Figure 1 -  A typical ball bearing.  Photo courtesy of The Timken Company

Magnetic bearings are an attractive alternative to ball bearings for several key reasons.  Magnetic bearings provide electromagnetic suspension of the inner race, therefore, replacing the ball component with an air gap.  An obvious advantage of magnetic bearings is that there is no physical contact between the inner and outer race, eliminating problems cause by friction such as maintenance, wear, and failure issues.  Magnetic bearings do not require lubrication, as ball bearings do.  An electromagnet levitates the inner shaft.  A sensor constantly monitors the position of the inner shaft, sending feedback back to the power supply.  The power supply then adjusts the current in the electromagnet in order to keep the inner shaft centered.  In complex systems, this process could happen thousands of times a second.  An exterior force is used to spin the inner shaft, or in our case, the steel ball.  Our sponsor is particularly interested in the design of magnetic bearings for flywheel applications.  For flywheel operation, a motor is used to spin a disc, creating kinetic energy.  The kinetic energy is stored in the spinning disc until it is needed, at which point the kinetic energy is converted back into electrical energy.  Magnetic bearings are ideal for high speed applications such as flywheels, because they can run at much higher speeds than ball bearings because there is no friction.

Figure 2 -  A radial magnetic bearing.  Photo courtesy of SKF.

Magnetic bearings consist of three main components; a magnetic actuator, a sensor, and a feedback controller.  To understand how magnetic bearings work, one must first understand the role of the three components used in levitation.

 The Magnetic Actuator:

 The purpose of the magnetic actuator is to provide a force which will bear the load that is being supported.  Since the bearing force is going to be magnetic, the object that is being suspended must be able to interact with magnetic forces.  This means the load must be made of either magnetic material or ferromagnetic material.  In theory, if you wanted to support a load under perfect conditions, a permanent magnet could be placed the correct distance away from an object to offset gravity or whatever other forces are acting on the object.  However, due to the nature of magnetic forces this is nearly impossible to achieve in practice and would not be very useful, since it would fail as soon as the load conditions changed.  This does not mean that permanent magnets are of no use at all.  A weak permanent magnet can be used to help offset some of the known loading, such as gravity.  The use of a permanent magnet helps to minimize current in very high load applications. 

 In practice, at least some of the magnetic force must be variable, therefore electromagnets must be used.  An electromagnet is created by wrapping wire about some type of core.  When current is run through a wire a magnetic field is created around the wire.  Since the direction of the field is related to the direction of current flow, wrapping the wire in a circular pattern directs all of the magnetic field into the center of the loop of wire, as shown in Figure 3.

.

Figure 3 - Magnetic field pattern

If the wire is wrapped many times, the magnetic fields will continually add, creating a strong magnetic field in the center of the wire loops.  If this wire is wrapped around a ferromagnetic core such as iron, the core acts to focus the magnetic field and creates a stronger magnet as well.  The other advantage of an electromagnet is that the strength of the field is related to the amount of current flowing through the wire, so the electromagnet can be adjusted to handle different loading conditions.  By itself, or in combination with permanent magnet, the electromagnet is the crucial element of a magnetic actuator. 

The Sensor:

The electromagnet described above gives the user the ability to easily adjust the strength of the magnetic field, but one needs to know when and by how much to adjust the strength in a real bearing application.  The sensor provides this information.  Rather than trying to detect the amount of force needed, it is only necessary to detect the distance between the magnetic actuator and the load.  The electromagnet can be continually adjusted to maintain that distance even as the loading force changes.  Therefore, the sensor needed to accomplish this is a distance sensor.  There are many different types of sensors that detect the distance between objects.  Capacitive sensors, inductive sensors, optical sensors and acoustic sensors are just a few of the types of sensors that can make this measurement.  Since this report focuses on levitation of a steel ball, the best sensor will be an optical sensor.  Optical sensors allow the measurement of distance across relatively large gaps while remaining fairly cheap compared to other technologies such as acoustic sensing.  Since the ball is shiny steel, a reflective sensor will be effective. 

The controller:

The sensor provides a signal that is proportional to the distance between the load and the magnetic actuator.  When the sensor detects that the load has moved from its nominal position, the controller sends a signal to the actuator to increase or decrease the strength of the magnetic field to compensate for the motion and bring the load back to its nominal position.  Due to the nonlinearities of this system and the nature of the problem, the control system will have to be more complicated than a simple Proportional, Integral, Derivative (PID) controller.  The literature indicates that a feed forward design has been effective in the past for this type of problem.  This will most likely be the design approach taken in this project.

Once the ball is levitated, a system must be employed to spin the ball about the z-axis.  There is no standard way of initiating the motion of the ball, so this aspect of the design is very much open to the team’s innovation.  Potential methods considered by the team are discussed later in the proposal.

While the magnetic bearing offers many clear advantages over a mechanical bearing, there are several drawbacks worth mentioning.  Due to the electromagnet’s constant need for current, a power supply must be available to run the bearing.  The windings of the electromagnet often heat up, wasting energy and becoming less efficient.  Also, since the system is inherently unstable, considerable time and money must be spent on control for each individual application to ensure the bearing will not fail under normal operating conditions.

Despite these obstacles the magnetic bearing seems to be a promising alternative for many applications.  The most obvious of these is in the area of high speed flywheels.  Flywheels are used to store energy.  When the flywheel is spun up, the energy is stored in the inertial energy of the spinning mass.  To make this work efficiently, one wants as little energy loss in the bearings as possible.  Since magnetic bearings do not have friction losses, it may be possible to design magnetic bearings that perform better than mechanical bearings.  Also, flywheels often operate at extremely high speeds, which cause a dangerous amount of heat generation in mechanical bearings.  Magnetic bearings alleviate that problem, because there is no friction, and therefore, none of the heat generation associated with it.  Since magnetic bearings do not require lubrication to run efficiently, they may also be useful to replace traditional bearings just for the environmental impact of using less oil.  

Project Objectives

Our main objective is to provide a practical, functional design of a magnetic bearing to Boeing.  This objective will be achieved through close coordination and teamwork between the team, faculty mentor, and Boeing.  Boeing has set the following project objective for the team:  Design, develop and demonstrate the use of a magnetic bearing to levitate a 2-inch diameter ball bearing.  Once the ball is levitated, spin the ball up to 200 rpm and let the ball spin freely demonstrating the value of a magnetic bearing.

Each team member recognizes the sponsor's generous contribution of both time and money and is dedicated to making this experience as rewarding to the sponsor as it will be to the design team.

The team's objectives are further laid out in the team's Formal Proposal.

To the top