interest comparatively, economic diversification be damned.

10 years behind Tesla, they are doomed

Never become dependent on doing hideously complicated things. You will eventually struggle to choose to do something more efficient, as the european auto industry is currently displaying. The car where thid motor will be used will, given current market sentiment, be a massive flop. Here they are showing off how complex the manufacturing process is. Surely we’d all be better off with simpler and cheaper processes.

  • This entire product is easier to build than the existing technology, has more simple tooling and will be simpler and cheaper.

    Designing the manufacturing machinery is exactly what happens in any manufacturing process. Those robots are general purpose that have been adapted for the required tasks, that's a normal process.

    Why would you build a motor that's twice as heavy with copper and much wider when you don't need to?

  • Efficiency and cost savings at scale usually involve an increase in complexity: in mass manufacturing, complexity is generally a fixed cost and so can be amortized over larger volumes.

  • A typical modern car is already hideously complicated and a different type of motor would not change this.

    What is the current market sentiment? Share of EVs is slowly rising so having a good motor as important as ever.

  • By that logic we should all just be writing assembly manually. Screw hideously complicated higher level languages. Screw LLMs in particular, so complicated!

  • > Never become dependent on doing hideously complicated things

    Is Mercedes stupid?

    How did Carl Benz dare to do something as hideously complicated as building the first gasoline-powered car in history?

    And why did they kept inventing complicated stuff that ended in all modern cars like ABS, adaptive cruise control, direct fuel injection, emergency brake assist, etc, etc?

    • Not all of those inventions are bad. But not all of them are coming from a place of necessity. All of them do increase complexity. My gripe with Mercedes is not that they are constantly pushing boundaries on what can be done with more tech. My main gripe is that the EVs they are building are essentially as complex as the ICE cars and follow largely the same design principles as the ICE cars. For instance, in the EQS, instead of applying engine breaking when the driver takes their foot off the pedal, they went to great lengths to _move the break pedal_ in proportion to the amount of engine breaking that is currently being applied as per the VCUs command. And yet the door cards on the EQS are not up to the standard of an S class.

      My main gripe with MB is that they have this new technology that could simplify things and let them build a better product. Instead of building around it, they shove it in to their existing designs. I was expecting an electric S class to be more akin to a Lucid Air sans the teething problems of a new company. Instead, we get weak attempts at solving non issues.

      And whilst they are certainly not in the market of producing affordable vehicles, I would hope that using EV tech they could create a better version of their existing fleet. I do not think anyone buying an A class cares about the 4 popper under the hood - losing it and simplifying radically, in my mind at least, would give them more budget and leeway to create a more compelling product.

      8 replies →

Great, they finally started mass-producing 19th century technology, let's cheer that! Nowadays, while Chinese and Americans are producing GPUs, AI and li-ion batteries, German high-tech is an engine invented by Faraday