Abstract
Although the researches into increasing the transmission torque and the torque density of magnetic gears are performed, setting the torque to the required value after assembling is almost not done. In this study, we examined adjusting the step-out torque by inserting a plane rectangular coil between opposed magnetic gears. It is possible that the magnetic field generated from a coil decreases the step-out torque. Therefore, the relationship between the characteristic value of the coil and the transmission torque was investigated experimentally. The theoretical value of the transmission torque decrease rate by calculation and the experimental result are compared. When the rotational frequency was less than 700 rpm, the transmission torque reduction rate by the coil was about the same as the calculated value, and adjusting the step-out torque by the coil is expected.
Introduction
A magnetic gear is a mechanical element that transmits the torque and rotation using the attractive and repulsive forces of the magnets. So, many researches paid attention to magnetic gear. It can decrease the wear, the dust, the vibration and the noise caused by the contact of the gear tooth compared to the mechanical gear. In addition, it is expected to be used in the place where needs a clean environment because it is not necessary to use lubricants. Moreover, the magnetic gear has the torque limiter effect that happens step-out and runs idle in case when the excessive torque is inputted. Due to the torque limiter effect of the magnetic gear, when the overload is applied to machines, step-out of the gear can prevent machines from damage. However, the step-out torque is decided by the magnet strength and distance of gears at assembling.
Many researches investigated properties of the magnetic gear. They include proposals of new types of magnetic gear [1–5], that increase the transmission torque and the torque density [6–10], reduce the torque cogging [11], accurately position [12–14] and so on. However setting the torque limit to the required value after assembling is almost not achieved. If it is possible to change the limit torque after the assembling the magnetic gear, it is useful to apply the magnetic gears and to use the torque limit function. The authors have been to try to adjust the step-out torque after the assembling by changing the distance and facing length in the cylindrical magnetic gear [15]. Although this method can make a big change of the torque limit, it leads to machine upsizing because it necessary to include the adjustment mechanisms. Therefore, in this study, as a simpler method of adjusting the torque, we examine inserting the rectangular coil between the facing magnetic gears. Furthermore, we investigated the effect of adjustment by the coil to be inserted.
Principle of adjusting torque
The step-out torque is changed according to the following principle. A plane rectangular coil is inserted between the magnetic gears (Figs 1, 5). When the magnetic gear rotates, the N poles and the S poles appear alternately between the magnetic gears. As the polarity of the magnetic gear changes, the magnetic flux penetrating the coil placed between the gears changes. According to the Faraday’s law of induction, an induced electromotive force is generated in a direction canceling the magnetic flux, thereby electrical current flows through the coil. Consequently, a magnetic field is generated inside the coil according to Biot-Savart law. The transmission torque between the magnetic gears is changed by the generated magnetic field because the magnetic gears transmit the torque by magnetic connection. Since the magnetic flux generated from the coil changes so as to cancel out the magnetic flux used for transmission, the magnetic flux density between the gears is weakened. So that the transmission torque is reduced.

Experimental setup.
Figure 1 shows the structure of the experimental setup. Figures 2 and 3 show the cylindrical magnetic gear used in this research. The input side is driven by the motor. The output side is connected to the powder brake as a load. Each torque is measured by detectors connected to the input and output shaft.

Photograph of magnetic gear.

Structure of magnetic gear.
Figure 2 shows the photograph of the cylindrical magnetic gear. The latter consists of the central electromagnetic soft iron and the neodymium magnets attached to the outside with no gap between each other (Fig. 3). The number of magnets is 8 and the number of pole pairs is 4. The facing length is the maximum since the gears are not offset in the axial direction. The same gears are used in both the input and the output sides. Thus the gear ratio is 1. In this research, the gap between the gears is 26 mm in order to insert a coil to investigate the possibility of adjustment.
Coil
Figure 4 shows the photograph of the coil used in this study. The coil has a rectangular hollow part of 10 mm × 20 mm at the center and the wire diameter is 0.5 mm. The winding number of the coil are of three types that are (a) 10, (b) 20 and (c) 40 turns. As shown in Fig. 5, the coil is inserted in the center of the gap of the magnetic gears in such away so that the longer direction is horizontal. In order at the center of the coil to maintain the same height with the center of the axis of the gear, the coil is held that is sandwiched between acrylic plates. The coil is a closed circuit with a single coil without connecting equipment, resistors and so on.

Photograph of coils.

Gears with coil.
The magnetic flux density of each side of the coil Bc is calculated from Eq. (1). Therefore, the magnetic flux density generated from the whole coil, is the sum of magnetic flux densities generated from each side.
The step-out torque was measured at each rotational frequency when the rotational frequency of the input side gear was changed. While keeping the rotational frequency constant, the current of the powder brake connected to the output side was gradually increased until step-out. The torque was measured at step-out. It was measured every 150 rpm from 100 rpm to 1000 rpm. From the data obtained by measurement, the average of 10 data just before step-out is treated as a step-out torque. Data from the torque detector is acquired at 10 msec intervals. Measurement was carried out five times under each condition, and the average value of five data was taken as the measurement result.

Input and output torque.
Step-out torque
Figure 6(a)–(d) shows the measurement results of the input and output torque in cases without the coil (called ‘unused’) and using the three types of coil (10, 20, 40 turns). These results are the step-out torque corresponding to each case. Because the gap between the magnetic gears is wide, the value of the transmission torque is not large. The purpose of this paper is to check ability to adjusting the step-out torque by using the coil.
At 100 rpm, the output gear stopped and the input gear was also stopped instead of stepping-out. As a result of the measurement, it was impossible to obtain a proper step-out torque. Although the input torque increases slightly as the rotation frequency grows, the output torque decreases in each case. With the same rotational frequency, the step out torque reduces as the winding number of the coil increases. These are considered as the effects of using coils.
Reduction rate by coil
The reduction rates of the output torque at each rotational frequency are shown in Fig. 7. They include the influence of both the rotation effect and the coil effect. So the reduction rate by the coil only is obtained by subtracting the measurement result without coil from the result using each coil. Figure 8 shows the modified reduction rates by the coil only and the calculated reduction rates from the Eq. (1).

Reduction rate.

Modified reduction rate by coil.
In the comparison between the calculated value and the modified measured value, the rate of reduction are about the same in the range of the rotational frequency of 250 rpm to 550 rpm. At the time of step out, the output side gear is stopped; by contrast the input side gear is rotating. Therefore, attractive force and repulsive force are alternately generated between the gears, so vibration is generated on the input shaft side that is rotating. With the high speed rotation, it is considered that the reason why the measurement result is smaller than the calculated value is that the vibration at the time of step-out is larger than with low speed rotation.
In this study, we examined adjusting the step-out torque by inserting a plane rectangular coil between the facing magnetic gears. The findings obtained by comparing the measured value and the calculated are shown below.
It confirmed the adjusting range of step-out torque being expanded by inserting the coil between the magnetic gears to change the transmission torque. When the rotational frequency was less than 700 rpm, the transmission torque reduction rate by the coil was about the same as the calculated value. Therefore, it is considered that it is possible to select the coil for required step-out torque by calculation in advance. Thus, it is inferred that it is possible to adjust the step-out torque if the selected coil is used. The coil is merely inserted between the gears and it is not controlled. If only changing the step-out torque slightly, it is inferred adjustment can be made at a relative low cost.
In this report, the results with the coil alone are shown. In future outlook on research, the effect of coil size should be treated. Also we plan to apply the electrical current to the coil synchronously with the gears rotation. It is expected that by passing current through the coil, it is possible to not only greatly adjust the step-out torque, but also to improve the transmission efficiency by decreasing the reduction rate as compared with the case without using the coil.
