SmCo vs NdFeB in Motor Applications: When the Switch Is Justified
The short version: NdFeB is the right choice for most motors, and SmCo is the right choice for a specific subset of applications where temperature or environment makes NdFeB impractical. The problem is that the boundary between those two categories is blurry in most engineering conversations. Engineers sometimes specify SmCo for thermal reasons when N45EH would have worked, and sometimes specify N45EH when the operating conditions clearly call for SmCo. Neither mistake is free.
This article gives concrete switching criteria - not general guidelines, but specific numbers - for the most common motor application scenarios where the NdFeB-to-SmCo transition comes up.
1. The Core Performance Tradeoff
NdFeB and SmCo both produce high-energy permanent magnetic fields, but their material properties diverge at elevated temperature in ways that matter for motor design.
NdFeB (sintered, N-grade): Br ranges from 1.08 T (N35) to 1.46 T (N52) at room temperature. Br temperature coefficient is approximately −0.11 to −0.13%/°C - so an N45 magnet at 150°C has Br approximately 1.25 T instead of the room-temperature 1.34 T, a 7% reduction. Hcj temperature coefficient is −0.55 to −0.60%/°C, which is the critical number: an N45 with Hcj = 1380 kA/m at 20°C has Hcj ≈ 620 kA/m at 150°C. The knee drops to approximately 310 kA/m demagnetizing field - meaning the magnet is relatively easy to demagnetize at elevated temperature under armature reaction.
SmCo Grade 26 (Sm₂Co₁₇): Br approximately 1.02–1.05 T at room temperature. Br temperature coefficient −0.030 to −0.035%/°C - roughly 3× better than NdFeB. At 200°C, Br is approximately 0.99 T, barely changed. Hcj temperature coefficient −0.20 to −0.25%/°C. SmCo Grade 26 with Hcj = 1500 kA/m at 20°C retains approximately 1000 kA/m at 200°C. The knee at 200°C is around 500 kA/m - still providing substantial demagnetization resistance under armature reaction.
The tradeoff is clear: at room temperature, N45 outperforms SmCo Gr26 by approximately 30% in Br. At 150°C, the gap narrows to 15%. Above 200°C, SmCo is the stronger magnet by Br as well. The price premium for SmCo is 4–6× per kilogram for equivalent volume.
2. Temperature-Based Switching Criteria
The switching decision belongs at a specific combination of peak magnet temperature and peak demagnetizing field, not at an arbitrary temperature threshold. But for practical first-pass decisions, these criteria cover most motor applications:
Below 120°C peak magnet temperature: NdFeB H or SH grade is almost always the correct choice. SmCo has no performance advantage at these temperatures that justifies the cost premium. The only exception is corrosive environments (marine motors, chemical process drives) where NdFeB's coating vulnerability matters more than the thermal properties.
120–150°C peak magnet temperature: This is the contested zone. For distributed winding motors with peak-to-continuous torque ratios below 3×, N45SH or N42SH is adequate and costs 60–70% less than SmCo. For concentrated winding motors with peak torque ratios above 4×, or for motor designs with PC below 4 (thin magnets in large air gaps), SmCo starts to make engineering sense because the Hcj margin calculation starts to fail for NdFeB even with SH grades.
150–200°C peak magnet temperature: SmCo is the correct material choice for most applications. NdFeB EH grades (Hcj > 3000 kA/m at 20°C) can sometimes be stretched into this range for low-armature-reaction designs, but you're paying EH prices (roughly 2× H grade) for a magnet that is still operating near its limits. SmCo Gr26 at 200°C retains 95% of its room-temperature Br and has Hcj three times higher than N45EH at the same temperature. The cost difference at 150–200°C is much less decisive than it looks at room temperature, because the NdFeB option requires the most expensive EH grades.
Above 200°C: SmCo only. NdFeB maximum continuous operating temperature is 200°C for EH grades, and that's with minimal armature reaction. For turbine auxiliary motors, downhole drilling motors, and high-temperature industrial drives that see sustained 200–250°C operation, SmCo Gr28–30 is the only practical option from the sintered rare earth category.
3. Application-Specific Switching Points
Temperature is the main driver, but it's not the only one. These specific applications require additional consideration:
Aerospace motor drives (actuators, fuel pumps, ECS fans): The primary driver is often not temperature but radiation hardness and outgassing requirements. NdFeB coatings (Ni-Cu-Ni, epoxy) can outgas in vacuum and have limited radiation resistance. SmCo has no organic components and much lower outgassing - measured TML (Total Mass Loss) for vacuum-grade SmCo is below 0.1%, versus 0.3–0.8% for epoxy-coated NdFeB. For NASA or ESA qualification, SmCo is standard not because of temperature but because of space environment compatibility.
High-speed motors (above 50,000 RPM): At very high speeds, the centrifugal stress in the rotor magnet structure becomes the limiting factor. SmCo has slightly higher compressive strength than NdFeB (approximately 1100 MPa vs 900 MPa compressive strength), but the more important factor is that high-speed motors with surface-mount magnets typically use a carbon fiber or Inconel retaining sleeve that sees the centrifugal load. At 100,000 RPM with a 10mm radius rotor, the magnet surface velocity is 100 m/s and the sleeve stress is the design constraint - the magnet material properties are secondary. SmCo's advantage here is corrosion resistance in the high-frequency eddy current environment, not mechanical strength.
Industrial process motors in corrosive environments: NdFeB corrodes rapidly in environments with humidity above 85% RH combined with ionic contamination (salt spray, chlorine, acid mist). The Ni-Cu-Ni coating provides adequate protection for standard industrial environments (IP54–IP65) but fails in marine or chemical environments within 2–5 years. SmCo is intrinsically corrosion-resistant - no coating required for marine-grade IP67 motors. If the motor is sealed in epoxy or the rotor is separately sealed, NdFeB can work in these environments with adequate encapsulation, but SmCo removes the design risk.
Long-lifetime motors with temperature cycling: Repeated thermal cycling from cold start to operating temperature creates differential thermal expansion between the magnet and the bonding adhesive. NdFeB has a thermal expansion coefficient of approximately 5–8 × 10⁻⁶/°C (depending on magnetization direction). SmCo has approximately 10–12 × 10⁻⁶/°C - actually higher than NdFeB. This means SmCo is not inherently better for thermal cycling; the bonding adhesive selection and magnet geometry dominate the fatigue life in both cases.
4. What SmCo Does Not Fix
SmCo is sometimes specified as a solution to problems it does not solve:
Eddy current losses: SmCo electrical resistivity (approximately 0.8–1.0 μΩ·m) is similar to NdFeB (approximately 1.4–1.6 μΩ·m). SmCo is not significantly better for reducing eddy current losses in high-frequency field environments. Segmentation of the magnet (cutting in the axial or circumferential direction) is the correct solution for both materials.
Cogging torque: Cogging torque is determined by the magnet pole shape, slot geometry, and their interaction - not the magnet material. SmCo doesn't reduce cogging torque relative to NdFeB with the same geometry. Magnet skew, slot skew, or fractional slot windings are the correct tools.
Cost reduction when EH is already required: When the operating conditions require NdFeB EH grade, some engineers switch to SmCo expecting a lower cost. In volume (>500 pieces), NdFeB EH is typically 30–50% cheaper than SmCo Gr26 for equivalent energy product. SmCo makes sense above a temperature threshold for engineering reasons - not as a cost reduction strategy when EH grades already price-equivalent to SmCo.
5. Practical Transition Protocol for Motor Design
When a motor design is being evaluated for SmCo versus NdFeB, the decision sequence should be:
Step 1: Calculate peak magnet temperature as described in the previous article (combining stator heat conduction and magnet self-heating). If below 120°C with SH grade margin, stop - NdFeB SH is the answer.
Step 2: If 120–180°C, calculate the demagnetization safety margin for N45SH and N42EH specifically. If both provide adequate margin (combined thermal + armature reaction field below the knee by 20% or more), N45SH is the answer.
Step 3: If EH grade is required but peak temperature is below 160°C, get current pricing for both N40EH and SmCo Gr26 for the specific geometry. At this temperature range, the cost differential is often 15–25% in NdFeB's favor for standard sizes. Above 160°C, request SmCo Gr26 pricing - the cost crossover point moves consistently in SmCo's direction as temperature increases.
Step 4: Regardless of temperature, specify SmCo if the application is aerospace vacuum-rated, sustained marine immersion without full epoxy encapsulation, or continuous operation above 200°C.
HIMAGNET supplies both NdFeB (standard through EH grades) and SmCo (Gr18 through Gr32), and can provide performance data at application temperatures for direct comparison in your load line calculations. If you have a motor design sitting in the 130–160°C peak magnet temperature range and aren't sure which material is justified, send us the operating profile and we'll provide the calculation rather than a recommendation based on price.
