France



I. INTRODUCTION&lt;br/&gt;Non-permanent magnet machines are having growing interest recently due to the volatility of permanent magnet (PM) prices. Variable flux reluctance machine (VFRM) is a non-permanent magnet machine and it is a candidate for high-performance applications such as electric vehicles [1,2]. There are several slot/pole combinations of VFRM, such as 6/4, 6/5, 6/7, and 12/10 [3,4]. The 6/4 topology has a large torque ripple [5]. The 6/5 and 6/7 topologies have low torque ripple, but they have an unbalanced magnetic force (UMF) issue which causes mechanical stress and acoustic noise due to the odd rotor poles [6]. The 12/10 topology has a low torque ripple and balanced UMF.&lt;br/&gt;Although the structure of the VFRM is similar to the switched reluctance machine (SRM), VFRM has DC winding in the stator as well as three-phase AC windings. Therefore, the back-EMF waveform of the armature winding is sinusoidal. The control of the VFRM is easier than SRM due to the sinusoidal back-EMF.&lt;br/&gt;There is a magnetic coupling between armature and field winding because they have a common flux path as seen in Fig. 1. However, the field winding is fed by a DC current while armature windings have an AC current. In this study, the magnetic coupling between armature and field winding is analyzed and how VFRM can operate having DC and AC windings is explained.&lt;br/&gt;&lt;br/&gt;II. FLUX LINKAGE ANALYSIS OF FIELD WINDING&lt;br/&gt;&lt;br/&gt;Every tooth of VFRM has a field winding as shown in Fig. 1. Blue ones are field windings, red, green and yellow ones are armature windings. Firstly, the field winding is separated and called DC1, D2, D3, DC4, DC5, and DC6. Field windings are excited by DC current while armature windings are not excited. The flux of a single excited magnetic circuit is determined by Eq. (1).&lt;br/&gt;&lt;br/&gt;θ=NI/R(θ&lt;sub&gt;r&lt;/sub&gt;) (1)&lt;br/&gt;&lt;br/&gt;Where N is turns number, I is current, R is magnetic reluctance and θ&lt;sub&gt;r&lt;/sub&gt; is rotor position. The magnetic reluctance of the motor depends on rotor position because effective airgap changes with rotor position. Flux linkages of DC field windings and resultant flux linkage of field winding are obtained using the FEA, as shown in Fig. 2. Reluctance function of DC1 winding is obtained by FFT analysis of flux linkage, turns number and DC current. R&lt;sub&gt;0&lt;/sub&gt; is the DC component of reluctance, n is harmonic order, R&lt;sub&gt;n&lt;/sub&gt; is the magnitude of nth harmonic, and α is the angle of harmonic with respect to rotor position.&lt;br/&gt;&lt;br/&gt;R&lt;sub&gt;DC1&lt;/sub&gt;=R&lt;sub&gt;0&lt;/sub&gt;+∑R&lt;sub&gt;n&lt;/sub&gt;sin(nθ&lt;sub&gt;r&lt;/sub&gt;+α) (2)&lt;br/&gt;&lt;br/&gt;Flux linkage of DC1 winding is obtained using Eq. 1 and Eq. 2. Each of stator and rotor poles are not aligned at the same time, so there is a phase difference between DC windings. Therefore, the flux linkage of the DC2 winding is given in Eq. 4.&lt;br/&gt;&lt;br/&gt;λ&lt;sub&gt;DC1&lt;/sub&gt;(θ&lt;sub&gt;r&lt;/sub&gt;)=λ&lt;sub&gt;0&lt;/sub&gt;+∑λ&lt;sub&gt;n&lt;/sub&gt;sin(nθ&lt;sub&gt;r&lt;/sub&gt;+α) (3)&lt;br/&gt;λ&lt;sub&gt;DC2&lt;/sub&gt;(θ&lt;sub&gt;r&lt;/sub&gt;)=λ&lt;sub&gt;0&lt;/sub&gt;+∑λ&lt;sub&gt;n&lt;/sub&gt;sin(nθ&lt;sub&gt;r&lt;/sub&gt;+α+π/3) (4)&lt;br/&gt;&lt;br/&gt;Eventually, resultant flux linkage of when all DC windings are connected series is obtained by Eq. 5.&lt;br/&gt;&lt;br/&gt;λ&lt;sub&gt;T&lt;/sub&gt;(θ&lt;sub&gt;r&lt;/sub&gt;)=6λ&lt;sub&gt;0&lt;/sub&gt;+∑λ&lt;sub&gt;n&lt;/sub&gt;sin(nθ&lt;sub&gt;r&lt;/sub&gt;+α)+∑λ&lt;sub&gt;n&lt;/sub&gt;sin(nθ&lt;sub&gt;r&lt;/sub&gt;+α+π/3) +∑λ&lt;sub&gt;n&lt;/sub&gt;sin(nθ&lt;sub&gt;r&lt;/sub&gt;+α+2π/3)+∑λ&lt;sub&gt;n&lt;/sub&gt;sin(nθ&lt;sub&gt;r&lt;/sub&gt;+α+3π/3)+∑λ&lt;sub&gt;n&lt;/sub&gt;sin(nθ&lt;sub&gt;r&lt;/sub&gt;+α+4π/3)+∑λ&lt;sub&gt;n&lt;/sub&gt;sin(nθ&lt;sub&gt;r&lt;/sub&gt;+α+5π/3) (5)&lt;br/&gt;λ&lt;sub&gt;T&lt;/sub&gt;(θ&lt;sub&gt;r&lt;/sub&gt;)=6λ&lt;sub&gt;0 &lt;/sub&gt;(6)&lt;br/&gt;&lt;br/&gt;As shown in Fig. 2, all of the AC components eliminate each other and the resultant flux linkage has only a DC component. In this analysis, armature windings are not excited. Also, both field and armature windings are excited and flux linkage of the field winding is obtained by FEA, core material is assumed it has infinite magnetic permeance. Although magnitudes of the harmonic orders are changed, all AC components are eliminated and resultant flux linkages consist of DC components. Flux linkage of field winding has little harmonic if core material is steel, due to the magnetic saturation and leakage flux. These results could not be presented in this digest due to the figure limits. However, all of the results will be given in the final paper.&lt;br/&gt;&lt;br/&gt;III. CONCLUSION&lt;br/&gt;VFRM has DC field winding and armature windings in the stator. Although armature and field windings have a common flux path and magnetic coupling, field winding can be excited by DC current. Thus, the flux linkage of the DC field winding is analyzed while armature windings are excited by AC current. The field winding is separated, and flux linkages of every winding are obtained by FEA. Every winding has a DC biased AC flux linkage. However, all DC windings are connected series, AC components of resultant flux are eliminated. Eventually, field winding has only a DC component even if there is a magnetic coupling between armature and field windings. The magnetic circuit of analysis and results at full load will be given in the final paper due to the figure limits of the digest.

**References:**

1. T. Raminosa, D. A. Torrey and A. M. El-Refaie, IEEE Trans. on Industry Applications, Vol. 52, p. 2129, (2016)&lt;br/&gt;2. T. Raminosa, A. M. El-Refaie and D. A. Torrey, IEEE Trans. on Industry Applications, Vol. 53, p. 3496, (2017)&lt;br/&gt;3. X. Liu and Z. Q. Zhu, IEEE Transactions on Magnetics&lt;i&gt;, &lt;/i&gt;vol. 49, p. 3020 (2013)&lt;br/&gt;4. . J. T. Shi, X. Liu and Z. Q. Zhu, IEEE Transactions on Magnetics,&lt;i&gt; &lt;/i&gt;Vol. 50, (2014).&lt;br/&gt;5. J. Feng, L. Huang and Z. Q. Zhu, IEEE Transactions on Energy Conversion, Vol. 34. p.1135 (2019)&lt;br/&gt;6. X. Liu and Z. Q. Zhu, IEEE Transactions on Industry Applications, Vol. 50. p. 3675 (2014)

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/746d6e18f4498238f67fb44609dbefd9.jpg)
Fig. 1 Structure and windings of 12/10 VFRM

![](https://assets.underline.io/uploads/markdown_image/1/image/1be7ef569c9266b81d7244046b273b53.jpg)
Fig. 2. Flux linkage variation of DC windings

INTERMAG 2021

 Analysis of Magnetic Coupling Between Armature and Field Windings of Variable Flux Reluctance Machine

INTERMAG 2021 Welcome Video



	
<figure class="video_container">
 <iframe src="https://screencast-o-matic.com/player/crflqFVnUpZ?v=6&ff=1&title=0&controls=1" width=640 height=360 frameborder="0" allowfullscreen="true"> </iframe>
</figure>

Want to buy a ticket? 

Register for INTERMAG 2021 here

[![](https://assets.underline.io/uploads/markdown_image/1/image/17ce76e7960db633258024eef8d6da7f.png)](https://intermag.org/register)

Got your ticket, but still can’t access the event?

Contact us and our team will look into it.

I. INTRODUCTION Non-permanent magnet machines are having growing interest recently due to the volatility of permanent magnet (PM) prices. Variable flux reluctance machine (VFRM) is a non-permanent magnet machine and it is a candidate for high-performance applications such as electric vehicles [1,2]. There are several slot/pole combinations of VFRM, such as 6/4, 6/5, 6/7, and 12/10 [3,4]. The 6/4 topology has a large torque ripple [5]. The 6/5 and 6/7 topologies have low torque ripple, but they have an unbalanced magnetic force (UMF) issue which causes mechanical stress and acoustic noise due to the odd rotor poles [6]. The 12/10 topology has a low torque ripple and balanced UMF. Although the structure of the VFRM is similar to the switched reluctance machine (SRM), VFRM has DC winding in the stator as well as three-phase AC windings. Therefore, the back-EMF waveform of the armature winding is sinusoidal. The control of the VFRM is easier than SRM due to the sinusoidal back-EMF. There is a magnetic coupling between armature and field winding because they have a common flux path as seen in Fig. 1. However, the field winding is fed by a DC current while armature windings have an AC current. In this study, the magnetic coupling between armature and field winding is analyzed and how VFRM can operate having DC and AC windings is explained. II. FLUX LINKAGE ANALYSIS OF FIELD WINDING Every tooth of VFRM has a field winding as shown in Fig. 1. Blue ones are field windings, red, green and yellow ones are armature windings. Firstly, the field winding is separated and called DC1, D2, D3, DC4, DC5, and DC6. Field windings are excited by DC current while armature windings are not excited. The flux of a single excited magnetic circuit is determined by Eq. (1). θ=NI/R(θr) (1) Where N is turns number, I is current, R is magnetic reluctance and θr is rotor position. The magnetic reluctance of the motor depends on rotor position because effective airgap changes with rotor position. Flux linkages of DC field windings and resultant flux linkage of field winding are obtained using the FEA, as shown in Fig. 2. Reluctance function of DC1 winding is obtained by FFT analysis of flux linkage, turns number and DC current. R0 is the DC component of reluctance, n is harmonic order, Rn is the magnitude of nth harmonic, and α is the angle of harmonic with respect to rotor position. RDC1=R0+∑Rnsin(nθr+α) (2) Flux linkage of DC1 winding is obtained using Eq. 1 and Eq. 2. Each of stator and rotor poles are not aligned at the same time, so there is a phase difference between DC windings. Therefore, the flux linkage of the DC2 winding is given in Eq. 4. λDC1(θr)=λ0+∑λnsin(nθr+α) (3) λDC2(θr)=λ0+∑λnsin(nθr+α+π/3) (4) Eventually, resultant flux linkage of when all DC windings are connected series is obtained by Eq. 5. λT(θr)=6λ0+∑λnsin(nθr+α)+∑λnsin(nθr+α+π/3) +∑λnsin(nθr+α+2π/3)+∑λnsin(nθr+α+3π/3)+∑λnsin(nθr+α+4π/3)+∑λnsin(nθr+α+5π/3) (5) λT(θr)=6λ0 (6) As shown in Fig. 2, all of the AC components eliminate each other and the resultant flux linkage has only a DC component. In this analysis, armature windings are not excited. Also, both field and armature windings are excited and flux linkage of the field winding is obtained by FEA, core material is assumed it has infinite magnetic permeance. Although magnitudes of the harmonic orders are changed, all AC components are eliminated and resultant flux linkages consist of DC components. Flux linkage of field winding has little harmonic if core material is steel, due to the magnetic saturation and leakage flux. These results could not be presented in this digest due to the figure limits. However, all of the results will be given in the final paper. III. CONCLUSION VFRM has DC field winding and armature windings in the stator. Although armature and field windings have a common flux path and magnetic coupling, field winding can be excited by DC current. Thus, the flux linkage of the DC field winding is analyzed while armature windings are excited by AC current. The field winding is separated, and flux linkages of every winding are obtained by FEA. Every winding has a DC biased AC flux linkage. However, all DC windings are connected series, AC components of resultant flux are eliminated. Eventually, field winding has only a DC component even if there is a magnetic coupling between armature and field windings. The magnetic circuit of analysis and results at full load will be given in the final paper due to the figure limits of the digest.

**References:**

1. T. Raminosa, D. A. Torrey and A. M. El-Refaie, IEEE Trans. on Industry Applications, Vol. 52, p. 2129, (2016) 2. T. Raminosa, A. M. El-Refaie and D. A. Torrey, IEEE Trans. on Industry Applications, Vol. 53, p. 3496, (2017) 3. X. Liu and Z. Q. Zhu, IEEE Transactions on Magnetics, vol. 49, p. 3020 (2013) 4. . J. T. Shi, X. Liu and Z. Q. Zhu, IEEE Transactions on Magnetics, Vol. 50, (2014). 5. J. Feng, L. Huang and Z. Q. Zhu, IEEE Transactions on Energy Conversion, Vol. 34. p.1135 (2019) 6. X. Liu and Z. Q. Zhu, IEEE Transactions on Industry Applications, Vol. 50. p. 3675 (2014)

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/746d6e18f4498238f67fb44609dbefd9.jpg)
Fig. 1 Structure and windings of 12/10 VFRM

![](https://assets.underline.io/uploads/markdown_image/1/image/1be7ef569c9266b81d7244046b273b53.jpg)
Fig. 2. Flux linkage variation of DC windings

technical paper

To achieve the force/torque control feature for robotic applications, some robotic devices such as robotic grippers or robotic arms use servo motors’ current or the structures’ mechanical strain to measure torques and forces. Additional commercial torque sensors are also commonly used to detect the output forces. Although these sensors can detect motor torques or contact forces, they are too stiff to render a soft stiffness [1]. If the load environment is stiff and the actuator is also stiff, the force controller will lead to low accuracy, as any tiny movements will cause a large force variety and components are rigid. To make robotic actuators perform more compliant with torque control and stiffness control, a series elastic actuation method was proposed [2]. The series elastic actuator (SEA) is constructed with an actuator and an elastic spring which can be added between the motor and the load to provide compliant features [3]. With displacement sensors attached at two sides of the spring, the output torque can be computed [4] by measuring the relative movement. With the SEA, a force control can then be transferred into a position control whatever the load environment is stiff or not. Magnetic coupling is a component that transmits torque from the motor to load without any direct mechanical contact, as attractive forces between N and S poles are used to transmit the force. When there is a relative angle difference between N and S poles, the attractive forces, which can be separated into axial forces and shear forces, generate a relative force that can be utilized as a virtual spring in the SEA system. The torsional stiffness of magnetic coupling was reported in [5] to be sensitive to strength of magnetic forces and air gap, axial displacement, and angular misalignment. When the load excesses maximum torque that can be transmitted by the magnetic coupling, it then switches to pole-slipping mode to avoid possible permanent deformation in structure [6]. To form a magnetic SEA (mSEA) system, an elastic magnetic coupling must be placed within the actuator and the load. This configuration of SEA connects the elastic component to the load directly, which apparently is sensitive to changes of the external load. By using the magnetic coupling, the attraction forces between two magnets can be transferred to the rotary torque when angular rotation occurs. With the angle solver attached to the load, the angular displacement of mSEA can be measured by the angle difference between the load’s angle solver and the motor’s encoder. The mathematic relation between output torque and rotating angle is the stiffness of magnetic coupling, which is an important variable for a mSEA controlling system. In this study, a 4-pole axially magnetized NdFeB ring magnet (Magtech Corp., TW) was chosen. The magnet’s outer and inner diameters are 24 and 13mm, respectively. The magnet’s thickness 3mm with flux density of 3530G. Because of its flat ring structure, the mSEA is more compact in size than other types of magnetic coupling. Beyond the size and material of the magnet, some other parameters affecting characteristics of the mSEA, like airgap length, pole number, and angular displacement were studied and analyzed in this paper toward understanding the torsional stiffness and attractive force for the axial magnetic coupling. The relation between airgap length and angular displacement is shown in Fig. 1(a), from which one can observes that the torque is a parabola curve with a saturation region between 30 o and 60o, with decreasing of torque when angular displacement is over 60 o. The maximum torque is observed to decrease significantly when the airgap, indicated as “e” in the figure, becomes larger. On the other hand, the axial force renders a linear relationship with the angular displacement as shown in Fig. 1(b). When the angle between the two magnets goes larger, the axial force decreases linearly, leading to zero value where the two magnets are separated. Along with these results, the airgap is found to be provide significant influence in axial attractive force and pull-out torque. With higher airgap, attractive force and pull-out torque both go lower. From this study, the torque for the magnetic couple is also found to be affected by the angular displacement of two magnets and the number of pole pairs. When the angular displacement goes higher, the pull-out torque follows a parabola curve, reaching to the maximum torque value 150 mN-m around 15 o with the airgap is 1mm as showed in Fig. 1. The range of angular displacement beyond the saturation region is the magnet coupler’s effective region. By using the model 4-pole axially magnetized ring magnet, the transmitting torque for the coupler can be then controlled, as the torsional stiffness can be rendered by the model of magnetic coupling. The proposed mSEA can render various magnitudes of torsional stiffness and the rendered results are showed in Fig. 2. The mSEA system can be controlled to perform desired motion when a virtual spring stiffness is set between 50 to 300mN-m/rad with ~5% stiffness rendering error compared to ideal stiffness. The maximum output torque of the mSEA is 130mN-m with force resolution of 0.1mN and control error less than 1mN. With the proposed mSEA put between actuator and load, the output torque and stiffness can certainly be controlled for robotic applications with fragile objects or human.

**References:**

[1] Y. Yu and C. Lan, "Design of a Miniature Series Elastic Actuator for Bilateral Teleoperations Requiring Accurate Torque Sensing and Control," IEEE Robotics and Automation Letters, vol. 4, no. 2, pp. 500-507 (2019) [2] J. Pratt, B. Krupp, and C. Morse, "Series elastic actuators for high fidelity force control," Industrial Robot: An International Journal, vol. 29, pp. 234-241, 06/01 (2002) [3] C. Lee, S. Kwak, J. Kwak, and S. Oh, "Generalization of Series Elastic Actuator Configurations and Dynamic Behavior Comparison," Actuators, vol. 6, p. 26, 08/22 (2017) [4] S. Oh and K. Kong, "High-precision robust force control of a series elastic actuator," IEEE/ASME Transactions on Mechatronics, vol. 22, no. 1, pp. 71-80 (2016) [5] T. Lubin, S. Mezani, and A. Rezzoug, "Simple Analytical Expressions for the Force and Torque of Axial Magnetic Couplings," IEEE Transactions on Energy Conversion, vol. 27, no. 2, pp. 536-546 (2012) [6] R. Montague, C. Bingham, and K. Atallah, "Servo Control of Magnetic Gears," IEEE/ASME Transactions on Mechatronics, vol. 17, no. 2, pp. 269-278 (2012)

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/e184980e50eb25832996607b5fbc0282.jpg)
Fig. 1 (a) torque and (b) axial force caused by angular displacement between the magnetic rings.

![](https://assets.underline.io/uploads/markdown_image/1/image/073c42b1fea45f07e5b3b6e362b3b838.jpg)
Fig. 2 Measured and predicted torque for the model mSEA.

 Design of a Stiffness Control Actuator Utilizing Magneto-Elastic Actuation

With the increasing focus on energy efficiency for the green transition, there is a desire to identify and reduce all losses possible. For rotary machines losses occur as friction at the bearings, and therefore frictionless magnetic bearings can be an important upgrade to high-speed machinery. However, as proven by Samuel Earnshaw in 1842 magnetic bearings are inherently unstable. Typically, magnetic bearings are stable in the lift (often vertical) direction. This results in instability in the radial direction and therefore some kind of stabilizing force is required in this direction [1]. For a magnetic bearing, the radial stiffness determines the active force needed to keep the bearing stable. Thus there is a compromise in designing a magnetic bearing which can both lift as high a load as possible and at the same time have a low and constant radial instability [2-5]. Here we describe a new magnetic bearing aiming to fulfill these requirements [6]. The bearing consists of two nested conical axi-symmetric permanent magnet rings with quadratic cross-sections, as shown in Fig. 1. This configuration is a superposition between a radially magnetized bearing and an axially magnetized bearing. The rotor and stator are centered on the same axis, and when load is applied on the rotor bearing this will move along the z-axis. The stator has a magnetization direction that is along the inwards normal to its surface, while the rotor has a magnetization direction outwards normal to its surface. Thus, there is repulsion between the rings. The idea behind this design is that the magnetic forces will both have a component providing a lifting force along the z-axis, but also a component along the radial direction, which might allow the stiffness along this direction to be controlled. Such a design of tilted conical rings of permanents magnets can easily be made of cubes of permanent magnets placed in a ring. We consider a design where there is a minimum distance between rotor and stator of 0.2 mm and where the conical rings that make up both the rotor and the stator have quadratic cross-section side lengths of 3 mm. The radius of the center of the tilted ring in the stator is fixed at 24 mm, while it is varied for the rotor. We consider different tilt angles of the stator/rotor rings in the bearing and we also consider ring geometries with twice the cross-section and different magnetization directions. The “single” bearing consists of two tilted concentric rings with opposite magnetization as described above, the “double” bearing has two magnets in each tilted ring, and finally the “alternating” bearing has the magnetization of one part of the ring reversed in both the stator and rotor. All configurations are illustrated in Fig. 1. We assume that the permanent magnets in the bearing have a remanence of 1.4 T and a relative permeability of 1. We determine the force and stiffness of the bearing using a magnetostatic finite element (FEM) axi-symmetric model in Comsol Multiphysics. As the simulations are 2D axi-symmetric a very high mesh resolution can be used, eliminating the known convergence issues with bearing models in 3D [7]. We have conducted a variation study to determine the optimal geometry of the bearing as function of the tilt angle and the bearing type. Tilt angles from 0 to 90 degrees in steps of 15 degrees were modelled. Combined with the three bearing types studied, this gives a total of 21 bearings modelled. For each of these we use an adaptive routine to determine the optimal rotor radius to a precision of 0.1 mm. We show that the optimal value of the rotor radius is almost identical for the different bearing geometries considered, with decreasing rotor radius for increasing tilt angle. Following this we determine which bearing would be easiest to control, i.e. has a constant stiffness value, but which also provides a high lifting force. We investigate this by calculating the relative standard deviation for both the axial force and stiffness. This is shown in Fig. 2 for all studied bearings which are stable in the axial direction. Choosing the best bearing depends on the desired characteristics of the bearing. If one desires a bearing that is easy to control, i.e. has a mostly constant stiffness, for a 3 mm axial displacement the double geometry with a tilt angle between 60-75 degrees is clearly the best. When a smaller axial displacement of 1 mm is considered, all bearings with a tilt angle of 60 degrees display almost constant stiffness. This is a remarkable property, as such a bearing is very easy to control with an active bearing. In this bearing the exact load on the rotor is not of importance, as long as it is between the minimum and maximum lift, and at the same time the stiffness is constant regardless of the load. We thus show that by varying the tilt angle and the bearing geometry it is possible to tailor a bearing to the desired characteristics and achieve an easily controllable bearing with a high lifting force.

**References:**

[1] J. P. Yonnet, IEEE Transactions on Magnetics, Vol. 17, p.1169-1173 (1981) [2] J. G. Detoni, Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science, Vol. 228, p.1829-1844 (2014) [3] J. Sun, D. Chen and Y. Ren. , IEEE Transactions on Instrumentation and Measurement, Vol. 62, p.2960-2965 (2013) [4] L. Zhang, H. Wu, P. Li, Y. Hu and C. Song, Mathematical Problems in Engineering, Vol. 2019, p.4265698 (2019) [5] K. K. Nielsen, C. R. H. Bahl, N. A. Dagnaes, I. F. Santos and R. Bjørk, IEEE Transactions on Magnetics, 10.1109/TMAG.2020.3042957 (2020) [6] R. Bjørk and C. R. H. Bahl, “A conical passive magnetic bearing with constant stiffness”, submitted for publication (2021) [7] N. A. Dagnæs-Hansen and I. F. Santos., Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science, Vol. 233, p.5280-5293 (2019)

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/da90a1119a35e4bf7055dcd0b30f5edb.jpg)
Fig.1: The bearing geometry considered. The stator, shown in green, and the rotor, shown in grey, each consists of two tilted rings of permanent magnets with a quadratic cross-section. Their direction of magnetization is also indicated. On the right is shown a cross-section of the bearing, with the three types of magnet configurations considered. The bearing illustrated here has a tilt angle of 30 degrees.

![](https://assets.underline.io/uploads/markdown_image/1/image/015e67cee587135f544e04ac077b1ccd.jpg)
Fig.2: The relative standard deviation of the stiffness in the z-direction in percent as function of the relative standard deviation of the force in the z-direction in percent.

 A constant stiffness magnetic bearing

I. Introduction The conventional modulating ring in a coaxial magnetic gear is composed of separate pieces of laminated steel known as flux modulators. The assembly of the separate modulators is not easy because the laminations of each modulator must be firmly held together by using a clamping rod. A modulating ring with bridges that integrate the steel pieces into a whole laminated ring can greatly save the assembly efforts [1]. From the electromagnetic perspective, thin bridges are undoubtedly preferred as thick bridges compromise the modulating effect and hence downgrade the pull-out torque [2], [3]. However, thin bridges challenge the mechanical design of the modulating ring considering the stress due to the magnetic pull. Fernando and Saha investigated the effect of different numbers of flux modulators, thicknesses and positions of bridge on the shear stress in the modulating ring, and ended with a double-bridge design [3]. Due to the effectively thick bridges, this design has a large loss of torque capacity though the double-bridge structure is very robust mechanically. This digests proposes a design procedure of bridged modulators that reduces the mechanical stress at the cost of torque capacity as small as possible. II. Magnetic Gear With Bridged Modulators Fig. 1 shows the prototype magnetic gear with bridged modulators. This magnetic gear was originally presented with separate modulators, and more information about its construction can be found in [4]. For the study in this paper, the separate modulators are replaced with a modulating ring integrated by the bridges near the high-speed rotor (HSR), as shown Fig. 1. Such a position of bridges was proved to have larger torque capacity than any other positions [2], [3]. The modulating ring laminations are kept together by using the end plates and the clamping rods that are inserted into three of the modulators. III. Design of Modulating Ring Considering Stress A. With Different Thicknesses of Bridge Two modulating ring designs with different thicknesses of bridge and equal ratio of pole arc to pole pitch (δ) (designs 1 and 2) are presented in Table I. The time-varying torques and stress distributions in these magnetic rings are calculated by three-dimensional (3-D) time-stepping finite element method (T-S FEM). The maximum mean torques extracted from the torque curves with different initial relative positions of the rotors, known as pull-out torques, are 15.70 Nm and 16.24 Nm at the outer rotors of designs 1 and 2, respectively. Table I shows the maximums of the fluctuating stress under approximately equal load condition. The stress of the thinner bridge is apparently higher than that of the thicker bridge. The maximum stress occurs at the junction of the bridge and the modulator that encloses a clamping rod. It should be noted that the modulating ring bears wildly fluctuating stress which is more prone to cause failure than a static stress of equivalent magnitude in a long-term operation, due to the metallic fatigue. Therefore, a sufficient margin of safety for yield must be reserved. B. With Different Ratios of Pole Arc to Pole Pitch Previous research indicates that the pull-out torque does not deviate significantly from the maximum for a certain large variation of δ in the vicinity of δm corresponding to the maximum pull-out torque [2]. This feature suggests that an appropriate value of δ can reduce the stress without downgrading the torque capacity. For the bridge thickness of 0.8 mm, the calculated pull-out torques with the δ values of 0.33, 0.375, 0.41, 0.45, 0.5 and 0.58 are 16.00 Nm, 16.33 Nm, 16.44 Nm, 16.53 Nm, 16.17 Nm and 15.11 Nm, respectively. When δ equals 0.33 to 0.5, the torque capacity is not apparently different from the maximum. Table I shows the maximum stress in the modulating ring with that range of δ, at approximately equal load. In comparison to the δm design, the design with δ of 0.33 reduces the maximum stress by 24.1% at the cost of only 3.2% lower torque capacity. C. With Different Fillet Radii at Junctions Filleting the sharp or right-angle corner is a commonly used mechanical design approach to reducing the stress concentration in a junction. Designs 1 to 6 all have right-angle junctions. Table I shows the maximum stresses with different δ and fillet radii (designs 7 to 15) at approximately equal load. In those cases, the increase of fillet radius relieves the stress only when δ equals 0.5. Filleting the junction may not relieve the stress in the modulating ring because the distribution of the magnetic load force changes together with the shape of bridge, unlike a mechanical load force. IV. Conclusion This digest proposes a design procedure for reduced mechanical stress in the bridged modulating ring of coaxial magnetic gear, at minimum cost of torque capacity. Step 1 of the design procedure is to place the bridges near the side of high-speed rotor and minimize the thickness of bridge for maximum torque capacity and acceptable stress. Step 2 is to vary the pole-arc-to-pole-pitch ratio of the modulating ring for reduced stress at acceptable cost of torque capacity. Step 3 is to fillet the junctions for the potential further reduction of stress. These steps could be repeated until the desired torque capacity and stress are achieved. Using more clamping rods could be another approach to reducing the stress. The full paper will investigate the effects of adding clamping rods on the stress and the eddy current loss in the modulating ring.

**References:**

[1] D. Z. Abdelhamid and A. M. Knight, IEEE Trans. Magn., vol. 53, no. 11, Art. no. 8112804 (2017). [2] Y. Zhan, et al, IEEE Trans. Magn., vol. 55, no. 6, Art. no. 8103904 (2019). [3] N. Fernando and S. Saha, IEEE Trans. Energy Convers., vol. 33, no. 1, pp. 49–58 (2018). [4] Y. Ge, Q. Xin, and C. Nie, Journal of Mech. Trans., vol. 36, no. 4, pp. 5–8, 13 (2012).

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/3d7ea0b425a5955b905b54bff4286211.jpg)
Fig. 1. Magnetic gear

![](https://assets.underline.io/uploads/markdown_image/1/image/8178c855ff5aa363505d14637e4d1e46.jpg)
Table I Load torques and stresses

 Design of Bridged Flux Modulators in Coaxial Magnetic Gear Considering Mechanical Stress

The electrification of vehicles and ships has been required to reduce the carbon dioxide emissions on global environment. In many cases, a general drive system of electric vehicles and ships consists of an electric motor and a mechanical gearbox to obtain desired speed and torque in terms of reducing size and weight. However, the mechanical gearbox has inherent problems caused by mechanical contacts, e.g., acoustic noise, wear and tear, vibration, and periodical maintenance. To solve the above problems, a magnetic-geared motor (MGM) has attracted attention [1]. The MGM consists of a permanent magnet (PM) motor and a flux-modulated type magnetic gear (MG), and they can transmit the torque without any contacts by an electromagnetic force. In addition, the MGM has the structure that a PM motor and an MG are combined magnetically and mechanically as one electric machine so that it can offer the higher torque density than the conventional drive system. Hence, the MGM is expected to be put into practical use. In previous studies, Refs. [2] and [3] have reported about a magnetic interaction of the MGM caused by magnetically coupling between a PM motor part and an MG part. These studies have discussed about the relationship between the magnetic interaction and the output torque of MGM. The magnetic interaction may influence only the output torque, but also the efficiency and the power factor. However, the influence of the magnetic interaction on the efficiency and the power factor have not been clarified. This paper investigates that the influence of the magnetic interaction on the efficiency and the power factor of the MGM by using a three-dimensional finite element method (3D-FEM). Furthermore, in order to verify the validity, the experimental results are compared to the calculated ones. Fig.1 shows the specifications of the MGM used for examinations. It consists of a PM motor part and an MG part, which share one high-speed rotor (HSR). The PM motor part consists of an armature and the shared HSR, while the MG part consists of the shared HSR, a magnetic modulator called pole-pieces (PP), and a low-speed-rotor (LSR). The numbers of pole-pairs of the shared HSR and the LSR are 4 and 23, respectively. Hence, the gear ratio is 5.75 which is given by the pole-pair ratio of the shared HSR and the LSR. First of all, the phase of the armature voltage is calculated by 3D-FEM in various load angles, namely, in various load torques, to clarify the influence of the magnetic interaction on the armature voltage. In 3D-FEM, the rotational speed and the current phase are kept constant. Fig. 2(a) indicates the calculated waveforms of the armature voltage at various load angles of 0 (no-load), 30, 45, and 90 degrees. In addition, an input current is also indicated in the figure as a reference of the phase. It is understood that the phase of the armature voltage is gradually shifted by increasing the load angles, i.e. load torque. The reason why the phase of the armature voltage is shifted is discussed based on the rotational coordinate system synchronized with HSR. Fig. 2 (b) shows the vector diagram of MGM based on the rotational coordinate system of the HSR when a load angle is δ and a current phase is β. In the figure, the no-load induce voltage is ea, the armature voltage is Va, the armature current is ia, the phase difference between Va and ia is Φ, d- and q-axis inductance are Ld and Lq, the magnetic flux of HSR magnet is ψh, the modulated magnetic flux of LSR magnet is ψml due to the magnetic interaction, respectively. The sum of four magnetic fluxes, which are Ldid, Lqiq, ψh, and ψml, is represented as the combined magnetic flux ψo. The phase of ψml changes with the load angle δ, thereby the phase of ψo changes too. Hence, the phase of Va shifts due to the magnetic interaction when the load angle changes, because the phase difference between Va and ψo is always 90 degrees. Based on the above discussions, it is considered that power factor also changes due to the magnetic interaction. Fig. 2 (c) indicates the calculated results of the displacement power factor of MGM in various load torques and current phases. The figure reveals that the displacement power factor changes with the load torques. Moreover, it is understood that the maximum point of the displacement power factor varies in each load torque. Therefore, it is concluded that if the current phase is advanced in accordance with increasing the load torque, the MGM can be a high-power-factor drive. Fig. 2 (d) indicates the measured results of the displacement power factor of the prototype MGM. The figure reveals that the measured displacement power factor changes with the load torque in the same way as the calculated ones shown in Fig. 2(c). Hence, the validity of the calculated results is proved.

**References:**

[1] K. T. Chau, Z. Dong, J. Z. Jiang, L. Chunhua, and Z. Yuejin, IEEE Trans. Magn., Vol. 43, pp. 2504-2506 (2007) [2] H. Suzuki, K. Hirata, N. Niguchi, A. Kohara and K. Takahara, Proc. of 19th International Symposium on Electromagnetic Fields in Mechatronics, Electrical and Electronic Engineering (2019) [3] K. Ito and K. Nakamura, IEEE Trans. Magn., doi: 10.1109/TMAG.2020.3020121 (2020)

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/2dd3339ff2411411f71aa3e80e14ad53.jpg)
Fig. 1 Specifications of the MGM used for examinations.

![](https://assets.underline.io/uploads/markdown_image/1/image/aee100ae06c85f4490a7c1b2f434fe0c.jpg)
Fig. 2 Calculation and experimental results, (a) Calculated waveforms of the armature voltage at various load angles, (b) Vector diagram considering the magnetic interaction, (c) Calculated results of the displacement power factor, (d) Measured results of the displacement power factor.

 Influence of Magnetic Interaction on Power Factor and Efficiency of IPM-type Magnetic-Geared Motor

Tidal stream offers predictable electrical energy generation at higher power densities than other renewable energy sources, such as wind. However, compared to wind turbines, the tidal turbine technology is arguably still in its infancy, and various types of drivetrains are being considered [1][2]. These include permanent magnet (PM) generators and doubly-fed induction generators. The paper presents an investigation into the application of magnetically geared Pseudo Direct Drive (PDD) electrical machine to a floating tethered 6kW, 100rpm marine current turbine. The turbine is designed to harness the energy from tidal and ocean currents, in regions with relatively high-speed water flows at depths, that are a long distance from the seabed, and cannot be economically exploited by conventional fixed turbines. The marine current turbine is a robust, easy to service and cost-effective solution for regions with variable or fixed sea depth of more than 60 metres. Fig. 1 shows the prototype tidal current turbine and cross-sectional view of the PDD generator, where the magnetic gear element consists of a 4-pole pair high speed PM rotor, 31 ferromagnetic pole piece output rotor and a 27-pole pair stationary Halbach PM array. The high-speed rotor of the magnetic gear is magnetically coupled to the stator, which employs a double layer concentrated winding, forming the PM brushless machine element. The PDD generator was designed for a rated torque of 580Nm at 100rpm, while the pull-out torque requirement of the magnetic gear was fixed at 640Nm. Fig. 2 shows the PDD generator efficiency for a range of input load torque and speed points. It can be seen, that the 6 kW PDD generator is capable of an efficiency of 94-95% for a substantial range of operation. The paper will present the electromagnetic design and analysis of the PDD generator, and predicted performance such as efficiency, will be compared with dynamometer measurments on a prototype PDD generator. Furthemore, test results on a demonstrator marine turbine in a towing tank will also be presented.

**References:**

[1] S. Benlghali, M. E. Benbouzid, J. F. Charpentier, IEEE Transactions on Oceanic Engineering, Vol. 37 (2012) [2] K. Touimi. M. Benbouzid and P. Tavner, IEEE International Power Electronics and Application Conference and Exposition (PEAC), (2018)

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/bd094d2afbc6f9458474c8afa56df111.jpg)
Fig. 1 Prototype tidal current turbine and PDD generator.

![](https://assets.underline.io/uploads/markdown_image/1/image/a6d8d36d60a6cd3fffebf64d92c3b836.jpg)
Fig. 2 Variation of PDD generator efficiency with torque.

 Pseudo Direct Drive Electrical Machine for a Floating Marine Current Turbine

Currently, subsea remote operated vehicles (ROVs) are equipped with hydraulically or electrically powered thruster units. For the former, the surface powered on-board electric motors are only used to drive hydraulic pumps, which distribute hydraulic power to hydraulically driven thrusters and auxiliary toolings [1]. For the elecrically powered thruster, direct-drive brushless permanent magnet (PM) machines are normally employed [2][3]. A novel all-electric ROV, which employs 25kW electric thruster units, exhibits increased system efficiency, in addition to a 20% reduction in weigh/size of the power delivery and drive systems, when paired with a high voltage DC supply umbilicals. Thus, enabling operation from smaller vessels. This paper presents an investigation into the electromagnetic design and analysis of a magnetically geared thruster motors, which can operate in closed loop control providing accurate tracking of designated routes or position holding of the ROV. Additionally, the novel thrusters provide 20% more thrust with 50% fewer moving parts, reducing the maintenance requirement, while providing hot-swap capability to reduce down-times of the ROV. The torque limit capability of the magnetic gear unit can passively isolate the propeller from the electric motor and re-engage normal operation when the excess load, on the output shaft, has been removed. In addition, the hermetically sealed magnetic gear topology, isolates the active electrical part of the thruster from the passive output stage of the magnetic gear. Thus, sea water ingress will not affect normal operation and the oil leak severity is reduced, since the output stage of the gear only requires a small quantity of oil for pressure compensation. Fig. 1 shows the main components of the novel magnetically geared thruster, and a cross-sectional view of the magnetic gear connecting the high-speed PM motor to the propeller, which is driven by the low speed rotor (LSR) of the magnetic gear. Furthermore, high-speed side of the thruster is hermetically sealed from the output LSR by the stationary pole piece structure. The high speed rotor (HSR) of the magnetic gear is rigidly connected to the rotor of the PM motor, which is designed for a rated output of 25kW. Fig. 2 shows the predicted efficiency map of the magnetically gear thruster. It can be observed that a high system efficiency is possible, despite the entire thruster being filled with mineral oil for pressure compensation. The paper will present the electromagnetic design and analysis of the magnetic gear and PM motor, and predictions are compared with experiments undertaken on a prototype. It is shown for the magnetic gear an airgap shear stress in excess of 100kPa can be achieved.

**References:**

[1] F. Zhou et al. OCEANS, 2013 [2] D. Uygun et al. IEEE 5th International Conference on Power Engineering, Energy and Electrical Drives (POWERENG), 2015. [2] H. Tian, Z. Mao, B. Song and Y. Li, OCEANS, 2018.

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/07a2ec71341290e2b0ff97f4389d3748.jpg)
Fig. 1 a) Magnetically geared thruster topology (b) Magnetic gear cross-section

![](https://assets.underline.io/uploads/markdown_image/1/image/39402114b3ca13f31cd673dc540843e5.jpg)
Fig. 2 Efficiency map of the thruster system

 Magnetically Geared Propulsion Motor for Subsea Remote Operated Vehicle

I. Introduction The conventional modulating ring in a coaxial magnetic gear is composed of a sequence of ferromagnetic blocks. The rotating cylinder planetary gear (RCPG) proposed by Davey et al borrowed the conventional topology but replaced the ferromagnetic blocks with free-spinning magnetized cylinders called planet gears [1]. The planet gears enhance the modulating effect and hence the torque density, since they themselves are a source of magnetomotive force (MMF). Davey's team reported 79% increase of torque density over a conventional coaxial gear of comparable dimension [1]. In the operation of the RCPG, the planet gears move to new angular position corresponding to each updated relative position of the two rotors. It is a prerequisite to know these angular positions of planet gear for the calculation of the performance such as torque. Since the planet gears are unconstrained, they move to zero-torque positions. However, finding these positions is mathematically a difficult optimization with a large number of unknowns and strong internal coupling. A custom optimization algorithm combining splines, quadratic interpolation and response surface method was implemented for these positions that minimize the maximum torque on the planet gears. This algorithm was proved to be the most effective though it converged more slowly than the other algorithms that Davey's team had tried [1]. Alternatively, Wang et al proposed that the total magnetic energy stored in the gear should be minimum (or maximum magnetic coenergy) when all the planet gears settle in their equilibrium positions [2]. Wang's team found the positions of planet gear corresponding to the minimum stored magnetic energy, by using gradient-based Broydon–Fletcher–Goldfarb–Shanno (BFGS) and three preconditioning techniques, respectively. The BFGS algorithms converge apparently faster than Davey'ss custom algorithm. However, two of the preconditioning techniques need flux-tracing finite element (FE) calculation with or without non-magnetized cylinders. The other technique seems analytical and consequently saves the preconditioning cost, but it was not described in detail [2]. This digest proposes a BFGS algorithm with an alternative analytical preconditioning technique for the angular positions of planet gear corresponding to the minimum stored magnetic energy. The algorithm can quickly find these angular positions for each relative position of the two rotors in the operation of the gear. II.Description of Proposed Preconditioned BFGS The proposed preconditioning technique estimates the angular position of a planet gear depending on the polarities of the closest neighboring magnets on the rotors. This estimate can be easily formulated if the two closest neighboring magnets have different polarity, but it is difficult for some special relative positions of the planet gear and the two rotors. For example, the closest neighboring magnets have same polarity, as shown in Fig. 1 (a) and (b), or the distances from the planet gear to the two neighboring magnets on a rotor are equal, as shown Fig 1 (c) and (d). The angular position of planet gear (given by the direction of polarity, shown as the red arrow in Fig. 1) in each special case is found by the BFGS without preconditioning. The preconditioned BFGS relies on this result to formulate the presetting of angular position for each special case in the performance calculation of RCPG. The calculation of gradient in the BFGS algorithm was also improved to ensure the convergence to the global optimum. The formulation of the proposed preconditioning and the improved gradient calculation will be given in the full paper. III.Performance Calculation With Preconditioned BFGS The proposed preconditioned BFGS was used to calculate the torque for each relative position of the planet gears and the rotors in the running of a magnetic gear with 4 inner poles, 40 outer poles and 22 planet gears. Randomly initialized and flux-tracing preconditioned BFGS algorithms were also used respectively to calculate the torque. The latter algorithm makes two-dimensional (2-D) FE calculation of the flux distribution in the magnetic gear with the conventional modulators of the same dimension as the planet gears, and initializes each planet gear's angular position for the BFGS optimization according to the field direction in the center of the corresponding conventional modulator. Fig. 2 (a) compares the performances of the three BFGS algorithms for the optimization at a certain relative position of the planet gears and the rotors. The randomly initialized, flux-tracing preconditioned and proposed preconditioned BFGS algorithms made 21, 16 and 14 iterations, and calculated the stored magnetic energy by FE for 474, 348 and 279 angular position cases in total, respectively, until the convergence. The running of the magnetic gear with different initial angles of the inner rotor was simulated by FE and the proposed preconditioned BFGS. Fig. 2 (b) shows the torque-initial-angle curve given by this simulation. IV.Conclusion This digest proposes an analytically preconditioned BFGS algorithm for the performance calculation of RCPG. With the preset angular positions of planet gear, the BFGS converges fast and significantly saves the cost for the FE simulations of the full-range operation of RCPG. The full paper will discuss the application of other algorithms to the analysis of RCPG.

**References:**

[1] K. Davey, et al, IEEE Transactions on Industry Applications, vol. 52, no. 3, pp. 2253–2260 (2016) [2] R. Wang, et al, IEEE Magnetics Letters, vol. 7, Art. no. 1303805 (2016)

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/24aefc2a2086278c9a4c40b19bcdde82.jpg)
Fig. 1. Special preconditioned cases

![](https://assets.underline.io/uploads/markdown_image/1/image/0e9bc3bec8e9ad352ffbbbf0931f3000.jpg)
Fig. 2. Performances of BFGS algorithm and torque capacity curve.

 Optimization Algorithms for Performance Calculation of Rotating Cylinder Planetary Gear

Axial flux magnetic-geared double-rotor machines (AMGDRMs) are characterized as a potential core component of the power splitting system in hybrid electric vehicles (HEVs). Conventional AMGDRMs suffer from low mechanical strength due to high axial magnetic force. In this paper, an interior-modulating-rotor axial flux magnetic-geared double-rotor machine (IMR-AMGDRM) is proposed to solve this problem. An analytical model of the IMR-AMGDRM is proposed to predict the magnetic field distribution and electromagnetic performances. Since analytical model is based on 2-D geometry, a 2-D approximation of IMR-AMGDRM which has similar output performance to 3-D finite element analysis (FEA) is presented based on the permanent magnet (PM) magnetic energy equivalent principle. Utilizing subdomain method, the magnetic field distribution and electromagnetic performances are calculated. Finally, the accuracy of the proposed analytical model is validated by 3-D FEA. Index Terms—Analytical model, axial flux, magnetic-geared machine, double-rotor machine. I. Introduction Axial flux magnetic-geared double-rotor machines (AMGDRMs), which enable high torque density, have been considered a promising candidate for power splitting in hybrid electric vehicles (HEVs) [1]. However, conventional AMGDRMs suffer from low mechanical strength due to high axial magnetic force. To enhance mechanical strength, an interior-modulating-rotor axial flux magnetic-geared double-rotor machine (IMR-AMGDRM) which can reduce axial magnetic force dramatically is proposed in this paper. Fig. 1 (a) and (b) illustrate the configuration of the proposed IMR-AMGDRM. As axial flux type machine, IMR-AMGDRMs have complex 3-D magnetic circuits. 3-D finite element analysis (FEA) can offer accurate results by taking into account the nonlinear factors such as magnetic saturation and flux leakage. But the major drawback of 3-D FEA is serious time cost. Analytical model which is much lower time-consuming provides an alternative for electromagnetic performance prediction in the initial design procedure. In this paper, an analytical model of the IMR-AMGDRM is proposed. The geometry of analytical model is derived based on PM magnetic energy equivalent principle to guarantee similar output performance to 3-D FEA. The solution to analytical model is calculated by subdomain method. To evaluate the accuracy of the proposed analytical model, the results computed by analytical model and that of 3-D FEA are compared. II. Configuration of IMR-AMGDRM In conventional configuration, the inner and outer rotors are the permanent magnet (PM) rotor and the modulating rotor, respectively. Since the modulating rotor is closer to the PM rotor than stator, the modulating rotor suffers from high axial magnetic force. Furthermore, the modulating rotor is hollow and thus the low mechanical strength of modulating rotor becomes the major drawback of conventional configuration [1]. To solve this problem, the IMR-AMGDRM which switches positions of the PM rotor and the modulating rotor is proposed. In interior-modulating-rotor configuration, as shown in Fig. 1 (a) and (b), the PM rotor is between the modulating rotor and stator, i.e. the inner and outer rotors are the modulating rotor and the PM rotor, respectively. Owing to that the PM rotor suffers from reverse axial magnetic forces from both sides simultaneously, the total axial magnetic force is dramatically reduced. Hence, although the outer rotor is hollow in both configurations, the axial magnetic force of interior-modulating-rotor configuration is significantly decreased, leading to higher mechanical strength and reliability. III.Geometry of 2-D Approximation Generally, the unrolled section of axial flux machine at mean radius is used as 2-D approximation. However, few references are found to evaluate the electromagnetic performance equivalence of such approximation. To make the electromagnetic performance of 3-D model and 2-D approximation equal, the amount of magnetic energy of the PM should remain unchanged after approximation, i.e. PM magnetic energy equivalent principle [2]. A 2-D approximation of IMR-AMGDRM which has identical output performance is presented by using PM magnetic energy equivalent principle. IV. Analytical Calculation of IMR-AMGDRM The proposed analytical model is illustrated in Fig. 1 (c). It can be seen that the model consists of four subdomains. Region i, I, II and III represent the ith slot of the modulating rotor, the inner air gap, the PM rotor and the outer air gap, respectively. Utilizing subdomain method, the magnetic field distributions in the inner and outer air gaps are obtained by the solution of Laplace’s and Poisson’s equations. Then, the electromagnetic performances such as no-load back EMF, torque and axial force can be derived from the calculated magnetic field distributions. Comparisons of computation accuracy between analytical model and 3-D FEA is shown in Fig. 2. As can be observed, the results of analytical model are well consistent with that of 3-D FEA, validating the proposed method.

**References:**

 [1] M. Q. Wang, D. C. Tong and P. Zheng, IEEE Trans. Ind. Electron., vol. 66, no. 1, pp. 806-817( 2019). [2] Kim J S, Lee J H, Song J Y, and Kim D W, IEEE Trans. Magn., vol. 53, no. 6, 8105304(2017).

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/6f461fd509eebd7cd34868edda2a46ab.jpg)
Fig. 1. IMR-AMGDRM. (a) Overall view of configuration. (b) Exploded view of configuration. (c) Analytical model.

![](https://assets.underline.io/uploads/markdown_image/1/image/464feedb2617c359e0d07874766cd2b5.jpg)
Fig. 2. Comparisons between analytical model and 3-D FEA. (a) Axial flux density distribution in the inner air gap. (b) Axial flux density distribution in the outer air gap. (c) The static torque characteristic. (c) The static axial force characteristic.

 Analytical Modeling of an Axial Flux Magnetic-Geared Double-Rotor Machine with Interior-Modulating-Rotor

Magnetic lead screw (MLS) is a new type of linear actuator with high force density. Its output force depends not only on the arrangement of the permanent magnet (PM), but also on the materials used for the PMs, and the different materials of the PM directly affect the cost of the MLS. Therefore, based on this, a proposed MLS is given based on the combination of PM structure and magnetization direction and the combination of different PM materials. Compared with the traditional structures, it can not only improve the output force of the MLS, but also reduce the cost of the MLS. The results show that the size of air gap, the size of PM and the proportion of different materials combinations have great influence on the performance and cost of MLS. Finally, the performance of the proposed structure is compared with the traditional structures. The results show that the output force is increased by 21.5% and the cost is reduced by 21.4% under the same volume of MLS and PM. Index Terms — MLS, PM structure, magnetization direction, output force, cost. I.Introduction In recent years, as a new type of high force density linear actuator, magnetic lead screw (MLS) has been widely used in many fields, ranging from CNC machine tools, wave power generation to artificial heart [1-3]. MLS is similar to mechanical lead screw, which can convert linear motion into rotary motion, and vice versa. Compared with mechanical lead screw, MLS has the characteristics of inherent overload, minimum friction and low maintenance [4]. But for the existing applications and MLS structure, in addition to its output force is still low, there is also a problem of high cost caused by the use of high-performance PM materials in order to improve the output force of the MLS. Based on the above problems, a proposed MLS is given based on the combination of PM structure and magnetization direction and the combination of different PM materials, which improves the output force and decreases its cost. Detailed proposed structure introduction and operation principle are introduced in the second part. In the third part, the influence of the main parameters on the output force and cost is designed and optimized, and finally the size parameters of the structure are given. Finally, the performance of the proposed model is simulated and compared with the traditional structures to verify the superiority and economy of the proposed model. II.Structure and Operation Principle Compared with the traditional MLS structures, the proposed structure is shown in Fig. 1. The blue block and red block are radially magnetized PMs, in which the blue block is rectangular shape and the red block is salient pole shape, and both are made of NdFeB material. Green and yellow block are axially magnetized PMs with rectangular shape, and are made of other materials. Compared with the traditional halbach magnetizing array, this structure has higher output force and lower cost, and the structure is not difficult to process. The detailed operation principle of MLS is introduced in this paper. Fig.1. The proposed model. III.Design and Optimization The main design point of MLS is how to design the 3D structure of PM. There are three kinds of processing and installation structures, which are cylindrical, spiral and discrete surface mounted. In this part, the three structures will be compared in detail and the advantages and disadvantages of the description, and finally choose the appropriate shape for processing. After the structure is determined, the structure of the PM is optimized, mainly discussing the influence of the size of the air gap and PM on the output force and its cost of the MLS. In addition, the combination of different materials also has a great impact on the performance and cost of MLS, such as NdFeB, Ferrite, SmCo and Alnico. Finally, the best set of structural parameters of the MLS is obtained through the optimization, and the specific description is shown in the full text. IV.Results and Conclusion After design, optimization and analysis, the output force of the proposed model is shown in Fig. 2. It can be seen that the output force has increased by 21.5%. Moreover, after calculation, the cost of the proposed MLS is reduced by 21.4%, and the weight of the proposed MLS is also reduced by 8.6% due to the use of the combined materials of NdFeB and Ferrite, in which the proportion of NdFeB is 75.9% and Ferrite is 24.1%. In particular, limited by the number of words and pictures, we choose the traditional MLS for comparison, in which the PMs are rectangular and are made of NdFeB material. Fig.2. Performance comparison of the proposed model.

**References:**

[1] R. K. Holm, N. I. Berg, M. Walkusch, P. O. Rasmussen and R. H. Hansen, "Design of a Magnetic Lead Screw for Wave Energy Conversion," in IEEE Transactions on Industry Applications, vol. 49, no. 6, pp. 2699-2708, Nov.-Dec. 2013. [2] N. I. Berg, R. K. Holm and P. O. Rasmussen, "Theoretical and Experimental Loss and Efficiency Studies of a Magnetic Lead Screw," in IEEE Transactions on Industry Applications, vol. 51, no. 2, pp. 1438-1445, March-April 2015. [3] J. Ji, Z. Ling, J. Wang, W. Zhao, G. Liu and T. Zeng, "Design and Analysis of a Halbach Magnetized Magnetic Screw for Artificial Heart," in IEEE Transactions on Magnetics, vol. 51, no. 11, pp. 1-4, Nov. 2015, Art no. 8108604. [4] Z. Ling, J. Ji, J. Wang and W. Zhao, "Design Optimization and Test of a Radially Magnetized Magnetic Screw with Discretized PMs," in IEEE Transactions on Industrial Electronics, vol. 65, no. 9, pp. 7536-7547, Sept. 2018.

**Images:**

![](https://assets.underline.io/uploads/markdown_image/1/image/813f2c217de6006d1e11f97bf88afb43.jpg)
The proposed model

![](https://assets.underline.io/uploads/markdown_image/1/image/dffae9c2b87389d02da5ebb1d06a6e7e.jpg)
Performance comparison of the proposed model

 Optimization Design of Performance and its Cost of a Novel Magnetic Lead Screw by Combination of Different Permanent Magnet Materials

Coaxial magnetic gears (MGs) create non-contact speed change by using a ferromagnetic slotted rotor between an inner and outer magnetic rotor with dissimilar magnet pole-pairs. In 2001, Atallah et al. demonstrated that a coaxial surface mounted permanent magnet MG with a gear ratio Gr = 5.5, could be designed to operate with a peak volumetric torque density of 100Nm/L [1]. This is approximately twice the continuous operating torque density of a motor. If a higher gear ratio is utilized, then the motor size can be reduced thereby increasing the overall performance of a magnetically geared drivetrain. To-date most studies have concluded that coaxial MGs are only capable of operating at a high volumetric torque density at a relatively low gear ratio, typically less than 1:8 [2]-[5]. The reason for the reduction in torque density at higher gear ratios is explained by the increased discrepancy between the inner and outer rotor magnet pole count as well as the pole pitch. For instance, Gouda [2] conducting a limited sweep analysis in which the magnet thickness and modulation thickness was varied and concluded that the peak volumetric torque density was close to a Gr=5 gear ratio and the torque density greatly reduced with increased gear ratios. More recently Gardner [3] came to the same conclusion. Gardner studied three different coaxial MG designs and computed that the lowest gear ratio design, Gr = 4, had a significantly higher torque density. In many applications a high gear ratio is required. In order to design a coaxial MG with a high gear ratio multistage MG designs must currently be utilized [6]. However, using a multistage MG complicates the mechanical structure and decreases the torque-per-kilogram of the rare earth permanent magnets. The object of this paper is to further investigate the performance of higher gear ratio designs, Gr&gt;20. In this paper the authors demonstrate, for the first time, that a coaxial MG can be simultaneously designed with both a high gear ratio, Gr&gt;20, and high torque density. In order to do this a complete design sweep of three key radial geometric sweep parameters (ri1, ro1, l2) is conducted using the MG design illustrated in Fig.1. In this MG design the laminated (M19) modulator is supported using an inner bridge and G11 Garolite rods (not shown). A pole-slot combination defined as (p1, n2, p3) = (1, 22, 21) is used to demonstrate the performance capability. When the outer rotor is stationary, a gear ratio of Gr = n2/p1 = 22 is obtained. In the design study the outer rotor is composed of a four-segment Halbach array (N48) rotor structure and the inner rotor is composed of a single pole-pair created by a single diametrically magnetizing NdFeB (N48) cylinder magnet. Note that the torque ripple is mitigated by choosing a pole-slot combination that minimizes the greatest common divisor (gcd) defined as: CT = gcd(2p1, n2) [7]. However, when CT = 1, an odd number modulation rotor slot number is obtained but this then creates a large radial force on the cage rotor due to the unsymmetrical slot number. Thus, the design with a CT = 2 was selected. In the illustrated design a 2-D finite element analysis model was used in which the outer rotor radius was fixed at ro3 = 60mm and the axial stack length was d = 60mm. The active material mass and volumetric torque density performance for the design example is shown in Fig. 2. Fig. 2 was created by simultaneously sweeping the three radial parameters. The inner rotor values were swept across values: (ri1, ro1)=([8, 10, … 30], [30, 32, … 50]) and the cage rotor radial thickness was simultaneously swept across the range shown in the legend of Fig. 2. The resulting pareto-front is evident. The peak volumetric torque occurred when (ri1, ro1, l2) = (8 mm,44 mm, 5/16”) and this yielded a peak computed volumetric and mass torque density of Tv = 380.5 Nm/L and Tm = 59.4 Nm/ kg respectively. The peak torque for the design was T = 258.2 Nm. The volumetric and mass torque density were computed using Tv=T/(dπro3^2) and Tm = T/(mm+ ms), where mm = mass of the magnets and ms = mass of the laminations. By studying Fig. 2. it can be seen that a design can be obtained that has an impressively high torque density whilst still having a high gear ratio. The authors believe this has not been demonstrated before. In order to understand the reason for this high performance, the full paper presents a comparison in performance for the lower performing high gear ratio design presented in [2] – [5] with the MG design presented here. A field and torque harmonic analysis are used to demonstrate that whist a higher gear ratio does lead to a lower torque density, if a complete sweep analysis is used, then the reduction in torque density performance with a high gear ratio is much less than previously demonstrated.

**References:**

[1] K. Atallah and D. Howe, IEEE Trans. Magn., vol. 37, no. 4, p. 2844, July 2001. [2] E. Gouda, S. Mezani, L. Baghli, IEEE Trans. Magn., vol. 47, no. 2, p. 439, Feb. 2011. [3] M. C. Gardner, M. Johnson and H. A. Toliyat, 2018 IEEE Energy Conv. Congress and Expo., Portland, OR, 2018, p. 5005. [4] D. J. Evans, Z. Q. Zhu, 2011 IEEE Intern. Elect. Mach. Drives Conf., Niagara Falls, ON, 2011, p. 1403. [5] E. Park, C. Kim, S. Jung, Intern. Conf. Elect. Mach. Sys., Jeju, Korea, p. 2529, Oct. 7-10, 2018. [6] M. C. Gardner, M. Johnson, H. A. Toliyat, IEEE Trans. Energy Conv. Vol. 34, no. 2, p. 665, June 2019. [7] G. Jungmayr, J. Loeffler, B. Winter, IEEE Trans. Ind. App., vol. 52, no. 5, p. 3822, Sept.-Oct. 2016.

**Images:**
![](https://assets.underline.io/uploads/markdown_image/1/image/ec91cbc6f43b7fed3076279e99e01552.jpg)

![](https://assets.underline.io/uploads/markdown_image/1/image/a250a46fc226ab68f10790bf829ca181.jpg)

Downloads

Next from INTERMAG 2021

Design of a Stiffness Control Actuator Utilizing Magneto-Elastic Actuation

Similar lecture

Influence of DC/AC Winding Split Ratio on Electromagnetic Performance of Hybrid Reluctance Machine with Synthetic Slot PM Excitation

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from INTERMAG 2021

Design of a Stiffness Control Actuator Utilizing Magneto-Elastic Actuation

Similar lecture

Influence of DC/AC Winding Split Ratio on Electromagnetic Performance of Hybrid Reluctance Machine with Synthetic Slot PM Excitation

Downloads