More specifically, the shaft can be steadily levitated and supported even without any control and simultaneously the stiffness and damping characteristics can be improved by adding a control function. This is a very efficient system which maximizes the advantages of the conventional bearings and overcomes the disadvantages.

BACKGROUND ART Most of the conventional passive magnetic bearings are based on attaining a stable levitating force caused by a reaction between two permanent magnets or one permanent magnet and an electromagnet that face each other at a close proximity. This is inefficient due to the reasons of low stability at the critical speed and an excessive distance which flux must travel through the air. Moreover, it requires a design which takes care of the unidirectional reaction force.

Although it has attracted a lot of attention from various areas of the industry in the mid 1980s, the commercialization of the active magnetic bearing came much too later than anticipated and its application is also limited to a few narrow areas. Mostly, there are two reasons for this, firstly, it is the low credibility from the users side due to its inherent instability, secondly, it is the high cost equipment involving non-contact variable sensor, high current amplifier and digital controller.

As a result, the research on the active magnetic bearing has been mostly focused on robust control, self-diagnostic and sensor replacement technology.

However, it has not reaped many harvests in terms of commercialization.

Amid the recent trend towards high speed rotation and miniaturization, an active magnetic bearing, that uses the Lorentz force which has the advantages of virtually negligible hysteresis and eddy current loss, and high efficiency at high rotation speeds, is introduced.

DISCLOSURE OF INVENTION The present invention is designed to overcome the above problems of prior art. The object of the invention is to provide a passive and active magnetic bearing where the coil with a running DC current would not be displaced from the designated position due to the Lorentz force by designing the air gap such that its flux density varies according to the position in the air gap and at a time, enabling an active control by measuring the movement of the rotor.

According to the present invention, the passive and active magnetic bearing comprises a permanent magnet which is located at the center of a rotor and an air gap that exists at the outside of the rotor, and a flux route in which the flux generated from the

North pole of a permanent magnet travels through the air gap at the outside to reach the South pole.

The air gap is designed to be narrower at the center and gets increasing wider along the outwards radial direction so that the flux density of the permanent magnet varies according to the radial position.

The coils located at the center of air gap are firmly fixed to a stator. If the rotor is displaced from the pre-designated position while a fixed current is running through the coil, the flux density between the two coils that face each other varies. As a result, the equilibrium in the Lorentz force breaks down causing a restoring force of the coil which results in the stable levitation of the rotor.

In addition to the function of a passive magnetic bearing as mentioned above, if the current running through the coil can be controlled by measuring the location of the rotor, it can also perform as an active magnetic bearing.

BRIEF DESCRIPTION OF THE DRAWINGS FIGS. la and lb show a cross section view of the passive and active magnetic bearing according to the

present invention.

FIG. 2 shows the principle of restoring force of the passive and active magnetic bearing.

FIG. 3 shows an example of a cross section view of the axial direction passive and active magnetic bearing when the principle of the present invention is applied to an axial direction bearing.

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

FIGS. la and lb show a cross section view of the passive and active magnetic bearing according to the present invention.

FIG. 2 shows the principle of restoring force of the. passive and active magnetic bearing.

FIG. 3 shows an example of a cross section view of the axial direction passive and active magnetic bearing when the principle of the present invention is applied to an axial direction bearing.

As shown in FIGS. la and lb, a rotor 10 at the center of magnetic bearing and each coil 14 is firmly fixed to a stator. The flux generated from the north

pole of a permanent magnet 12 which is inserted to the rotor 10 travels to the south pole of the permanent magnet via a air gap outside.

Here, a special attention has to be paid to the shape of the air gap. The width of the void gets increasingly narrower along the inward radial direction and the flux density is higher at the narrower side. More specifically, a larger Lorentz force is applied at the narrower side even if the same current flows.

As shown in FIG. 1, if the same current flows through all the coils while the rotor is place at the exact position, the coils are under the same force that pushes them outwards. As a result, reaction forces are generated around the rotor which push the rotor 10 inwards and consequently are canceled each other out.

However, if the rotor 10 moves slightly to the right as shown in FIG. 2, the right coil 24 is located in a narrow side and the left coil is located in a wide side. As a result, the equilibrium in the Lorentz force breaks down and the force which acts on the right coil becomes larger. Hence the force which pushes the stator to the right becomes larger. As a reaction, the rotor encounters a force which pushes it

to the left causing it to return to the original position, hence, maintaining a stable levitating position.

Although the stable stiffness increases with a larger variation rate of the air gap, there is a limit to it. The air gap can be designed in a third order function rather than a linear function according to the desirable variation rate.

For the purpose of the function of a passive magnetic bearing, all the coils can be connected and only one amplifier is sufficient. However, in case an active control is required, four amplifiers are necessary.

Other system construction for an active control is similar to the conventional active magnetic bearing systems.

Generally, if more than 3 coils are present in the air gap, it becomes possible to actively control the rotor. The control signals according to the movement of the rotor can be added to the coils with a running current in order to perform an active control.

Also, the cost of manufacturing can be reduced since the speed can be measured or deduced instead of the displacement signal to be used as a control input.

The reliability of the system can be improved since

there is no fatal damage even in case of a controller malfunction.

According to the present invention, in order to fully levitate the rotor, a 5-axis support is required, hence, two magnetic bearings along the radial direction as shown in FIG. 1 and an additional axial direction magnetic bearing are necessary.

As shown in FIG. 3, the basic principle of the axial direction bearing is as explained previously.

The numeral 30 on the drawing represent a rotor, 32a, 32b are permanent magnets and 34a, 34b are coils.

Only a difference is that the direction of Lorentz force is rotated in axial direction. In this case, a single body type coil in the shape of a donut can not be utilized.

The passive and active magnetic bearing according to the present invention can establish a levitation stability without a controller, and possess the advantages of an active magnetic bearing such as precision location control, high speed rotation and variable dynamic characteristic and has a high stability and efficiency while free from hysteresis and eddy current loss. Also, there are advantages of having a high reliability while functioning as an active bearing and a free transfer between an active

type to a passive type is possible upon request.

Also, the present system can be used as a bearing element as well as various type of levitation systems and actuators. Specifically, if it is used as an oscillator which continuously vibrates at a fixed frequency, an efficient design is possible by maximizing this characteristic.