What Is Inside the Rotor Decides Everything Else
Every traction motor has essentially the same stator. The interesting choice — cost, efficiency, supply-chain exposure, controller complexity — is what spins inside it.
Magnets and Motors · Part 3 — The Machine · Chapter 7 · 19 min read
Every traction motor has essentially the same stator: three-phase windings in a laminated steel bore. The interesting choice — cost, efficiency, supply-chain exposure, controller complexity — is what spins inside it.
96–97 %
IPM peak efficiency
1–2 kg
NdFeB per IPM motor
2–4 pts
Induction’s efficiency penalty
65–75 %
Hairpin slot fill vs 40–45 % round wire
7.1 — Five rotors, one stator
The V-shaped magnet arrangement in the IPM rotor is doing two jobs at once: it holds the magnets against 20,000 rpm centrifugal load, and it deliberately creates the flux barriers that generate the reluctance torque of chapter 6.
7.2 — What each topology commits you to
| Topology | What it gives you | What it costs |
|---|---|---|
| IPM — interior permanent magnet | Peak efficiency 96–97 %, the highest torque density available, mechanical retention and reluctance torque from the same buried geometry, and a wide field-weakening range. The incumbent. | 1–2 kg of NdFeB per motor, with heavy rare earths for temperature. Back-EMF is always present, so an inverter fault at high speed produces uncontrolled generation and braking torque. |
| SPM — surface permanent magnet | Simple to model, near-sinusoidal back-EMF, very low torque ripple. Common in servos and industrial drives. | L_d ≈ L_q, so no reluctance torque at all and poor field weakening. Magnets need a retaining sleeve above modest speeds. Rarely used in traction. |
| Induction — IM | A shorted cage of aluminium or copper bars. Robust, cheap, no magnets, no supply exposure, and it free-wheels harmlessly when de-energised. | Rotor I²R loss exists by definition — the rotor has to carry current to work — so efficiency runs 2–4 points below IPM, worst at light load, which is most of a drive cycle. Rotor heat is hard to remove. |
| EESM — externally excited synchronous | The rotor is an electromagnet fed through slip rings or a rotating transformer, so field strength becomes an independent control variable: full flux for launch, weak flux at cruise. No back-EMF when de-excited, so no uncontrolled generation. Zero rare earths. | Rotor copper loss, brushes or a contactless exciter to maintain, and a more complex inverter. BMW and Renault ship it at scale, so the objections are engineering rather than fundamental. |
| SynRM and ferrite-assisted SynRM | Pure reluctance torque from a rotor of stamped flux barriers. No magnets, no rotor current, no rotor loss, and very cheap. Adding low-cost ferrite into the barriers lifts torque density substantially without touching a rare earth. | Low power factor drives up inverter VA rating, and torque ripple is high. The ferrite variant inherits ferrite’s cold-demagnetisation behaviour from chapter 5. |
| Axial flux | Flux runs parallel to the shaft in a pancake geometry. Because torque scales as D²L, a large-diameter short-axial machine is intrinsically torque-dense — 2–3× the N·m/kg of a comparable radial motor. | Large axial magnetic attraction forces demand stiff structures and precise gaps. Harder to manufacture and to cool. Genuinely excellent for in-wheel and space-constrained installations. |
7.3 — Why hairpins replaced round wire
Slot fill factor is the fraction of the slot occupied by copper rather than air, insulation and varnish. Randomly wound round wire reaches 40 to 45 per cent. Rectangular hairpin conductors — pre-formed U-shaped bars inserted and laser-welded — reach 65 to 75.
- •More copper in the same slot means lower resistance, which means less I²R loss and better continuous rating from the same volume.
- •It also gives a far better thermal path from conductor to stator iron, because rectangular bars sit flat against the slot wall instead of touching it at points.
Important
The catch appears at high speed. A large solid conductor in a slot suffers skin and proximity effects. At 800 Hz electrical the current crowds into the outer few millimetres of the bar — and a hairpin bar is a few millimetres. The AC resistance can be several times the DC value exactly where the motor spends its highway life.
7.3.1 — The tension is structural
The hairpin’s fill-factor win and its AC-loss penalty come from the same property: a large solid conductor. Chapter 8 works through the skin-depth arithmetic and the geometric mitigations.
Quick check: test yourself
1.An inverter faults at 15,000 rpm. What happens in an IPM motor that does not happen in an EESM?
Show answer
2.Induction motors are 2 to 4 points less efficient. Why is that worse than it sounds?
Show answer
3.Hairpin windings raise slot fill from 45 to 70 per cent. Why is that not a straightforward win?
Show answer
Chapter summary
- ✓The stator is essentially the same in every traction motor. The rotor choice decides cost, efficiency, supply exposure and controller complexity.
- ✓IPM leads on efficiency and torque density, and pays with 1 to 2 kg of NdFeB and permanent back-EMF that becomes uncontrolled generation in a fault.
- ✓EESM makes field strength a control variable and eliminates both rare earths and uncontrolled generation, at the cost of rotor copper loss and an exciter.
- ✓Ferrite-assisted synchronous reluctance is the most credible rare-earth-free path for cost-sensitive vehicles — and inherits ferrite’s cold-demagnetisation risk.
- ✓Hairpin windings raise slot fill from 45 to 70 per cent and improve the thermal path, but the same geometry causes AC loss at highway frequencies.
Frequently asked questions
Why is IPM the default traction motor?+
Because burying the magnets does three jobs at once: it gives peak efficiency of 96 to 97 per cent and the highest torque density available, it retains the magnets mechanically against centrifugal load, and it creates the flux barriers that generate reluctance torque. It also gives a wide field-weakening range. The costs are 1 to 2 kg of NdFeB per motor with heavy rare earths for temperature, and back-EMF that is always present — an inverter fault at high speed produces uncontrolled generation and braking torque.
What is EESM and why is it growing?+
An externally excited synchronous machine makes the rotor an electromagnet, fed through slip rings or a rotating transformer. Field strength becomes an independent control variable — full flux for launch torque, weak flux at cruise — and with no permanent magnets there is no back-EMF when de-excited, so no uncontrolled generation, and zero rare earth exposure. The costs are rotor copper loss, brushes or a contactless exciter to maintain, and a more complex inverter. BMW and Renault ship it at volume, so the objections are engineering rather than fundamental.
Why is induction less efficient than IPM?+
Because the rotor has to carry current to work, so rotor I²R loss exists by definition. Efficiency runs 2 to 4 points below IPM, and the gap is worst at light load — which is where a vehicle spends most of a drive cycle. Rotor heat is also hard to remove. In exchange you get no magnets, no supply exposure, mechanical robustness, and harmless free-wheeling when de-energised.
Why did hairpin windings replace round wire, and what is the catch?+
Slot fill factor. Randomly wound round wire reaches 40 to 45 per cent copper in the slot; rectangular hairpin bars, pre-formed and laser-welded, reach 65 to 75. More copper means lower resistance, less I²R loss and better continuous rating from the same volume, plus a far better thermal path because flat bars sit against the slot wall rather than touching at points. The catch is AC loss: at 800 Hz electrical the skin depth in copper is about 2.3 mm, which is the same order as a hairpin bar’s thickness, so current crowds to the surface exactly where the motor spends its highway life.
Reviewed by
Magnets and Motors is an original educational series on permanent magnets and electric machines. Datasheet ranges, temperature coefficients and efficiency figures are representative standard-condition literature values, not measurements of a specific product, and sources differ on several. Always verify against the specific magnet and lamination datasheets in use before making design, procurement or certification decisions.