Radial flux and axial flux describe two fundamentally different ways an electric motor channels its magnetic field, and that difference shapes everything from how compact the motor can be to where it performs best. In a radial-flux motor, the magnetic field crosses the air gap perpendicular to the shaft, radiating outward like spokes on a wheel. In an axial-flux motor, the field crosses the gap parallel to the shaft, flowing along the axis from one disc-shaped component to another. Most electric motors in the world today are radial-flux designs, but axial-flux machines are pushing into electric vehicles, aircraft propulsion, and wind energy because of their unusual ability to pack high torque into a flat, lightweight package.
How the Two Designs Differ Physically
A radial-flux motor looks like a cylinder. The rotor sits inside the stator (or occasionally outside it), and the magnetic field jumps across a cylindrical air gap. This is the geometry most people picture when they think of an electric motor, and it has been the industry default for well over a century. Scaling a radial motor up in torque usually means making it longer or wider, and torque output scales with both the axial length and the square of the diameter.
An axial-flux motor, by contrast, looks more like a pancake or a thick disc. One or more disc-shaped rotors face disc-shaped stators, and the magnetic field crosses a flat air gap. Torque in these machines scales with the cube of the diameter but is essentially independent of axial length, which is why they naturally adopt a short, wide profile and can achieve high torque density simply by growing in diameter rather than stretching longer.1Chinese Journal of Aeronautics. Multi-physics evaluation of radial- and axial-flux YASA high-torque-density PMSMs for aircraft electric propulsion considering size constraint effects That cubic scaling relationship is one of the main reasons engineers get excited about axial-flux designs for applications where space is tight.
Where Radial Flux Motors Still Dominate
Despite the buzz around axial-flux machines, radial-flux motors remain overwhelmingly dominant in industrial use, home appliances, power tools, and the vast majority of electric vehicles currently on the road. There are good reasons for this. Radial-flux stators can be stamped from flat sheets of electrical steel in high volumes at low cost. The cylindrical geometry makes it straightforward to press-fit the stator into a housing and mount the rotor on bearings. Decades of supply-chain optimization mean that radial-flux motor components are commodity items.
Radial-flux motors also handle certain mechanical challenges more gracefully. The air gap in a cylindrical design is maintained by conventional radial bearings, which are cheap and well understood. Thermal management is relatively simple because the stator is a solid ring of laminated steel pressed against a cylindrical housing that can be cooled with a water jacket or fins. And because the rotor sits concentrically inside the stator, there is no strong net axial force trying to slam the rotor into the stator, a problem that axial-flux machines have to manage carefully.
Why Axial Flux Is Gaining Ground
The core appeal of axial-flux motors is torque density, meaning how much torque you get per kilogram or per unit of volume. Because torque scales with the cube of the diameter rather than its square, a flat disc-shaped motor can deliver the same output as a much heavier cylindrical one when diameter is not constrained. This makes axial-flux designs attractive in electric vehicles (where every kilogram matters), in aircraft propulsion (where both weight and packaging matter enormously), and in direct-drive wind turbines (where eliminating a gearbox requires a motor that produces extremely high torque at low speed).
The flat shape also opens packaging options that cylindrical motors cannot match. An axial-flux motor can be tucked into a wheel hub, sandwiched between a gearbox and a differential, or mounted directly onto a propeller shaft. Several electric-vehicle startups and at least one major automaker have adopted axial-flux motors for exactly this reason, using the freed-up space for batteries or cabin room.
In aerospace, the combination of low weight and high torque density has made axial-flux motors a leading candidate for electric vertical-takeoff-and-landing (eVTOL) aircraft, where multiple motors drive distributed propulsion systems and each motor needs to be as light as possible.2CrossRef. Design and Evaluation of a High-Power Density Axial-Flux Motor-Propeller System for eVTOL Distributed Electric Propulsion The disc shape also helps in this context because a flat motor can be integrated directly behind a propeller, keeping the nacelle compact.
The Thermal Problem
One of the biggest engineering hurdles for axial-flux motors is heat. In a radial-flux machine, the stator is a thick ring of steel that conducts heat outward into a housing, which can be water-cooled with a simple jacket. In an axial-flux machine, the stator often sits between two rotors (a layout sometimes called YASA, for “yokeless and segmented armature”), and getting coolant to the stator in that sandwich arrangement is tricky. The stator is essentially trapped between spinning discs, so conventional cooling methods do not translate directly.
Engineers have responded with increasingly creative thermal designs. One approach uses integrated water-cooling channels built directly into the stator assembly. A recent study on YASA-style motor cooling found that an optimized independent-coolant-pipeline design cut peak stator temperatures by roughly 37 to 53 percent compared to a baseline model, keeping temperatures below about 75 °C even under demanding operating conditions.3Case Studies in Thermal Engineering. Design of the water-cooling system for the stator assembly of YASA motors Those numbers matter because permanent magnets begin to lose their magnetization if they get too hot, and the stator windings can fail if insulation temperature limits are exceeded. Effective cooling is what separates a motor that performs well on a test bench for a few minutes from one that can sustain high power in a real vehicle or aircraft.
Bearing Forces and Structural Stiffness
Axial-flux motors generate large axial forces, the kind that try to pull the rotor and stator together like two strong magnets facing each other. In a radial-flux motor, these forces are balanced radially and mostly cancel out, so standard deep-groove ball bearings work fine. In an axial-flux machine, the net axial pull can be substantial, particularly in single-rotor designs where there is no opposing rotor to balance the force.
Handling these forces typically requires angular-contact or tapered-roller bearings rated for high axial loads. One engineering analysis used a single-row angular-contact ball bearing with a static thrust rating of 15 kN to handle the unidirectional axial forces in a prototype machine.4IET Electric Power Applications. Axial Force Negative Stiffness in Axial‐Flux Electric Machines Beyond the bearings themselves, the housing and rotor mounting need enough stiffness to prevent bending under those forces. If the structure flexes even slightly, the air gap changes, and in an axial-flux motor the air gap is already thin. That same analysis noted that axial-flux machines often use a tape-wound laminated construction for the stator and rotor, which inherently has limited bending stiffness, so the surrounding mechanical assembly has to compensate.5IET Electric Power Applications. Axial Force Negative Stiffness in Axial‐Flux Electric Machines
This is one reason you hear less about axial-flux motors in heavy industrial applications like compressors and conveyor drives. The mechanical engineering required to maintain tight air-gap tolerances under high loads adds cost and complexity. For a factory compressor running 24/7, the simpler bearing arrangement of a radial-flux motor is often the more practical choice.
Cogging Torque and How Designers Minimize It
Cogging torque is the jerky, pulsing resistance you sometimes feel when you turn a motor shaft slowly by hand. It comes from the permanent magnets on the rotor snapping toward the nearest stator tooth, and it creates vibration, noise, and rough low-speed performance. Axial-flux motors are susceptible to cogging because their disc geometry means the magnets sweep past stator segments at a relatively small radius, and the interaction can be strong.
Magnet shaping is the most direct way to reduce cogging. A comparison of several magnet geometries in an axial-flux permanent-magnet machine found that a combined arc-top and skewed magnet shape reduced cogging torque by about 54 percent compared to a conventional flat-top trapezoidal magnet, and further optimization brought the reduction to around 57 percent.6Results in Engineering. Improved performance of single sided axial flux for reduction in cogging torque (IMPACT) The important finding is that these cogging-reduction techniques did not sacrifice useful output torque; they just cleaned up the undesirable pulsation.
Noise reduction follows a similar path. One analytical approach combined varying pole-arc coefficients (essentially changing how wide each magnet pole is relative to the stator slot pitch) with circumferential shifting of magnets around the rotor disc. The result was a meaningful drop in acoustic noise without losing torque output.7IET Electric Power Applications. Noise reduction of axial‐flux motors by combining various pole‐arc coefficients and circumferential shifting of permanent magnets: analytical approach For consumer-facing products like electric cars, where cabin noise is a selling point, these design refinements matter as much as raw power numbers.
Magnet Arrangements That Boost Performance
The layout of permanent magnets on the rotor disc makes a surprisingly large difference in how well an axial-flux motor performs. A standard arrangement places magnets with alternating north-south polarity in a simple ring. A Halbach array, by contrast, arranges the magnets so that the field on one side of the array is reinforced while the field on the opposite side is largely canceled. This concentrates useful magnetic flux toward the stator and reduces leakage flux on the back side of the rotor, which means a thinner or even nonexistent rotor yoke (back-iron) can be used, saving weight.
A dual-skewed Halbach-array design for a double-sided axial-flux motor aimed at electric vehicles takes this further. By skewing the Halbach magnets in two directions, the design improves flux focusing and minimizes losses by directing the magnetic field more precisely toward the stator windings.8Scientific Reports. Design optimization of a novel dual-skewed Halbach-array double-sided axial flux permanent magnet motor for electric vehicles Removing or thinning the rotor yoke thanks to the Halbach effect also reduces the rotor’s moment of inertia, which improves dynamic response, meaning the motor can speed up and slow down faster, a useful trait in a vehicle drivetrain.
Printed Circuit Board Stators
One of the more unconventional developments in axial-flux motor design is replacing traditional copper wire windings with printed circuit boards. Instead of winding copper wire around stator teeth, designers etch flat copper traces onto PCB layers and stack multiple layers together to form the stator. The axial-flux geometry lends itself to this because the stator is a flat disc, which matches the flat board naturally.
The main advantage is manufacturing. PCB fabrication is precise, fast, and cheap at scale, making PCB axial-flux machines a viable alternative to conventional wire-wound machines for certain applications.9IET Electric Power Applications. Comparison of PCB winding topologies for axial‐flux permanent magnet synchronous machines The trade-off is that copper traces on a PCB are thin compared to round wire, so the current-carrying capacity per layer is limited. Designers compensate by stacking many layers, sometimes a dozen or more, but the motor still tends to be lower power than a wire-wound equivalent of the same size. PCB stators show up most often in smaller motors, drones, cooling fans, and compact robotics joints, where ease of manufacturing and ultra-thin profiles outweigh the power limitations.
There is also a quality-control benefit. Wire-wound stators have some variability from unit to unit because the winding process, even when automated, introduces slight differences in how tightly and evenly the wire is packed. PCB stators are identical from one board to the next, which means more consistent motor performance across a production run.
Direct-Drive Wind Turbines
Wind turbines present an interesting case for radial-versus-axial-flux comparisons. Most large wind turbines use a gearbox to step up the slow blade rotation (typically 10 to 20 rpm) to a speed that a conventional generator can handle efficiently. Gearboxes are heavy, expensive, and a common failure point in offshore turbines where maintenance is costly. Direct-drive generators eliminate the gearbox entirely, but they need to produce enormous torque at very low speed, which means a very large, heavy generator.
Axial-flux generators are one path to making direct drive more practical. A study demonstrated a 1 MW direct-drive axial-flux permanent-magnet generator using a novel flat-wire winding technique that achieved a rated torque of 480 kNm at a weight of 16 tons.10IET Renewable Power Generation. Axial flux generator with novel flat wire for direct‐drive wind turbines For context, that torque-to-weight ratio is competitive with much larger radial-flux direct-drive generators. The flat-wire approach also improves the copper fill factor (how much of the available winding space is actually filled with conductor), which boosts efficiency compared to round-wire designs.
Most of the world’s installed direct-drive wind turbines still use radial-flux generators, largely because the supply chain and manufacturing processes are more mature. But axial-flux prototypes keep pushing the performance envelope, and as offshore wind installations demand lighter, more reliable generators, the axial-flux approach has a plausible path to wider adoption.
Common Misconceptions About the Two Topologies
A frequent claim online is that axial-flux motors are “always more efficient” than radial-flux ones. This oversimplifies things. Efficiency depends on the specific operating point, the quality of the magnets, the winding design, the cooling system, and dozens of other engineering choices. A well-designed radial-flux motor can match or exceed the efficiency of a mediocre axial-flux motor. Where axial-flux machines genuinely tend to win is in torque density and power density at a given weight, not necessarily in peak electrical efficiency.
Another misconception is that axial-flux motors are a new invention. Faraday’s original motor experiments in the 1820s actually used an axial-flux-like geometry. The cylindrical radial-flux design became dominant later because it was easier to manufacture with 19th- and 20th-century tooling. The recent resurgence of axial-flux machines is driven not by a new discovery but by better permanent magnets (especially high-energy neodymium magnets), better manufacturing techniques like PCB stators and precision machining, and application demands, such as electric aviation, that put an extreme premium on weight savings.
A third misunderstanding is that axial-flux motors cannot be scaled to high power levels. While early commercial examples were mostly small, the wind-turbine work mentioned earlier shows machines at the megawatt scale, and several companies are developing axial-flux traction motors for full-size passenger cars and commercial vehicles. The scaling challenges are real, particularly around thermal management and structural stiffness, but they are engineering problems with known solutions, not fundamental physical limits.
How Magnet Supply Affects Both Designs
Both radial-flux and axial-flux permanent-magnet motors depend on rare-earth magnets, primarily neodymium-iron-boron (NdFeB). The supply chain for these magnets is concentrated in a handful of countries, and price fluctuations can significantly affect motor cost. Axial-flux motors sometimes use more magnet material per kilowatt than radial-flux machines because the disc geometry places magnets across a larger surface area, though the total motor weight is often lower.
This dependence has spurred research into ferrite-magnet axial-flux designs, which use cheaper, more abundant magnets at the cost of lower magnetic strength. Ferrite-based axial-flux motors tend to be larger for the same power output, partially offsetting the compactness advantage that makes axial flux attractive in the first place. Some designers are exploring hybrid approaches that use a small amount of rare-earth material in combination with ferrite or with reluctance torque (torque produced by the shape of the rotor rather than by magnets alone) to reduce rare-earth dependence without giving up too much performance.
For consumers, the practical takeaway is that the cost premium sometimes associated with axial-flux motors is not purely about the motor topology itself. It also reflects magnet costs, lower production volumes, and the fact that the manufacturing supply chain for axial-flux components is less mature than for radial-flux ones. As production scales up and magnet alternatives improve, that gap is likely to narrow.