← Darsh Patel

Rare-earth-free joint motors: feasibility study

EMPE Lab · Feasibility study · FEMM 4.2 and Elmer · simulation only AI-drivenFeasibility study

On this page

This feasibility study asks whether a rare-earth-free joint motor can reach 70% of the NdFeB baseline's continuous torque per active mass. None of the rare-earth-free designs meets that target. The best ferrite spoke rotor reaches 49% in 2D and 43% after the Elmer 3D end-leakage check. All results are simulated; no motor has been built.

The radial comparison holds the active envelope at 91 mm diameter, an 8 mm stack and 36 stator slots on a 48 V bus. Every design gets the same 17.8 W copper loss, set by the NdFeB baseline at an assumed 1.0 N·m continuous torque. The 70% target was chosen before the sweep.

I also checked axial-flux layouts and sized complete actuators for MABEL's arm and torso joints. The best axial ferrite design reaches 32% of the radial baseline, and axial NdFeB reaches 95%. In the main 64:1 actuator case, ferrite fails the mass criterion at all four joints. The sections below separate the fixed-envelope motor comparison, coupled actuator sizing and final per-joint 3D checks.

Humanoid upper body ↗ · code ↗

Six motor cross-sections in a grid: an NdFeB surface-magnet outer rotor, inner and outer ferrite spoke rotors, a ferrite-assisted reluctance rotor, a reluctance rotor and a hybrid rotor.
The six rotor designs in the same envelope: NdFeB in red, ferrite in blue, arrows for magnetisation direction.
Cutaway 3D renders of the six motors, showing laminations, coloured magnets, copper coils and shafts.
3D models built from the same geometry FEMM solves, with one quadrant cut away.

Results across all designs

88 designs were swept, plus 219 hybrid pairings. The table gives the best design in each family at the same 17.8 W copper loss, with magnets at 20 °C. The verdict holds from −20 °C to 100 °C: the best spoke reaches 50% of the baseline when cold and 45% when hot.

FamilyContinuous torqueTorque per active massvs NdFeBPeak at 4× current
NdFeB surface magnet (baseline)1.00 N·m4.86 N·m/kg100%3.92 N·m
Ferrite spoke, inner rotor0.67 N·m2.37 N·m/kg49%2.28 N·m
Ferrite spoke, outer rotor0.39 N·m1.97 N·m/kg41%1.62 N·m
Hybrid (ferrite spoke + reluctance)0.49 N·m1.65 N·m/kg34%1.71 N·m
Ferrite-assisted reluctance0.18 N·m0.51 N·m/kg10%1.01 N·m
Reluctance, no magnets0.06 N·m0.11 N·m/kg2%0.78 N·m
Best design per family against the baseline, with the 70% line.
Best design per family against the baseline, with the 70% line.
Every design in the sweep: continuous torque per active mass against peak torque.
Every design in the sweep: continuous torque per active mass against peak torque.
The five finalists on six metrics.
The five finalists on six metrics.
Flux density at rated torque for all five finalists, on one 0 to 2.2 T scale.
Flux density at rated torque for all five finalists, on one 0 to 2.2 T scale.
Torque ripple, back-EMF distortion and PWM current ripple for each finalist.
Torque ripple, back-EMF distortion and PWM current ripple for each finalist.
Rated torque against magnet temperature. Ferrite loses torque about 70% faster than NdFeB as it heats.
Rated torque against magnet temperature. Ferrite loses torque about 70% faster than NdFeB as it heats.

NdFeB surface-magnet outer rotor (baseline)

The baseline is a 42-pole outer rotor with 2 mm N42 arc magnets over a 36-slot tooth-coil stator, the usual layout for this class of joint motor. It makes 1.0 N·m at 17.8 W, with 26.9 g of magnet and 4.86 N·m/kg of active mass. At 4× current and 100 °C it keeps a 0.06 T margin above the demagnetisation knee.

It is the smoothest machine here: 1.4% torque ripple and 1.1% cogging. Its weak point is for the drive, not the motor. With only 8 turns per coil its inductance is about 65 µH, so PWM current ripple at 20 kHz is about 30% of rated current. About 60 kHz brings that under 10%.

3D model of the baseline: outer rotor can with NdFeB arc magnets (red) over the tooth-coil stator.
3D model of the baseline: outer rotor can with NdFeB arc magnets (red) over the tooth-coil stator.
Flux density at rated torque. The stator teeth carry about 1.2 to 1.5 T.
Flux density at rated torque. The stator teeth carry about 1.2 to 1.5 T.
Cogging torque over one cogging period: 11.3 mN·m peak to peak, 1.1% of rated torque.
Cogging torque over one cogging period: 11.3 mN·m peak to peak, 1.1% of rated torque.
Torque against rotor position at rated current: 1.4% ripple, mostly the 12th electrical order from the 36 slots.
Torque against rotor position at rated current: 1.4% ripple, mostly the 12th electrical order from the 36 slots.
No-load back-EMF and its harmonics. The phase distortion is almost all third harmonic, which cancels line to line.
No-load back-EMF and its harmonics. The phase distortion is almost all third harmonic, which cancels line to line.
Torque against current angle. Maximum torque per amp sits at zero angle, so all the torque comes from the magnets.
Torque against current angle. Maximum torque per amp sits at zero angle, so all the torque comes from the magnets.
PWM current ripple against switching frequency at 48 V. The low inductance needs a fast switching drive.
PWM current ripple against switching frequency at 48 V. The low inductance needs a fast switching drive.
Rated torque against magnet temperature: about −0.12% per °C, so 91% of its 20 °C torque at 100 °C.
Rated torque against magnet temperature: about −0.12% per °C, so 91% of its 20 °C torque at 100 °C.
Elmer 3D no-load field of the baseline, 120° cut away, flux density on the cut faces. Peak 2.43 T.
Elmer 3D no-load field of the baseline, 120° cut away, flux density on the cut faces. Peak 2.43 T.
The same cutaway with flux density painted on every surface.
The same cutaway with flux density painted on every surface.
The complete baseline with flux density on its outside surfaces.
The complete baseline with flux density on its outside surfaces.
Baseline magnets alone and iron alone, painted with flux density.
Baseline magnets alone and iron alone, painted with flux density.
Baseline saturation map: coloured only where flux density is above 1.5 T. Only small spots at the tooth tips.
Baseline saturation map: coloured only where flux density is above 1.5 T. Only small spots at the tooth tips.
Baseline field lines through one pole pitch. Lines looping above and below the stack are end leakage.
Baseline field lines through one pole pitch. Lines looping above and below the stack are end leakage.
Baseline, 2D equivalent against the real 3D stack. The gap field matches at mid-stack and falls to 0.55 of it at the end face.
Baseline, 2D equivalent against the real 3D stack. The gap field matches at mid-stack and falls to 0.55 of it at the end face.
Animated orbit around a cutaway of the NdFeB outer-rotor motor with flux density painted on its surfaces.
Camera orbit around the painted baseline cutaway.

Ferrite spoke rotors

A spoke rotor stands ferrite magnets on edge between iron pole pieces, so the flux of two magnet faces adds into each pole. This is the strongest way to use ferrite, whose remanence is 0.40 T against 1.30 T for NdFeB. I swept 36 spoke designs, inner and outer rotor, from 8 to 42 poles. 24 to 30 poles works best: 8 poles collapses, and at 42 poles the narrow spokes leak too much flux.

The best design is a 30-pole inner rotor with 14 mm deep spokes. It makes 0.67 N·m and 2.37 N·m/kg, 49% of the baseline in 2D and 43% after the 3D end-leakage correction. It also has the lowest rotor inertia here, 77 against 131 kg·mm². Matching the baseline's torque would need a stack of about 15.8 mm instead of 8 mm and 2.6 times the active mass.

Its real limit is demagnetisation. Ferrite is weakest cold. At −20 °C about 4% of the magnet area sits below the knee even with no current, at the spoke corners, and current-driven damage starts spreading between 2× and 2.5× rated. So its usable peak is about 2× rated, not 4×.

The outer spoke rotor drops into the baseline's own stator. It reaches 41%, because the same stator carries only half the flux it does in the baseline.

3D model of the best design: 30-pole ferrite spoke inner rotor (blue magnets) inside the tooth-coil stator.
3D model of the best design: 30-pole ferrite spoke inner rotor (blue magnets) inside the tooth-coil stator.
Inner spoke rotor: flux density at rated torque.
Inner spoke rotor: flux density at rated torque.
Spoke torque per active mass by pole count. 24 to 30 poles is the best range.
Spoke torque per active mass by pole count. 24 to 30 poles is the best range.
Inner spoke at −20 °C: overload torque and the share of magnet area below the knee. Damage spreads past about 2× rated.
Inner spoke at −20 °C: overload torque and the share of magnet area below the knee. Damage spreads past about 2× rated.
Flux density along the magnetisation across the magnet area at 4× current, against the knee.
Flux density along the magnetisation across the magnet area at 4× current, against the knee.
Stack length and active mass the inner spoke needs to match the baseline, with the 3D end loss included: about 15.8 mm and 2.6 times the mass.
Stack length and active mass the inner spoke needs to match the baseline, with the 3D end loss included: about 15.8 mm and 2.6 times the mass.
Inner spoke cogging: 17.1 mN·m peak to peak, 2.6% of its rated torque. Grey: the baseline.
Inner spoke cogging: 17.1 mN·m peak to peak, 2.6% of its rated torque. Grey: the baseline.
Inner spoke torque against rotor position at rated current: 3.4% ripple.
Inner spoke torque against rotor position at rated current: 3.4% ripple.
Inner spoke no-load back-EMF and its harmonics: 1.2% phase distortion.
Inner spoke no-load back-EMF and its harmonics: 1.2% phase distortion.
Inner spoke torque against current angle. Like the baseline, it is almost all magnet torque.
Inner spoke torque against current angle. Like the baseline, it is almost all magnet torque.
Inner spoke PWM current ripple against switching frequency: about 10% of rated current at 20 kHz.
Inner spoke PWM current ripple against switching frequency: about 10% of rated current at 20 kHz.
Inner spoke rated torque against magnet temperature: about −0.20% per °C, so 84% of its 20 °C torque at 100 °C.
Inner spoke rated torque against magnet temperature: about −0.20% per °C, so 84% of its 20 °C torque at 100 °C.
Elmer 3D no-load field of the inner spoke, same cut and colour scale as the baseline. Peak 2.0 T.
Elmer 3D no-load field of the inner spoke, same cut and colour scale as the baseline. Peak 2.0 T.
The same cutaway with flux density painted on every surface.
The same cutaway with flux density painted on every surface.
The complete inner spoke with flux density on its outside surfaces.
The complete inner spoke with flux density on its outside surfaces.
Inner spoke magnets alone and iron alone, painted with flux density. The ferrite runs at well under 0.5 T.
Inner spoke magnets alone and iron alone, painted with flux density. The ferrite runs at well under 0.5 T.
Inner spoke saturation map: almost nothing exceeds 1.5 T, only small spots at the rotor bridges and tooth tips.
Inner spoke saturation map: almost nothing exceeds 1.5 T, only small spots at the rotor bridges and tooth tips.
Inner spoke field lines through one pole pitch. Lines that loop out above the rotor are end leakage.
Inner spoke field lines through one pole pitch. Lines that loop out above the rotor are end leakage.
Inner spoke, 2D equivalent against the real 3D stack. The gap field is 0.80 of the 2D value all the way up the stack.
Inner spoke, 2D equivalent against the real 3D stack. The gap field is 0.80 of the 2D value all the way up the stack.
Animated orbit around a cutaway of the ferrite spoke inner-rotor motor with flux density painted on its surfaces.
Camera orbit around the painted inner spoke cutaway.
3D model of the outer spoke rotor on the baseline's stator.
3D model of the outer spoke rotor on the baseline's stator.
Outer spoke rotor: flux density at rated torque. The stator teeth run at about half the baseline's flux.
Outer spoke rotor: flux density at rated torque. The stator teeth run at about half the baseline's flux.
Outer spoke cogging: 8.0 mN·m peak to peak.
Outer spoke cogging: 8.0 mN·m peak to peak.
Outer spoke torque against rotor position at rated current: 3.0% ripple.
Outer spoke torque against rotor position at rated current: 3.0% ripple.
Outer spoke no-load back-EMF and its harmonics.
Outer spoke no-load back-EMF and its harmonics.
Outer spoke torque against current angle.
Outer spoke torque against current angle.
Outer spoke PWM current ripple against switching frequency.
Outer spoke PWM current ripple against switching frequency.
Outer spoke rated torque against magnet temperature.
Outer spoke rated torque against magnet temperature.

Ferrite-assisted reluctance rotor

This rotor uses flux barriers to make reluctance torque and puts ferrite in the barriers to add magnet torque. The best of 26 designs, a 24-pole rotor with tooth coils, reaches 0.18 N·m: 10% of the baseline. At this light loading the 0.4 mm ribs at the air gap saturate and take most of the reluctance torque with them, so at rated current all its torque comes from the magnets.

It needs a 20° current angle at rated, has a power factor of 0.64 and ripples 13% at rated and 41% at 4×.

3D model of the 24-pole ferrite-assisted reluctance rotor.
3D model of the 24-pole ferrite-assisted reluctance rotor.
Flux density at rated torque. The ribs at the air gap saturate at about 2 T while the rotor body carries little.
Flux density at rated torque. The ribs at the air gap saturate at about 2 T while the rotor body carries little.
Torque against current angle. The best angle moves from 20° at rated to 32° at 4×.
Torque against current angle. The best angle moves from 20° at rated to 32° at 4×.
Cogging: 6.7 mN·m peak to peak, 3.7% of its rated torque.
Cogging: 6.7 mN·m peak to peak, 3.7% of its rated torque.
Torque against rotor position at rated current: 13% ripple, mostly the 6th electrical order.
Torque against rotor position at rated current: 13% ripple, mostly the 6th electrical order.
No-load back-EMF and its harmonics.
No-load back-EMF and its harmonics.
PWM current ripple against switching frequency.
PWM current ripple against switching frequency.
Rated torque against magnet temperature: about −0.31% per °C.
Rated torque against magnet temperature: about −0.31% per °C.

Reluctance rotor, no magnets

A synchronous reluctance rotor has no magnets at all. The best of 26 designs, a 6-pole rotor with two barriers per pole and a distributed winding, reaches 0.056 N·m: 2% of the baseline. Reluctance torque grows with the square of current, and 17.8 W on an 8 mm stack is a light loading. At 4× current it recovers to 0.78 N·m.

Low pole counts also need distributed windings, whose end turns make the mean turn 4 times longer than a tooth coil's. Its power factor is 0.37 and it ripples 42% at rated without skew, so it does not suit a small joint motor.

3D model of the 6-pole reluctance rotor with its distributed winding.
3D model of the 6-pole reluctance rotor with its distributed winding.
Flux density at rated torque.
Flux density at rated torque.
Torque against current angle. The best angle is 46° at rated and 54° at 4×.
Torque against current angle. The best angle is 46° at rated and 54° at 4×.
Torque against rotor position at rated current: 42% ripple, mostly the 12th electrical order.
Torque against rotor position at rated current: 42% ripple, mostly the 12th electrical order.
PWM current ripple against switching frequency. Its high inductance keeps ripple near 1% of rated current.
PWM current ripple against switching frequency. Its high inductance keeps ripple near 1% of rated current.

Hybrid rotor

The hybrid stacks a ferrite spoke segment and a reluctance segment on one shaft, offset so their best current angles line up. Its torque per mass is a weighted mean of its two halves, so it can never beat the better one. The best of 219 pairings is 75% spoke and reaches 34% of the baseline; its 24-pole reluctance segment adds almost nothing. It does use less magnet and has less cogging.

3D model of the best hybrid: 75% ferrite spoke and 25% reluctance along the stack. The top is the reluctance segment, and the spokes show on the cut faces.
3D model of the best hybrid: 75% ferrite spoke and 25% reluctance along the stack. The top is the reluctance segment, and the spokes show on the cut faces.
Hybrid torque per active mass against the share of stack given to the spoke segment. The best is 1.65 N·m/kg at 75% spoke.
Hybrid torque per active mass against the share of stack given to the spoke segment. The best is 1.65 N·m/kg at 75% spoke.

3D field views

These views come from the Elmer 3D no-load solves used for the end-leakage check. Each solve covers a 60° sector and half the stack, and the full machine is rebuilt from it using the sector's anti-periodic symmetry and the mid-stack mirror. The field is from the magnets alone at one rotor position. Each machine's own 3D views are in its section above.

The 3D model shows where the spoke loses its torque. Its magnets push a broad ring of flux out of the rotor ends, and its gap field is about 20% below the 2D model even at mid-stack. The baseline's gap field matches the 2D model at mid-stack and only falls off near the end faces.

Field 1 mm above the stack and on a side cut through one magnet, on one scale. The baseline leaks only at its magnet ring; the spoke leaks over the whole rotor. The 12% and 89% count flux leaving the end face, not lost flux linkage, which is 2.1% and 13.9%.
Field 1 mm above the stack and on a side cut through one magnet, on one scale. The baseline leaks only at its magnet ring; the spoke leaks over the whole rotor. The 12% and 89% count flux leaving the end face, not lost flux linkage, which is 2.1% and 13.9%.
The same views, each divided by its own machine's mid-stack gap field.
The same views, each divided by its own machine's mid-stack gap field.
Radial, tangential and axial field components in the gap. The radial machines are mostly radial; the axial-flux machine is 96% axial.
Radial, tangential and axial field components in the gap. The radial machines are mostly radial; the axial-flux machine is 96% axial.
Direction of the gap flux as unit arrows for the two radial machines and the axial-flux machine.
Direction of the gap flux as unit arrows for the two radial machines and the axial-flux machine.

Axial flux

Could an axial-flux layout change the verdict? I swept 99 axial designs in the same package: 91 mm across, 11.4 mm long (the radial machine's 8 mm stack plus its end turns) and the same 17.8 W of copper loss. The families were an NdFeB yokeless and segmented armature (YASA) double rotor, an NdFeB single-sided machine, ferrite surface-magnet and ferrite spoke double rotors, and a magnet-free reluctance rotor. Each design is solved as five FEMM slices at different radii, run on the home cluster, and the best designs were checked in Elmer 3D.

Axial flux makes ferrite worse here. The best axial ferrite design reaches 32% of the baseline's torque per kilogram, against 43% for the radial spoke after its 3D correction. In an 11.4 mm package every millimetre of magnet comes out of the stator, so the thinnest magnets win and the torque stays near 0.27 N·m. The radial spoke's deep magnets cost no stack length.

Axial NdFeB does not raise the bar either. The YASA reaches 95% of the radial baseline and the single-sided machine 88%, so the 70% line stays tied to the radial baseline. The axial machines lose little at their edges in 3D: 0.993 for the NdFeB YASA and 0.984 for the best ferrite design, against 0.861 for the radial spoke.

Axial pull becomes a design load. At peak current each NdFeB YASA rotor is pulled towards the stator with 558 N. A double rotor cancels that only when centred, and off centre it is pushed further with about 1.8 N per µm. A single-sided machine puts 711 N on its bearing. Ferrite pulls about 75 N. Widening the air gap from 0.5 mm to 0.75 mm costs 20% of the torque, so the gap has to be held tightly against that pull.

The 3D views below are the NdFeB YASA design with 36 slots and 42 poles used for the Elmer check, not the best YASA in the sweep. Its coils are not modelled, so none are drawn.

Best design per family, radial and axial, against the 70% line. Neither set has its 3D correction here.
Best design per family, radial and axial, against the 70% line. Neither set has its 3D correction here.
All 99 axial designs: peak torque against continuous torque per kilogram, with the two radial reference machines.
All 99 axial designs: peak torque against continuous torque per kilogram, with the two radial reference machines.
Axial pull on one rotor disc at peak current, best design per family.
Axial pull on one rotor disc at peak current, best design per family.
Axial-flux YASA, 120° cut away, flux density on the cut faces. Peak 2.19 T.
Axial-flux YASA, 120° cut away, flux density on the cut faces. Peak 2.19 T.
The same cutaway with flux density painted on every surface.
The same cutaway with flux density painted on every surface.
The complete axial-flux machine with flux density on its outside surfaces.
The complete axial-flux machine with flux density on its outside surfaces.
Axial-flux magnets alone and iron alone, painted with flux density.
Axial-flux magnets alone and iron alone, painted with flux density.
Axial-flux saturation map: the rotor back iron runs above 1.5 T between magnets.
Axial-flux saturation map: the rotor back iron runs above 1.5 T between magnets.
Axial-flux field lines through one pole pitch, crossing both gaps along the shaft.
Axial-flux field lines through one pole pitch, crossing both gaps along the shaft.
Animated orbit around a cutaway of the NdFeB axial-flux YASA motor with flux density painted on its surfaces.
Camera orbit around the painted axial-flux cutaway.

Method and checks

Every model is the full 360° machine in FEMM 4.2, run headless on my home Linux server. Torque at each current comes from a search over current angle. Three torque methods agree within 0.3% on the baseline, and torque moves 0.02% between 100,000 and 500,000 mesh elements. For the 3D check, Elmer solves a 60° sector of the baseline and the best spoke. A 2D-equivalent version reproduces FEMM's flux linkage within 0.6%. The finite 8 mm stack then gives the end-leakage factor: 0.979 for the baseline and 0.861 for the spoke. Laminations, the far boundary and a finer axial mesh each move it by 0.3% or less.

The stator and magnet dimensions are estimates of the WGG8108, not measurements. The 2D models leave out spoke bridges and iron losses, which would both lower the ferrite numbers further.

Three torque methods across four mesh sizes on the baseline.
Three torque methods across four mesh sizes on the baseline.
PWM current ripple against switching frequency for all finalists.
PWM current ripple against switching frequency for all finalists.

Per-joint actuators on MABEL

The sections above use one stator size. This section asks the same question for the joints of the MABEL robot, with a separate actuator for each joint. Arm and torso actuators are sized per joint from a library of 145 FEMM motor designs: NdFeB surface-magnet outer rotors against ferrite spoke inner rotors. Each actuator is a motor, a gearbox, a housing and my 48 V board stack, chosen together. Each set is sized again with its own masses and gearbox inertia, so a heavier shoulder actuator raises the torso load. The study calls this coupled sizing. The main case limits the gearbox to a two-stage planetary of at most 64:1. The lift, head and hands are stock MABEL parts. The swerve drive in the base was not changed in any variant, because it was not in the design scope of this feasibility study.

Before any results, the pass line was set at ferrite weighing at most 1.43 times the NdFeB actuator, which is the same 70% of NdFeB torque per kilogram used above. This is a judgement line from the original brief, chosen before the results. It is not computed. In the main case ferrite fails it at all four joints, with ratios of 1.44 to 1.49. One arm's actuators weigh 3.68 kg with NdFeB and 5.38 kg with ferrite.

The break-even cutoff is the line at which a joint would flip, and for the main case it is 67% to 69%. The same picks pass at a 60% or 65% line and fail at 70%, 75% and 80%. The ratio depends on how much gearing is allowed, as the next section shows. The whole-robot MuJoCo check of these sets is on the humanoid upper body page.

What stock means. In the mass and size comparisons, stock is MABEL's own DAMIAO actuators. In the MuJoCo and Simulink comparisons, stock is the GIM reference set (GIM8115-36, GIM8108-36 and GIM4310-40), because the electrical data of MABEL's DAMIAO actuators is not modelled.

Cogging warning. The mass-only optimiser chose 36-slot, 24-pole ferrite spokes for most of the ferrite arm joints, including J1, J3 and J5 here. Those motors have a cogging torque of 20% to 38% of their continuous torque and a torque ripple of up to 45%, against 3.5% to 7.6% for 30-pole spokes. The library sweep did not compute cogging, so the optimiser could not see it. A low-cogging ferrite set with no 24-pole motors weighs about 1% more (5.45 kg instead of 5.38 kg for one arm in the main case) and gives the same verdict at every joint. The new 3D flux overlays at the end of this page use the low-cogging picks. The 24-pole ferrite plots here retain the original mass-only result; they are not the recommended design.

The per-pick Elmer 3D no-load solves are complete. Painted field views, field lines and the finite-stack end-leakage checks are at the end of this page. Torque against current angle, PWM ripple, temperature and cold demagnetisation are separate loaded-motor checks.

Jump to solved 3D flux overlays ↗

JointNdFeB pick (mass, size)Ferrite pick (mass, size)Ferrite / NdFeBBreak-even cutoff
Shoulder J10.78 kg, 97 × 36 mm1.15 kg, 108 × 40 mm1.49 fail67%
Upper arm J30.55 kg, 86 × 32 mm0.79 kg, 95 × 35 mm1.44 fail69%
Wrist J50.35 kg, 56 × 40 mm0.51 kg, 64 × 41 mm1.46 fail69%
Torso pitch1.16 kg, 110 × 42 mm1.71 kg, 124 × 45 mm1.48 fail68%

Every pick above is a two-stage planetary at 64:1. J2 repeats J1 and J4 repeats J3. The picks for J6 and J7 differ slightly from J5. Sizes are the outer diameter and length of the whole actuator.

Rotor and stator only, main picks, to scale. NdFeB outer rotors in the top row, ferrite spoke inner rotors in the bottom row. Stator laminations are grey, windings copper, NdFeB magnets red and ferrite magnets blue.
Rotor and stator only, main picks, to scale. NdFeB outer rotors in the top row, ferrite spoke inner rotors in the bottom row. Stator laminations are grey, windings copper, NdFeB magnets red and ferrite magnets blue.
CAD cutaways of the main per-joint picks, with the stock MABEL actuators as plain outlines, all to one scale. Each cutaway shows the motor, gearbox and board stack inside its housing.
CAD cutaways of the main per-joint picks, with the stock MABEL actuators as plain outlines, all to one scale. Each cutaway shows the motor, gearbox and board stack inside its housing.
The 64:1 optimisation, from the coupled sizing: actuator mass per joint under four constraints, NdFeB against ferrite spoke, with the ferrite to NdFeB ratio above each pair. The lightest case with the gear ratio up to 64:1 is the headline, and ferrite fails the 1.43 line at all four joints there (1.49, 1.44, 1.46 and 1.48). Ferrite only gets near the line with more gearing, and it fails clearly once the ratio and reflected inertia are held to stock values or the drive is quasi-direct.
The 64:1 optimisation, from the coupled sizing: actuator mass per joint under four constraints, NdFeB against ferrite spoke, with the ferrite to NdFeB ratio above each pair. The lightest case with the gear ratio up to 64:1 is the headline, and ferrite fails the 1.43 line at all four joints there (1.49, 1.44, 1.46 and 1.48). Ferrite only gets near the line with more gearing, and it fails clearly once the ratio and reflected inertia are held to stock values or the drive is quasi-direct.
Ferrite to NdFeB mass ratio against the highest gear ratio allowed, with every gearbox type capped together, from the coupled sizing. At the 64:1 cap all four joints fail. Ferrite passes only with more gearing: at a 100:1 cap J3, J5 and the torso pass and J1 does not. This differs from the mass_100 row in the tables, where only the planetary gearbox is allowed up to 100:1.
Ferrite to NdFeB mass ratio against the highest gear ratio allowed, with every gearbox type capped together, from the coupled sizing. At the 64:1 cap all four joints fail. Ferrite passes only with more gearing: at a 100:1 cap J3, J5 and the torso pass and J1 does not. This differs from the mass_100 row in the tables, where only the planetary gearbox is allowed up to 100:1.
The lightest actuator that meets each limit on reflected inertia, with all gearboxes up to 64:1, from the coupled sizing. The stock actuator is marked. A lighter actuator costs reflected inertia, and the ferrite curve sits above and to the right of the NdFeB curve at every joint.
The lightest actuator that meets each limit on reflected inertia, with all gearboxes up to 64:1, from the coupled sizing. The stock actuator is marked. A lighter actuator costs reflected inertia, and the ferrite curve sits above and to the right of the NdFeB curve at every joint.
Torso duty cycle used to size the torso motor: a 60° bend and a return of 1 s each, with a brake hold in between and 3 kg per hand. The motor limits are 1.25 times the move peak (211 N·m) and 1.25 times the RMS over the 4 s cycle (96 N·m). The brake carries the static hold.
Torso duty cycle used to size the torso motor: a 60° bend and a return of 1 s each, with a brake hold in between and 3 kg per hand. The motor limits are 1.25 times the move peak (211 N·m) and 1.25 times the RMS over the 4 s cycle (96 N·m). The brake carries the static hold.
Every motor the per-joint study picked across the 18 coupled sets, to scale, by topology and stator diameter. Blank cells were not picked by any set.
Every motor the per-joint study picked across the 18 coupled sets, to scale, by topology and stator diameter. Blank cells were not picked by any set.

Ferrite to NdFeB mass ratio in every case

The ratios come from the coupled sizing, and a pass is 1.43 or less. The mass_100 case is the only one where ferrite passes at all four joints, but the whole-robot check on the humanoid page fails that ferrite set. The backdrive and QDD cases limit the gear ratio and the reflected inertia, which is where the ferrite actuators grow the most. In the mass_50 case the ferrite J3 and J5 picks switch to a 160:1 strain wave gearbox, so that comparison is a strain wave against a planetary. A torso entry reads ferrite infeasible when no ferrite actuator can meet that set's own torso constraint.

CaseJ1J3J5Torso
main (64:1)1.49 fail1.44 fail1.46 fail1.48 fail
mass_641.49 fail1.44 fail1.46 fail1.48 fail
mass_501.52 fail1.45 fail1.36 pass1.52 fail
mass_1001.40 pass1.42 pass1.40 pass1.36 pass
mass_64_reva661.48 fail1.44 fail1.35 pass1.47 fail
size1.51 fail1.44 fail1.50 failferrite infeasible
backdrive2.80 fail2.42 fail2.21 failferrite infeasible
backdrive_modelref2.66 fail2.14 fail2.21 failferrite infeasible
qdd1.84 fail1.65 fail1.64 failferrite infeasible

Machines for the shoulder J1

J1, and J2 with the same pick. NdFeB: 91 mm motor with an 8 mm stack, 48 slots and 56 poles, in a 97 × 36 mm actuator of 0.78 kg. Ferrite: 104 mm motor with a 10 mm stack, 36 slots and 24 poles in a spoke inner rotor, in a 108 × 40 mm actuator of 1.15 kg. The ferrite spoke has a cogging torque of 37.7% of its continuous torque and a torque ripple of 44.6%, against 1.6% for both in the NdFeB motor. Iron loss is a stator estimate.

Shoulder J1, NdFeB: CAD cutaway of the pick, 97 × 36 mm, 0.78 kg, with the two-stage 64:1 planetary and the board stack. First quadrant cut away.
Shoulder J1, NdFeB: CAD cutaway of the pick, 97 × 36 mm, 0.78 kg, with the two-stage 64:1 planetary and the board stack. First quadrant cut away.
Shoulder J1, NdFeB: Flux density at the continuous operating point, solved in FEMM. 99.5th percentile |B| 1.63 T at 13.0 A/mm².
Shoulder J1, NdFeB: Flux density at the continuous operating point, solved in FEMM. 99.5th percentile |B| 1.63 T at 13.0 A/mm².
Shoulder J1, NdFeB: Cogging torque over 120 electrical degrees, at the motor. Cogging 14.3 mN·m peak to peak at the motor, 1.6% of its continuous torque.
Shoulder J1, NdFeB: Cogging torque over 120 electrical degrees, at the motor. Cogging 14.3 mN·m peak to peak at the motor, 1.6% of its continuous torque.
Shoulder J1, NdFeB: Torque ripple at the continuous and peak points. Torque ripple 1.6% peak to peak at continuous, 1.3% at peak.
Shoulder J1, NdFeB: Torque ripple at the continuous and peak points. Torque ripple 1.6% peak to peak at continuous, 1.3% at peak.
Shoulder J1, NdFeB: No-load back-EMF waveform and its harmonics. Fundamental 11.36 V peak per phase at 96 rad/s; THD 5.7% phase, 0.1% line to line.
Shoulder J1, NdFeB: No-load back-EMF waveform and its harmonics. Fundamental 11.36 V peak per phase at 96 rad/s; THD 5.7% phase, 0.1% line to line.
Shoulder J1, NdFeB: Output torque against phase current through the gearbox. Motor kt 0.178 N·m/A at low current; saturation bends the curve above that.
Shoulder J1, NdFeB: Output torque against phase current through the gearbox. Motor kt 0.178 N·m/A at low current; saturation bends the curve above that.
Shoulder J1, NdFeB: Losses at the continuous duty point. Efficiency 75% including the stator iron-loss estimate (76% without it).
Shoulder J1, NdFeB: Losses at the continuous duty point. Efficiency 75% including the stator iron-loss estimate (76% without it).
Shoulder J1, ferrite spoke: CAD cutaway of the pick, 108 × 40 mm, 1.15 kg, with the two-stage 64:1 planetary and the board stack. First quadrant cut away.
Shoulder J1, ferrite spoke: CAD cutaway of the pick, 108 × 40 mm, 1.15 kg, with the two-stage 64:1 planetary and the board stack. First quadrant cut away.
Shoulder J1, ferrite spoke: Flux density at the continuous operating point, solved in FEMM. 99.5th percentile |B| 1.42 T at 8.8 A/mm².
Shoulder J1, ferrite spoke: Flux density at the continuous operating point, solved in FEMM. 99.5th percentile |B| 1.42 T at 8.8 A/mm².
Shoulder J1, ferrite spoke: Cogging torque over 120 electrical degrees, at the motor. Cogging 362.4 mN·m peak to peak at the motor, 37.7% of its continuous torque.
Shoulder J1, ferrite spoke: Cogging torque over 120 electrical degrees, at the motor. Cogging 362.4 mN·m peak to peak at the motor, 37.7% of its continuous torque.
Shoulder J1, ferrite spoke: Torque ripple at the continuous and peak points. Torque ripple 44.6% peak to peak at continuous, 37.9% at peak.
Shoulder J1, ferrite spoke: Torque ripple at the continuous and peak points. Torque ripple 44.6% peak to peak at continuous, 37.9% at peak.
Shoulder J1, ferrite spoke: No-load back-EMF waveform and its harmonics. Fundamental 10.98 V peak per phase at 96 rad/s; THD 7.5% phase, 7.5% line to line.
Shoulder J1, ferrite spoke: No-load back-EMF waveform and its harmonics. Fundamental 10.98 V peak per phase at 96 rad/s; THD 7.5% phase, 7.5% line to line.
Shoulder J1, ferrite spoke: Output torque against phase current through the gearbox. Motor kt 0.171 N·m/A at low current; saturation bends the curve above that.
Shoulder J1, ferrite spoke: Output torque against phase current through the gearbox. Motor kt 0.171 N·m/A at low current; saturation bends the curve above that.
Shoulder J1, ferrite spoke: Losses at the continuous duty point. Efficiency 74% including the stator iron-loss estimate (74% without it).
Shoulder J1, ferrite spoke: Losses at the continuous duty point. Efficiency 74% including the stator iron-loss estimate (74% without it).

Machines for the upper arm J3

J3, and J4 with the same pick. NdFeB: 80 mm motor with a 6 mm stack, 48 slots and 56 poles, in an 86 × 32 mm actuator of 0.55 kg. Ferrite: 91 mm motor with an 8 mm stack, 36 slots and 24 poles, in a 95 × 35 mm actuator of 0.79 kg.

Upper arm J3, NdFeB: CAD cutaway of the pick, 86 × 32 mm, 0.55 kg, with the two-stage 64:1 planetary and the board stack. First quadrant cut away.
Upper arm J3, NdFeB: CAD cutaway of the pick, 86 × 32 mm, 0.55 kg, with the two-stage 64:1 planetary and the board stack. First quadrant cut away.
Upper arm J3, NdFeB: Flux density at the continuous operating point, solved in FEMM. 99.5th percentile |B| 1.65 T at 15.0 A/mm².
Upper arm J3, NdFeB: Flux density at the continuous operating point, solved in FEMM. 99.5th percentile |B| 1.65 T at 15.0 A/mm².
Upper arm J3, NdFeB: Cogging torque over 120 electrical degrees, at the motor. Cogging 4.8 mN·m peak to peak at the motor, 1.0% of its continuous torque.
Upper arm J3, NdFeB: Cogging torque over 120 electrical degrees, at the motor. Cogging 4.8 mN·m peak to peak at the motor, 1.0% of its continuous torque.
Upper arm J3, NdFeB: Torque ripple at the continuous and peak points. Torque ripple 1.0% peak to peak at continuous, 0.7% at peak.
Upper arm J3, NdFeB: Torque ripple at the continuous and peak points. Torque ripple 1.0% peak to peak at continuous, 0.7% at peak.
Upper arm J3, NdFeB: No-load back-EMF waveform and its harmonics. Fundamental 10.76 V peak per phase at 96 rad/s; THD 5.0% phase, 0.1% line to line.
Upper arm J3, NdFeB: No-load back-EMF waveform and its harmonics. Fundamental 10.76 V peak per phase at 96 rad/s; THD 5.0% phase, 0.1% line to line.
Upper arm J3, NdFeB: Output torque against phase current through the gearbox. Motor kt 0.168 N·m/A at low current; saturation bends the curve above that.
Upper arm J3, NdFeB: Output torque against phase current through the gearbox. Motor kt 0.168 N·m/A at low current; saturation bends the curve above that.
Upper arm J3, NdFeB: Losses at the continuous duty point. Efficiency 71% including the stator iron-loss estimate (71% without it).
Upper arm J3, NdFeB: Losses at the continuous duty point. Efficiency 71% including the stator iron-loss estimate (71% without it).
Upper arm J3, ferrite spoke: CAD cutaway of the pick, 95 × 35 mm, 0.79 kg, with the two-stage 64:1 planetary and the board stack. First quadrant cut away.
Upper arm J3, ferrite spoke: CAD cutaway of the pick, 95 × 35 mm, 0.79 kg, with the two-stage 64:1 planetary and the board stack. First quadrant cut away.
Upper arm J3, ferrite spoke: Flux density at the continuous operating point, solved in FEMM. 99.5th percentile |B| 1.33 T at 10.0 A/mm².
Upper arm J3, ferrite spoke: Flux density at the continuous operating point, solved in FEMM. 99.5th percentile |B| 1.33 T at 10.0 A/mm².
Upper arm J3, ferrite spoke: Cogging torque over 120 electrical degrees, at the motor. Cogging 177.1 mN·m peak to peak at the motor, 34.1% of its continuous torque.
Upper arm J3, ferrite spoke: Cogging torque over 120 electrical degrees, at the motor. Cogging 177.1 mN·m peak to peak at the motor, 34.1% of its continuous torque.
Upper arm J3, ferrite spoke: Torque ripple at the continuous and peak points. Torque ripple 40.2% peak to peak at continuous, 33.0% at peak.
Upper arm J3, ferrite spoke: Torque ripple at the continuous and peak points. Torque ripple 40.2% peak to peak at continuous, 33.0% at peak.
Upper arm J3, ferrite spoke: No-load back-EMF waveform and its harmonics. Fundamental 10.17 V peak per phase at 96 rad/s; THD 7.1% phase, 7.1% line to line.
Upper arm J3, ferrite spoke: No-load back-EMF waveform and its harmonics. Fundamental 10.17 V peak per phase at 96 rad/s; THD 7.1% phase, 7.1% line to line.
Upper arm J3, ferrite spoke: Output torque against phase current through the gearbox. Motor kt 0.159 N·m/A at low current; saturation bends the curve above that.
Upper arm J3, ferrite spoke: Output torque against phase current through the gearbox. Motor kt 0.159 N·m/A at low current; saturation bends the curve above that.
Upper arm J3, ferrite spoke: Losses at the continuous duty point. Efficiency 68% including the stator iron-loss estimate (69% without it).
Upper arm J3, ferrite spoke: Losses at the continuous duty point. Efficiency 68% including the stator iron-loss estimate (69% without it).

Machines for the wrist J5

J5. NdFeB: 50 mm motor, 24 slots and 28 poles, in a 56 × 40 mm actuator of 0.35 kg with a 13 mm stack. Ferrite: 60 mm motor, 36 slots and 24 poles, in a 64 × 41 mm actuator of 0.51 kg with a 15 mm stack. These charts are for the motor cross-section at one stack length, 12 mm for NdFeB and 6 mm for ferrite, which differ from the stacks of the picks. The flux map, torque ripple and back-EMF shapes hardly depend on the stack. The cogging torque is scaled to the drawn stack. The torque-current and loss charts of the ferrite wrist motor are left out, because they were drawn for a different gearbox and stack than the pick.

Wrist J5, NdFeB: CAD cutaway of the pick, 56 × 40 mm, 0.35 kg, with the two-stage 64:1 planetary and the board stack. First quadrant cut away.
Wrist J5, NdFeB: CAD cutaway of the pick, 56 × 40 mm, 0.35 kg, with the two-stage 64:1 planetary and the board stack. First quadrant cut away.
Wrist J5, NdFeB: Flux density at the continuous operating point, solved in FEMM. 99.5th percentile |B| 1.59 T at 15.3 A/mm². Drawn at a 12 mm stack, the pick uses 13 mm.
Wrist J5, NdFeB: Flux density at the continuous operating point, solved in FEMM. 99.5th percentile |B| 1.59 T at 15.3 A/mm². Drawn at a 12 mm stack, the pick uses 13 mm.
Wrist J5, NdFeB: Cogging torque over 120 electrical degrees, at the motor. Cogging 1.7 mN·m peak to peak at the motor, 0.8% of its continuous torque. Drawn at a 12 mm stack, the pick uses 13 mm.
Wrist J5, NdFeB: Cogging torque over 120 electrical degrees, at the motor. Cogging 1.7 mN·m peak to peak at the motor, 0.8% of its continuous torque. Drawn at a 12 mm stack, the pick uses 13 mm.
Wrist J5, NdFeB: Torque ripple at the continuous and peak points. Torque ripple 0.8% peak to peak at continuous, 0.6% at peak. Drawn at a 12 mm stack, the pick uses 13 mm.
Wrist J5, NdFeB: Torque ripple at the continuous and peak points. Torque ripple 0.8% peak to peak at continuous, 0.6% at peak. Drawn at a 12 mm stack, the pick uses 13 mm.
Wrist J5, NdFeB: No-load back-EMF waveform and its harmonics. Fundamental 10.48 V peak per phase at 96 rad/s; THD 7.7% phase, 0.2% line to line. Drawn at a 12 mm stack, the pick uses 13 mm.
Wrist J5, NdFeB: No-load back-EMF waveform and its harmonics. Fundamental 10.48 V peak per phase at 96 rad/s; THD 7.7% phase, 0.2% line to line. Drawn at a 12 mm stack, the pick uses 13 mm.
Wrist J5, NdFeB: Output torque against phase current through the gearbox. Motor kt 0.164 N·m/A at low current; saturation bends the curve above that. Drawn at a 12 mm stack, the pick uses 13 mm.
Wrist J5, NdFeB: Output torque against phase current through the gearbox. Motor kt 0.164 N·m/A at low current; saturation bends the curve above that. Drawn at a 12 mm stack, the pick uses 13 mm.
Wrist J5, NdFeB: Losses at the continuous duty point. Efficiency 67% including the stator iron-loss estimate (67% without it). Drawn at a 12 mm stack, the pick uses 13 mm.
Wrist J5, NdFeB: Losses at the continuous duty point. Efficiency 67% including the stator iron-loss estimate (67% without it). Drawn at a 12 mm stack, the pick uses 13 mm.
Wrist J5, ferrite spoke: CAD cutaway of the pick, 64 × 41 mm, 0.51 kg, with the two-stage 64:1 planetary and the board stack. First quadrant cut away.
Wrist J5, ferrite spoke: CAD cutaway of the pick, 64 × 41 mm, 0.51 kg, with the two-stage 64:1 planetary and the board stack. First quadrant cut away.
Wrist J5, ferrite spoke: Flux density at the continuous operating point, solved in FEMM. 99.5th percentile |B| 1.05 T at 13.6 A/mm². Drawn at a 6 mm stack, the pick uses 15 mm.
Wrist J5, ferrite spoke: Flux density at the continuous operating point, solved in FEMM. 99.5th percentile |B| 1.05 T at 13.6 A/mm². Drawn at a 6 mm stack, the pick uses 15 mm.
Wrist J5, ferrite spoke: Cogging torque over 120 electrical degrees, at the motor. Cogging 23.8 mN·m peak to peak at the motor, 22.2% of its continuous torque. Drawn at a 6 mm stack, the pick uses 15 mm.
Wrist J5, ferrite spoke: Cogging torque over 120 electrical degrees, at the motor. Cogging 23.8 mN·m peak to peak at the motor, 22.2% of its continuous torque. Drawn at a 6 mm stack, the pick uses 15 mm.
Wrist J5, ferrite spoke: Torque ripple at the continuous and peak points. Torque ripple 25.9% peak to peak at continuous, 20.1% at peak. Drawn at a 6 mm stack, the pick uses 15 mm.
Wrist J5, ferrite spoke: Torque ripple at the continuous and peak points. Torque ripple 25.9% peak to peak at continuous, 20.1% at peak. Drawn at a 6 mm stack, the pick uses 15 mm.
Wrist J5, ferrite spoke: No-load back-EMF waveform and its harmonics. Fundamental 9.69 V peak per phase at 240 rad/s; THD 4.8% phase, 4.8% line to line. Drawn at a 6 mm stack, the pick uses 15 mm.
Wrist J5, ferrite spoke: No-load back-EMF waveform and its harmonics. Fundamental 9.69 V peak per phase at 240 rad/s; THD 4.8% phase, 4.8% line to line. Drawn at a 6 mm stack, the pick uses 15 mm.

Machines for the torso pitch

Torso pitch. NdFeB: 104 mm motor with an 11 mm stack, 48 slots and 56 poles, in a 110 × 42 mm actuator of 1.16 kg. Ferrite: 120 mm motor with a 13 mm stack, 36 slots and 30 poles, in a 124 × 45 mm actuator of 1.71 kg. The torso holds its pose on a brake, so the motor is sized for motion only. The torque-current and loss charts of the NdFeB torso motor are left out, because they were drawn for a 36:1 gearbox and the pick is 64:1.

Torso pitch, NdFeB: CAD cutaway of the pick, 110 × 42 mm, 1.16 kg, with the two-stage 64:1 planetary and the board stack. First quadrant cut away.
Torso pitch, NdFeB: CAD cutaway of the pick, 110 × 42 mm, 1.16 kg, with the two-stage 64:1 planetary and the board stack. First quadrant cut away.
Torso pitch, NdFeB: Flux density at the continuous operating point, solved in FEMM. 99.5th percentile |B| 1.65 T at 11.4 A/mm².
Torso pitch, NdFeB: Flux density at the continuous operating point, solved in FEMM. 99.5th percentile |B| 1.65 T at 11.4 A/mm².
Torso pitch, NdFeB: Cogging torque over 120 electrical degrees, at the motor. Cogging 31.3 mN·m peak to peak at the motor, 1.9% of its continuous torque.
Torso pitch, NdFeB: Cogging torque over 120 electrical degrees, at the motor. Cogging 31.3 mN·m peak to peak at the motor, 1.9% of its continuous torque.
Torso pitch, NdFeB: Torque ripple at the continuous and peak points. Torque ripple 1.8% peak to peak at continuous, 0.9% at peak.
Torso pitch, NdFeB: Torque ripple at the continuous and peak points. Torque ripple 1.8% peak to peak at continuous, 0.9% at peak.
Torso pitch, NdFeB: No-load back-EMF waveform and its harmonics. Fundamental 11.14 V peak per phase at 54 rad/s; THD 6.5% phase, 0.1% line to line.
Torso pitch, NdFeB: No-load back-EMF waveform and its harmonics. Fundamental 11.14 V peak per phase at 54 rad/s; THD 6.5% phase, 0.1% line to line.
Torso pitch, ferrite spoke: CAD cutaway of the pick, 124 × 45 mm, 1.71 kg, with the two-stage 64:1 planetary and the board stack. First quadrant cut away.
Torso pitch, ferrite spoke: CAD cutaway of the pick, 124 × 45 mm, 1.71 kg, with the two-stage 64:1 planetary and the board stack. First quadrant cut away.
Torso pitch, ferrite spoke: Flux density at the continuous operating point, solved in FEMM. 99.5th percentile |B| 1.32 T at 7.8 A/mm².
Torso pitch, ferrite spoke: Flux density at the continuous operating point, solved in FEMM. 99.5th percentile |B| 1.32 T at 7.8 A/mm².
Torso pitch, ferrite spoke: Cogging torque over 120 electrical degrees, at the motor. Cogging 73.3 mN·m peak to peak at the motor, 4.5% of its continuous torque.
Torso pitch, ferrite spoke: Cogging torque over 120 electrical degrees, at the motor. Cogging 73.3 mN·m peak to peak at the motor, 4.5% of its continuous torque.
Torso pitch, ferrite spoke: Torque ripple at the continuous and peak points. Torque ripple 5.8% peak to peak at continuous, 3.9% at peak.
Torso pitch, ferrite spoke: Torque ripple at the continuous and peak points. Torque ripple 5.8% peak to peak at continuous, 3.9% at peak.
Torso pitch, ferrite spoke: No-load back-EMF waveform and its harmonics. Fundamental 8.17 V peak per phase at 64 rad/s; THD 3.8% phase, 1.0% line to line.
Torso pitch, ferrite spoke: No-load back-EMF waveform and its harmonics. Fundamental 8.17 V peak per phase at 64 rad/s; THD 3.8% phase, 1.0% line to line.
Torso pitch, ferrite spoke: Output torque against phase current through the gearbox. Motor kt 0.191 N·m/A at low current; saturation bends the curve above that.
Torso pitch, ferrite spoke: Output torque against phase current through the gearbox. Motor kt 0.191 N·m/A at low current; saturation bends the curve above that.
Torso pitch, ferrite spoke: Losses at the continuous duty point. Efficiency 72% including the stator iron-loss estimate (73% without it).
Torso pitch, ferrite spoke: Losses at the continuous duty point. Efficiency 72% including the stator iron-loss estimate (73% without it).

Material cost estimate

A rough bill of materials for the 15 motors of one robot (seven joints in each arm and the torso). These are estimates for materials only, not quotes. A built motor costs far more than its materials at low volume, and manufacturing drives that cost, not the magnets. The assumed prices are NdFeB N42 at $60 to $120/kg, ferrite at $3 to $10/kg, magnet wire at $12 to $18/kg, M-19 steel at $2 to $4/kg and 1010 steel at $1 to $2/kg.

In the main case ferrite needs more than twice the active material and three times the magnet mass, but its magnets are far cheaper, so the total materials cost about the same. In the backdrivable and quasi-direct-drive sets the ferrite motors grow so much that their materials cost more.

Item (15 motors)NdFeBFerrite spoke
Active material, main set2.0 kg4.5 kg
Magnet mass, main set0.27 kg0.83 kg
Magnet cost, main set$16 to $33$2 to $8
Total material, main set$23 to $44$19 to $36
Total material, backdrivable set$36 to $67$64 to $119
Total material, quasi-direct-drive set$67 to $125$74 to $141

Electronics and limits of the per-joint study

Both optimised variants use my own electronics architecture: 48 V DRV8363 gate drivers with a 100 V FET stage, an STM32G474 on the logic board, and the support and brake boards. Only the REVA power stage is laid out in KiCad. The other boards are simulated, not laid out, so a pick that needs a different board outline is a simulated board.

Iron loss is only a stator estimate. Cogging torque is not in the backdrive torque. There is no thermal FEM: the temperature limit uses one heat-transfer coefficient calibrated on stock actuators. Gearbox flex is one stiffness number. The MuJoCo check welds the base. These are screening results, not qualification. All of this is simulation, and nothing has been built.

Solved 3D flux overlays for the joint motors

All 16 Elmer solves are complete: a 2D-equivalent slab and a finite-stack model for each of eight motors, at its picked stack length. The slab flux linkage agrees with FEMM to about 1%. The table gives finite-stack flux linkage divided by slab flux linkage. Equivalent end loss is the stack length multiplied by the fraction lost.

These new views compare the main NdFeB picks with the low-cogging ferrite picks, which all have 30 poles. That ferrite set weighs 5.45 kg per arm against 3.68 kg for NdFeB and still fails the 70% mass criterion at every joint. The earlier 24-pole arm plots above document the original mass-only optimisation, whose cogging was too high.

The colours are the solved magnetic flux density mapped onto the motor surfaces and cut faces. Each full machine is reconstructed from its solved sector and the mid-stack mirror. These are no-load fields from the permanent magnets, with zero winding current, at one rotor position. They show end leakage; loaded torque, heating and demagnetisation need their own checks. The gearbox, housing and controller boards are outside these field models.

NdFeB retains 96.9% to 98.7% of its slab flux; ferrite retains 86.1% to 90.6%. Ferrite's equivalent end loss grows to 1.44 mm at the torso, compared with the 1.11 mm assumed in sizing. This is a check on that assumption, not a new coupled optimisation with the measured losses substituted.

Field-image provenance and solve checks (JSON) ↗

Joint and motorMotor diameter × stackSlots / polesFlux retainedEquivalent end loss
Shoulder J1, NdFeB91 × 8 mm48 / 5697.7%0.19 mm
Shoulder J1, low-cogging ferrite104 × 10 mm36 / 3087.1%1.29 mm
Upper arm J3, NdFeB80 × 6 mm48 / 5696.9%0.19 mm
Upper arm J3, low-cogging ferrite91 × 8 mm36 / 3086.1%1.11 mm
Wrist J5, NdFeB50 × 13 mm24 / 2898.7%0.16 mm
Wrist J5, low-cogging ferrite70 × 10 mm36 / 3090.6%0.94 mm
Torso pitch, NdFeB104 × 11 mm48 / 5698.5%0.16 mm
Torso pitch, low-cogging ferrite120 × 13 mm36 / 3088.9%1.44 mm

Shoulder J1: solved 3D fields

Shoulder J1, NdFeB, 91 mm motor, 8 mm stack. Cutaway and close-up, with flux density painted on the surfaces and cut faces. All painted views use the same 0 to 2.2 T scale; peaks above it are stated in the image.
Shoulder J1, NdFeB, 91 mm motor, 8 mm stack. Cutaway and close-up, with flux density painted on the surfaces and cut faces. All painted views use the same 0 to 2.2 T scale; peaks above it are stated in the image.
Shoulder J1, NdFeB, 91 mm motor, 8 mm stack. Complete motor with flux density painted on its outside surfaces, at the same no-load operating point.
Shoulder J1, NdFeB, 91 mm motor, 8 mm stack. Complete motor with flux density painted on its outside surfaces, at the same no-load operating point.
Shoulder J1, NdFeB, 91 mm motor, 8 mm stack. Field lines through one pole pitch, seeded at the middle of the air gap, from the saved no-load field.
Shoulder J1, NdFeB, 91 mm motor, 8 mm stack. Field lines through one pole pitch, seeded at the middle of the air gap, from the saved no-load field.
Shoulder J1, NdFeB, 91 mm motor, 8 mm stack. 2D-equivalent slab against the finite stack: field through a magnet and radial air-gap flux density along the stack.
Shoulder J1, NdFeB, 91 mm motor, 8 mm stack. 2D-equivalent slab against the finite stack: field through a magnet and radial air-gap flux density along the stack.
Shoulder J1, low-cogging ferrite, 104 mm motor, 10 mm stack. Cutaway and close-up, with flux density painted on the surfaces and cut faces. All painted views use the same 0 to 2.2 T scale; peaks above it are stated in the image.
Shoulder J1, low-cogging ferrite, 104 mm motor, 10 mm stack. Cutaway and close-up, with flux density painted on the surfaces and cut faces. All painted views use the same 0 to 2.2 T scale; peaks above it are stated in the image.
Shoulder J1, low-cogging ferrite, 104 mm motor, 10 mm stack. Complete motor with flux density painted on its outside surfaces, at the same no-load operating point.
Shoulder J1, low-cogging ferrite, 104 mm motor, 10 mm stack. Complete motor with flux density painted on its outside surfaces, at the same no-load operating point.
Shoulder J1, low-cogging ferrite, 104 mm motor, 10 mm stack. Field lines through one pole pitch, seeded at the middle of the air gap, from the saved no-load field.
Shoulder J1, low-cogging ferrite, 104 mm motor, 10 mm stack. Field lines through one pole pitch, seeded at the middle of the air gap, from the saved no-load field.
Shoulder J1, low-cogging ferrite, 104 mm motor, 10 mm stack. 2D-equivalent slab against the finite stack: field through a magnet and radial air-gap flux density along the stack.
Shoulder J1, low-cogging ferrite, 104 mm motor, 10 mm stack. 2D-equivalent slab against the finite stack: field through a magnet and radial air-gap flux density along the stack.

Upper arm J3: solved 3D fields

Upper arm J3, NdFeB, 80 mm motor, 6 mm stack. Cutaway and close-up, with flux density painted on the surfaces and cut faces. All painted views use the same 0 to 2.2 T scale; peaks above it are stated in the image.
Upper arm J3, NdFeB, 80 mm motor, 6 mm stack. Cutaway and close-up, with flux density painted on the surfaces and cut faces. All painted views use the same 0 to 2.2 T scale; peaks above it are stated in the image.
Upper arm J3, NdFeB, 80 mm motor, 6 mm stack. Complete motor with flux density painted on its outside surfaces, at the same no-load operating point.
Upper arm J3, NdFeB, 80 mm motor, 6 mm stack. Complete motor with flux density painted on its outside surfaces, at the same no-load operating point.
Upper arm J3, NdFeB, 80 mm motor, 6 mm stack. Field lines through one pole pitch, seeded at the middle of the air gap, from the saved no-load field.
Upper arm J3, NdFeB, 80 mm motor, 6 mm stack. Field lines through one pole pitch, seeded at the middle of the air gap, from the saved no-load field.
Upper arm J3, NdFeB, 80 mm motor, 6 mm stack. 2D-equivalent slab against the finite stack: field through a magnet and radial air-gap flux density along the stack.
Upper arm J3, NdFeB, 80 mm motor, 6 mm stack. 2D-equivalent slab against the finite stack: field through a magnet and radial air-gap flux density along the stack.
Upper arm J3, low-cogging ferrite, 91 mm motor, 8 mm stack. Cutaway and close-up, with flux density painted on the surfaces and cut faces. All painted views use the same 0 to 2.2 T scale; peaks above it are stated in the image.
Upper arm J3, low-cogging ferrite, 91 mm motor, 8 mm stack. Cutaway and close-up, with flux density painted on the surfaces and cut faces. All painted views use the same 0 to 2.2 T scale; peaks above it are stated in the image.
Upper arm J3, low-cogging ferrite, 91 mm motor, 8 mm stack. Complete motor with flux density painted on its outside surfaces, at the same no-load operating point.
Upper arm J3, low-cogging ferrite, 91 mm motor, 8 mm stack. Complete motor with flux density painted on its outside surfaces, at the same no-load operating point.
Upper arm J3, low-cogging ferrite, 91 mm motor, 8 mm stack. Field lines through one pole pitch, seeded at the middle of the air gap, from the saved no-load field.
Upper arm J3, low-cogging ferrite, 91 mm motor, 8 mm stack. Field lines through one pole pitch, seeded at the middle of the air gap, from the saved no-load field.
Upper arm J3, low-cogging ferrite, 91 mm motor, 8 mm stack. 2D-equivalent slab against the finite stack: field through a magnet and radial air-gap flux density along the stack.
Upper arm J3, low-cogging ferrite, 91 mm motor, 8 mm stack. 2D-equivalent slab against the finite stack: field through a magnet and radial air-gap flux density along the stack.

Wrist J5: solved 3D fields

Wrist J5, NdFeB, 50 mm motor, 13 mm stack. Cutaway and close-up, with flux density painted on the surfaces and cut faces. All painted views use the same 0 to 2.2 T scale; peaks above it are stated in the image.
Wrist J5, NdFeB, 50 mm motor, 13 mm stack. Cutaway and close-up, with flux density painted on the surfaces and cut faces. All painted views use the same 0 to 2.2 T scale; peaks above it are stated in the image.
Wrist J5, NdFeB, 50 mm motor, 13 mm stack. Complete motor with flux density painted on its outside surfaces, at the same no-load operating point.
Wrist J5, NdFeB, 50 mm motor, 13 mm stack. Complete motor with flux density painted on its outside surfaces, at the same no-load operating point.
Wrist J5, NdFeB, 50 mm motor, 13 mm stack. Field lines through one pole pitch, seeded at the middle of the air gap, from the saved no-load field.
Wrist J5, NdFeB, 50 mm motor, 13 mm stack. Field lines through one pole pitch, seeded at the middle of the air gap, from the saved no-load field.
Wrist J5, NdFeB, 50 mm motor, 13 mm stack. 2D-equivalent slab against the finite stack: field through a magnet and radial air-gap flux density along the stack.
Wrist J5, NdFeB, 50 mm motor, 13 mm stack. 2D-equivalent slab against the finite stack: field through a magnet and radial air-gap flux density along the stack.
Wrist J5, low-cogging ferrite, 70 mm motor, 10 mm stack. Cutaway and close-up, with flux density painted on the surfaces and cut faces. All painted views use the same 0 to 2.2 T scale; peaks above it are stated in the image.
Wrist J5, low-cogging ferrite, 70 mm motor, 10 mm stack. Cutaway and close-up, with flux density painted on the surfaces and cut faces. All painted views use the same 0 to 2.2 T scale; peaks above it are stated in the image.
Wrist J5, low-cogging ferrite, 70 mm motor, 10 mm stack. Complete motor with flux density painted on its outside surfaces, at the same no-load operating point.
Wrist J5, low-cogging ferrite, 70 mm motor, 10 mm stack. Complete motor with flux density painted on its outside surfaces, at the same no-load operating point.
Wrist J5, low-cogging ferrite, 70 mm motor, 10 mm stack. Field lines through one pole pitch, seeded at the middle of the air gap, from the saved no-load field.
Wrist J5, low-cogging ferrite, 70 mm motor, 10 mm stack. Field lines through one pole pitch, seeded at the middle of the air gap, from the saved no-load field.
Wrist J5, low-cogging ferrite, 70 mm motor, 10 mm stack. 2D-equivalent slab against the finite stack: field through a magnet and radial air-gap flux density along the stack.
Wrist J5, low-cogging ferrite, 70 mm motor, 10 mm stack. 2D-equivalent slab against the finite stack: field through a magnet and radial air-gap flux density along the stack.

Torso pitch: solved 3D fields

Torso pitch, NdFeB, 104 mm motor, 11 mm stack. Cutaway and close-up, with flux density painted on the surfaces and cut faces. All painted views use the same 0 to 2.2 T scale; peaks above it are stated in the image.
Torso pitch, NdFeB, 104 mm motor, 11 mm stack. Cutaway and close-up, with flux density painted on the surfaces and cut faces. All painted views use the same 0 to 2.2 T scale; peaks above it are stated in the image.
Torso pitch, NdFeB, 104 mm motor, 11 mm stack. Complete motor with flux density painted on its outside surfaces, at the same no-load operating point.
Torso pitch, NdFeB, 104 mm motor, 11 mm stack. Complete motor with flux density painted on its outside surfaces, at the same no-load operating point.
Torso pitch, NdFeB, 104 mm motor, 11 mm stack. Field lines through one pole pitch, seeded at the middle of the air gap, from the saved no-load field.
Torso pitch, NdFeB, 104 mm motor, 11 mm stack. Field lines through one pole pitch, seeded at the middle of the air gap, from the saved no-load field.
Torso pitch, NdFeB, 104 mm motor, 11 mm stack. 2D-equivalent slab against the finite stack: field through a magnet and radial air-gap flux density along the stack.
Torso pitch, NdFeB, 104 mm motor, 11 mm stack. 2D-equivalent slab against the finite stack: field through a magnet and radial air-gap flux density along the stack.
Torso pitch, low-cogging ferrite, 120 mm motor, 13 mm stack. Cutaway and close-up, with flux density painted on the surfaces and cut faces. All painted views use the same 0 to 2.2 T scale; peaks above it are stated in the image.
Torso pitch, low-cogging ferrite, 120 mm motor, 13 mm stack. Cutaway and close-up, with flux density painted on the surfaces and cut faces. All painted views use the same 0 to 2.2 T scale; peaks above it are stated in the image.
Torso pitch, low-cogging ferrite, 120 mm motor, 13 mm stack. Complete motor with flux density painted on its outside surfaces, at the same no-load operating point.
Torso pitch, low-cogging ferrite, 120 mm motor, 13 mm stack. Complete motor with flux density painted on its outside surfaces, at the same no-load operating point.
Torso pitch, low-cogging ferrite, 120 mm motor, 13 mm stack. Field lines through one pole pitch, seeded at the middle of the air gap, from the saved no-load field.
Torso pitch, low-cogging ferrite, 120 mm motor, 13 mm stack. Field lines through one pole pitch, seeded at the middle of the air gap, from the saved no-load field.
Torso pitch, low-cogging ferrite, 120 mm motor, 13 mm stack. 2D-equivalent slab against the finite stack: field through a magnet and radial air-gap flux density along the stack.
Torso pitch, low-cogging ferrite, 120 mm motor, 13 mm stack. 2D-equivalent slab against the finite stack: field through a magnet and radial air-gap flux density along the stack.