A High-Efficiency Radial Flux Generator Using Finemet and Soft Magnetic Composite Materials: Performance and Techno-Economic Comparison with Conventional Aerospace Designs

The proposed design shown in Figure 1 represents a radial-flux permanent-magnet synchronous generator (RFPMG). The left panel illustrates the 3-dimensional isometric geometry, while the right panel presents the radial cross-sectional view of the machine. A solid shaft (grey) supports the inner rotor core (cyan/aqua-blue), over which surface-mounted permanent magnets (yellow) are uniformly placed to generate radial magnetic flux across the air gap. Surrounding the rotor, the stator structure (red) forms the main magnetic return path and provides mechanical rigidity. The stator tooth regions (green) guide magnetic flux towards the yoke and contain slot spaces intended for multi-phase armature windings (not shown in the schematic for clarity). This surface-mounted inner-rotor configuration ensures strong flux linkage, reduced eddy losses and high-speed electromagnetic stability. The compact geometry and material layout make the design suitable for next-generation applications, such as aerospace power units, high-density portable power systems and renewable energy conversion platforms (wind, hydrokinetic or micro-turbine-based), where efficiency, lightweight implementation and reliability are critical.

Figure 1.

3D and cross-sectional views of a radial flux proposed design generator.

The internal behaviour of the generator under load conditions is depicted by the per-phase equivalent circuit shown in Figure 2. This circuit elucidates the relationship between the generated electromotive force (E_ph), armature impedance (Z_a) and terminal voltage (V_t) during operation. The impedance Z_a comprises armature resistance (R_a), which accounts for copper loss and synchronous reactance (X_s), representing the combined effects of leakage reactance (X_L) and armature reaction (X_a). These parameters directly influence voltage regulation, current magnitude and the power angle (δ), all of which govern power transfer capability and steady-state stability. This equivalent circuit forms the analytical foundation for evaluating voltage regulation, load dynamics, power factor and overall efficiency in high-speed aerospace-grade synchronous generators, where compactness, thermal reliability and operational stability are mission-critical.

Figure 2.

Per-phase equivalent circuit of a high-speed radial flux generator.

Table 1 delineates the principal geometric parameters of the proposed high-speed aerospace generator, which was optimised for compactness, mechanical robustness and electromagnetic performance at 10,000 RPM and 500 Hz. The design incorporates a 6-pole, 9-slot fractional-slot winding configuration, which is recognised for its minimal cogging torque and elevated winding factor, making it particularly suitable for high-speed aerospace applications. The core dimensions, including a 2 mm air gap, 120 mm stator outer diameter, 56 mm rotor outer diameter and 7 mm magnet thickness, were optimised to achieve a balance between efficiency, stability and thermal safety. A 1 mm Inconel 718 rotor band and 30 mm shaft were employed to ensure safe operation under significant centrifugal and torsional stress. The geometry adheres to the IEEE 1812-2014, IEC 60034-1 and SAE ARP4990 standards, which encompass mechanical strength, magnetic loading and thermal margins (IEC60034-1:2022 | IEC, n.d; IEEE SA - IEEE1812-2014, n.d; SAE International | Advancing mobility knowledge and solutions, n.d). The finalised model, developed using ANSYS Motor-CAD, incorporates all design dimensions to facilitate an accurate simulation of electromagnetic losses and temperature increase, thereby ensuring both manufacturability and readiness for aerospace certification.

To evaluate the behaviour of the proposed high-speed generator under full-load aerospace operating conditions, a combined electromagnetic and thermal model is developed using multiphysics simulation tools. The generator operates at 10,000 RPM and 500 Hz, which introduces challenges such as magnetic saturation, core losses and elevated temperature rise.

The electromagnetic behaviour is governed by Maxwell’s equations, particularly Faraday’s and Ampere’s Laws: (1) ∇×E=−∂B∂t \nabla \times E = - {{\partial B} \over {\partial t}} (2) ∇×H=J+∂D∂t \nabla \times H = J + {{\partial D} \over {\partial t}}

To solve these equations in finite element simulation, the magnetic vector potential method is used: (3) B=∇×A, ∇×∇μ×A=J B = \nabla \times A,\;\nabla \times \left( {{\nabla \over \mu } \times A} \right) = J where non-linear B–H curves of Finemet and SMC materials are used to account for the magnetic saturation and permeability behaviour. The rotor magnets are modelled with defined remanence and coercivity to simulate realistic flux density profiles. Simulation is performed in ANSYS Maxwell, both in 2D and 3D environments.

Copper loss in the stator winding is calculated using: (4) Pcu=I2R {P_{cu}} = {I^2}R where R includes the AC resistance due to skin and proximity effects, for which Litz wire is used to reduce eddy losses. Core losses in the stator and rotor are estimated using the generalised Steinmetz equation (Roy et al., 2017): (5) Pcore=kh f Bα+ke f2 B2 {P_{core}} = {k_h}\;f\;{B^\alpha } + {k_e}\;{f^2}\;{B^2} where f is frequency, B is peak flux density and k_h, k_e, α are material constants (specific to Finemet and SMC). Magnet losses and additional eddy current losses in rotating surfaces are also captured through time-domain loss calculation in Maxwell. The heat generated due to total losses is applied as input to a thermal model in Motor-CAD. The machine’s temperature distribution is simulated using the heat conduction equation (Kazeem Iyanda et al., 2023): (6) ρ cp∂T∂t=∇⋅k∇T+Q \rho \;{c_p}{{\partial T} \over {\partial t}} = \nabla \cdot \left( {k\nabla T} \right) + Q where ρ is the material density, c_p is the specific heat capacity, k is the thermal conductivity, T is the temperature distribution inside the corresponding material region (stator core, rotor core, magnet and windings) and Q represents the volumetric heat source generated by copper loss, core loss and magnet loss. The material-dependent thermal properties used in the simulation include Finemet ( 23 Wmk 23\;{\raise0.7ex\hbox{$W$} \!\mathord{\left/ {\vphantom {W {mk}}}\right.} \!\lower0.7ex\hbox{${mk}$}} ), SMC ( 6 Wmk 6\;{\raise0.7ex\hbox{$W$} \!\mathord{\left/ {\vphantom {W {mk}}}\right.} \!\lower0.7ex\hbox{${mk}$}} ) and copper ( 385 Wmk 385\;{\raise0.7ex\hbox{$W$} \!\mathord{\left/ {\vphantom {W {mk}}}\right.} \!\lower0.7ex\hbox{${mk}$}} ). Appropriate boundary conditions were applied to model both natural and forced convection, with the ambient temperature set to 40°C and convective heat transfer imposed on the outer casing and shaft surfaces. The thermal simulation results demonstrated that the temperatures of both the winding and the stator core remained well within their safe operational limits, reaching only 89°C in the proposed Finemet–SMC design compared to 142°C in the conventional generator. The magnetic flux-density distribution confirmed linear magnetic behaviour throughout the stator and rotor, with no signs of local saturation. Overall, the Finemet–SMC-based generator exhibited reduced thermal stress, improved cooling effectiveness and significantly enhanced stability under aerospace-rated loading conditions, strengthening its suitability for lightweight, high-speed electric propulsion applications.

Material properties are one of the key parameters in determining the feasibility and performance of electric machines for aerospace applications. The rotor core and stator core of the new generator configuration employ ultra-highly optimised soft magnetic materials. They possess high electrical resistivity with extremely thin lamination thicknesses, which significantly reduce the eddy current losses under high-frequency operation. This is especially required in UHGs where reduction of loss means system efficiency and thermal stability enhancement directly. Table 2 also show that the new materials have relatively high yield stress and low Young’s modulus but are structurally strong enough to resist excessive centrifugal forces at flight and mechanically flexible. Lower densities in the new design are directly translated to mass savings a highly desirable advantage for aerospace applications requiring higher power-to-weight ratios. Table 2 also shows that the traditional design, as high as its Young’s modulus and density, is disadvantageously lower in electrical resistivity and has more laminations, which correspond to higher iron losses. As it is currently, it is less effective and less suitable for small, lightweight and low-power applications. Among various soft magnetic materials, Hiperco 50 is chosen as the baseline reference due to its extensive use in high-speed aerospace machines. By enabling lower losses, lower weight and improved thermal management, the material configuration described here assures improved overall system performance, reliability and energy efficiency under harsh aerospace environments.

The comprehensive simulation workflow depicted in Figure 3 delineates the entire design and analysis procedure for the proposed aerospace generator. This workflow methodically integrates the electromagnetic and thermal domains through a sequential approach, commencing with the definition of the application requirements and culminating in the final performance evaluation. The process encompasses topology selection, material assignment, boundary condition configuration, finite element simulation using ANSYS Motor-CAD, thermal modelling and integration of multiphysics data. This systematic methodology ensures precise performance prediction and establishes the Finemet–SMC-based generator as a dependable and high-efficiency solution for aerospace applications.

Figure 3.

Simulation workflow for performance evaluation of proposed and conventional aerospace generators.

Figure 4 compares the terminal voltage waveforms for both generator designs in aerospace. Figure 4(a) presents the terminal voltages for the new generator design in aerospace, while Figure 4(b) presents the same in voltage amplitude consistency and waveform symmetry. In Figure 5(a), the amplitude of the peak terminal voltages between phases is approximately ±430 V (Ph1–Ph2, Ph2–Ph3 and Ph3–Ph1). Figure 4(b) illustrates that the identical voltages reach an amplitude of merely ±330 V. The higher voltage amplitude of the proposed design would denote better electromagnetic flux linkage and better optimisation of coil configuration, likely due to the Litz wire windings and magnetic circuit improvement. Furthermore, Figure 4(a) waveform is also more sinusoidal and symmetrical, indicative of a balanced phase output necessary to minimise harmonics and ensure smooth torque production in aerospace applications. Figure 4(b), by comparison, is somewhat more distorted in waveform and less symmetrical, potentially implying less efficiency and greater thermal losses at increased speeds. The improved waveform quality and higher voltage rating of the new generator demonstrate its potential for use in high-reliability, high-efficiency aerospace power systems, especially where space and weight constraints demand optimum performance from miniature electrical machines.

Figure 4.

Phase-to-phase terminal voltages: (a) proposed Finemet–SMC-based generator and (b) Hiperco 50-based conventional generator. SMC, soft magnetic composites.

Figure 5(a) presents the torque–rotational speed characteristics of the proposed Finemet–SMC-based synchronous generator. Although the rated operating speed is 10,000 rpm, the simulation is extended to higher rotational speeds to evaluate over-speed tolerance, torque stability and field-weakening behaviour – critical performance indicators for aerospace power applications. The design maintains an almost constant torque across the nominal operating region for all applied advance angles. Beyond approximately 17,000–18,000 rpm, the machine transitions into the field-weakening region, where torque gradually decreases. Notably, the torque envelope remains stable beyond 30,000 rpm, demonstrating that the Finemet–SMC combination significantly reduces iron losses and enables reliable high-speed operation with an extended constant-power range. Figure 5(b) illustrates the torque–rotational speed response of the conventional Hyperco-50-based generator. While torque remains nearly constant at low speeds, field-weakening begins earlier at ≈15,000–16,000 rpm, and torque declines sharply with increasing speed. The reduced high-speed capability is attributed to the higher hysteresis and eddy-current losses of Hyperco-50, which limit magnetic performance at elevated frequencies. Overall, the comparison confirms that the proposed Finemet–SMC generator exhibits superior high-speed characteristics, delayed field-weakening onset, enhanced torque sustainability and a significantly wider effective operating bandwidth than the conventional machine.

Figure 5.

Torque–speed characteristics of: (a) proposed Finemet–SMC-based generator and (b) Hiperco 50-based conventional generator. SMC, soft magnetic composites.

Figure 6(a) and Figure 6(b) depict that the power–speed characteristics of the proposed Finemet–SMC generator and the conventional Hiperco 50-based design, respectively, evaluated across various current advance angles. A distinct performance difference is evident between the two machines throughout the entire operating range. In Figure 6, the proposed generator demonstrates a significantly higher power extraction capability from low to high speeds. The consistently steeper power gradient is attributed to the superior electromagnetic loading facilitated by the Finemet stator and the high-resistivity SMC rotor. These materials effectively minimise leakage flux, reduce high-frequency losses and enhance the effective air-gap flux density. As a result, the power envelope remains higher for all advance angles (120°–180°), indicating strong suitability for high-speed aerospace operations. Conversely, Figure 6(b) reveals that the conventional Hiperco 50-based generator experiences a more rapid decline in power as rotational speed increases. The earlier collapse of the power envelope is primarily due to higher eddy-current loss and increased magnetic drag in the laminated steel core, which limits sustained power delivery above approximately 20,000–24,000 rpm. A direct comparison of the two envelopes shows that the proposed machine delivers 20–35% more usable power in the mid-speed range (10,000–18,000 rpm) and maintains a substantially broader and more stable high-speed operating band beyond 20,000 rpm. This advantage confirms that the Finemet–SMC configuration enhances torque production, minimises loss-induced derating and ensures stable output for high-speed aerospace propulsion systems. Although the generator is nominally rated at 10,000 rpm, the power–speed curves are extended up to 30,000 rpm to assess the overspeed capability margin required in aerospace platforms. Short-duration overspeed events can occur during rapid throttle commands, emergency power demands, or regenerative braking sequences in hybrid-electric propulsion architectures. Certification guidelines – such as IEC 60034-1 and SAE ARP4990 – recommend validating performance at 200%–300% of the nominal speed to ensure mechanical and electromagnetic robustness under such transient conditions. The proposed Finemet–SMC generator maintains stable behaviour across this extended speed range, confirming that it possesses adequate safety margins and can withstand the aerodynamic, thermal and structural stresses encountered in modern high-speed aviation environments.

Figure 6.

Power–speed envelopes of (a) the proposed Finemet–SMC-based generator and (b) the conventional Hiperco 50 generator. SMC, soft magnetic composites.

Figure 7(a) and Figure 7(b) show the magnetic flux density distribution of the new and conventional generator designs, respectively, operating under similar working conditions. In the new structure, in which Finemet is employed in the stator and SMC in the rotor, the flux density is at a maximum of approximately 1.48 T with even and symmetric flux lines along the rotor-stator interface. The colour map is utilised to indicate the flux density magnitude, with blue signifying the low flux regions (∼0 T) and red being the high flux regions approaching saturation (≥1.5 T). Most of the magnetic path in Figure 7(a) is within the green to light yellow region (∼0.7–1.4 T), indicating that the magnetic circuit is well within the linear region of the core materials. This ensures lower core losses and a higher overall efficiency.

Figure 7.

Comparison of magnetic flux density distribution: (a) proposed Finemet–SMC-based generator and (b) Hiperco 50-based conventional generator. SMC, soft magnetic composites.

On the contrary, the conventional design in Figure 7(b), with a based on high-performance cobalt–iron alloy core material (Hiperco 50), possesses a maximum flux density of approximately 1.7 T that is illustrated by orange to red regions concentrated at the stator teeth tips and yoke corners. They are prone to higher hysteresis and eddy current losses, especially at high frequency operation. Besides that, flux lines in the classical model appear denser and slightly more rounded in high-density areas, reflecting less efficient magnetic utilisation and greater local thermal stress. This comparative flux visualisation emphasises the benefit of using Finemet and SMC in the design at hand to suppress saturation effects, maintain magnetic linearity and assure high-efficiency operation under aerospace-grade conditions.

Figure 8(a) and Figure 8(b) present, respectively, the comparative Back EMF waveforms and harmonic content of the proposed Finemet–SMC generator and the conventional Hiperco 50-based generator based on Ansys Motor-CAD simulation. All three, line-to-line Back EMF waveform, harmonic amplitude spectrum and harmonic phase angles, are presented in each figure. In the specified generator of Figure 8(a), the waveform is highly sinusoidal and symmetrical with peak voltages of approximately ±600 V. There is a prominent fundamental component with small higher-order harmonics in the harmonic spectrum, which confirms reduced waveform distortion. The trend is a reflection of improved magnetic circuit and efficient winding design, which find their implementation in lower torque ripple and improved electromagnetic performance. On the contrary, Figure 8(b) shows that the conventional design generates a little smaller peak Back EMF of about ±530 V and also shows prominent 5th and other harmonic orders. These harmonics place higher electromagnetic stress and presumably higher torque ripple under working conditions. Moreover, the phase angle spread is less uniform in the conventional configuration, indicating less predictable phase behaviour under dynamic conditions. Overall, the new design embodies improved waveform quality, improved harmonic suppression and more fundamental voltage output, ideal for aerospace uses where performance, reliability and thermal efficiency are key.

Figure 8.

Back EMF and harmonic signature analysis: (a) proposed Finemet–SMC-based generator and (b) Hiperco 50-based conventional generator. SMC, soft magnetic composites.

Figure 9 illustrates the harmonic analysis of both generator designs, as evaluated using ANSYS Motor-CAD. The Finemet–SMC generator (Figure 9[a]) exhibited a smoother torque profile with a reduced fluctuation amplitude compared to the Hiperco 50 model (Figure 9[b]). The Fast Fourier Transform (FFT) results corroborate that the proposed design significantly diminishes the 6th and 9th harmonic components, thereby minimising torque ripple and electromagnetic noise. Despite both designs having similar phase distributions, the reduced harmonic magnitudes in the Finemet–SMC generator suggest a more symmetrical magnetic field and decreased vibration. Overall, the Finemet–SMC design demonstrated that superior harmonic suppression, reduced cogging torque and smoother electromagnetic torque attributed to Finemet’s high permeability and low core loss, as well as SMC’s low eddy current losses, rendering it particularly suitable for high-speed aerospace applications.

Figure 9.

Torque waveform and FFT analysis: (a) proposed Finemet–SMC-based generator and (b) Hiperco 50-based conventional generator. FFT, fast Fourier transform; SMC, soft magnetic composites.

Thermal performance comparison in Table 3 reveals the spectacular superiority of the new generator design over the conventional Hiperco 50-based solution. The new Finemet + SMC design is spectacularly lowering overall thermal losses (637.4 W to 165.4 W) through new core materials and Litz wire windings. Significantly, the stator iron loss is going down spectacularly from 226.6 W to a mere 0.6027 W, reflecting magnetic efficiency improvement.

In addition, the maximum temperatures of the curving, stator and rotor are diminished significantly in the new design, enhancing operating safety and component longevity. The margin of thermal limit is expanded from 15% to 42% and that gives a larger margin of ruggedness against thermal cycling typical of aerospace operating conditions. As may be seen from Table 3, the passive convection reference system is replaced by a forced-air technique utilising aluminium fins and ducting. The novel thermal management not only maintains system temperatures well below critical values, but it also enables long-term sustainable operation in extreme aerospace conditions.

Thermal Rise Estimation Based on Losses.

We use the basic thermal equation (Hamila et al., 2017): Δ T=Ploss×θ \Delta \;T = {P_{{\rm{loss}}}} \times \theta where Δ T is the temperature rise in °C, P_loss is the total thermal loss in W, and θ is the thermal resistance (°C/W), typically 0.3°C/W for compact aerospace machines.

The thermal performance comparison highlights a substantial enhancement in the proposed design. The conventional generator exhibited a total power loss of 737 W, resulting in a temperature increase of 221.1°C at a thermal resistance of 0.3°C/W. Conversely, the proposed design reduces the power loss to 296.4 W, thereby limiting the temperature rise to 88.9°C – a reduction of 132.2°C. This marked decrease in the operating temperature improves the thermal stability, extends the insulation life and enhances the overall system reliability, thereby affirming the superior thermal management capabilities of the proposed generator for high-power aerospace applications.

Table 4 tabulates the detailed comparative review of material content, weight distribution and cost breakdown of the traditional aerospace generator against the suggested high-speed generator design. The traditional arrangement extensively employs Hiperco 50 in the stator back as well as tooth laminations, contributing mainly to its weight and heavy manufacturing cost. In turn, the new design replaces Hiperco 50 with Finemet FT-3M, which possesses superior magnetic performance and facilitates extreme reductions in component cost and weight. The rotor core, usually composed of Hiperco 50, is replaced by Soft Magnetic Composite (SMC HB1), possessing higher thermal insulation and lower eddy current loss. The armature winding method is also improved via replacement of regular copper with Litz wire, which minimises AC losses at high frequency (500 Hz) and serves to minimise weight as well. Crucially, the total estimated cost of the new design is $2,159.61, compared to $4,601.67 for the standard setup, which represents a 53.1% decrease in capital expenditure. This cost-effectiveness, coupled with minimum architecture and material integration with high performance, emphasises commercial and technological viability of the proposed generator for use in future aero-sciences.

The thermal and electrical efficiency of any aerospace generator is directly proportional to its internal power losses. Based on a comparison between the two, the conventional aerospace generator with Hiperco 50 has a total power loss of 737 W, whereas the new high-speed generator with a Finemet stator, SMC rotor, and Litz wire windings has a significantly lower total loss of 296.4 W.

The difference in loss, ΔPloss=737−296.4=440.6 W \matrix{ {\Delta {{\rm{P}}_{{\rm{loss}}}} = 737 - 296.4} \cr { = 440.6\;{\rm{W}}} \cr } constitutes a 59.8% reduction in overall losses.

Taking a normal operation time of 4,000 h/year, which is common for aerospace auxiliary or continuous electrical systems, the equivalent yearly energy savings can be determined as: Esaved per year=440.6 W×4000 h=1762.4KWh/year \matrix{ {{E_{{\rm{saved}}\;{\rm{per}}\;{\rm{year}}}}} \hfill & { = 440.6\;{\rm{W}} \times 4000\;{\rm{h}}} \hfill \cr {} \hfill & { = 1762.4{\rm{KWh}}/{\rm{year}}} \hfill \cr }

Such energy saving reduces the load on the onboard power system, along with reduced cooling requirements, longer component life and system reliability overall. From an ecological perspective as well as a monetary one, such a saving can translate to less fuel consumption on generator-powered aircraft, making operations environmentally friendly as well as cost-effective in the long term.

Due to the severe operating conditions in aerospace environments, such as frequent thermal cycling, mechanical vibrations and regular maintenance, an optimal service life of 15 years is generally assumed for onboard electrical generators. Based on the calculated annual energy savings of 1,762.4 kWh/year and a conservative global average electricity price of $0.12 per kWh, annual operating cost savings can be estimated as: Esaved per year×0.12=$211.488/year {E_{{\rm{saved}}\;{\rm{per}}\;{\rm{year}}}} \times 0.12 = \$ 211.488/{\rm{year}}

Total Lifetime Energy Cost Savings: Annual Energy Cost Savings×15=$3,172.32 {\rm{Annual\; Energy\; Cost\; Savings}} \times 15 = \$ 3,172.32

These significant lifecycle cost savings of over $3100 highlights the economic benefits of employing the planned generator architecture. For defence and commercial aviation applications, where operating efficiency and cost of ownership are both major considerations, such long-term energy savings can play a large part in reduced fuel consumption, lower emissions and enhanced return on investment. Additionally, the reduced thermal and electrical stress on system elements lowers maintenance schedules and enhances reliability, substantiating its value in mission-critical aerospace operations.