Rotating Machinery and Controls Laboratory




Current Research


Current Members


Member Services


ROMAC People


Annual Meetings


Short Courses





Passive Magnetic Bearing for Flywheels

Student: Alexei Filatov

Advisor: Eric Maslen

Project Start Date: March 2000


Project Objectives:

Develop a new type of magnetic bearings capable of non-contact supporting high-speed rotors without using external energy sources and external control systems. The expected advantages of the new bearing are high efficiency, reliability and low cost. One of the possible applications is in flywheel energy storage systems, which are considered to be an attractive alternative to electrochemical batteries due to higher stored energy density, higher life term, deterministic state of charge and ecologically clean nature. While demand for reliable and inexpensive energy storage systems grew up dramatically in the last few years because of rapid progress of computer systems, which require higher quality of energy supplies than can be provided by public electrical networks, the progress in development of flywheel energy storage systems was in large degree hold by the lack of reliable, efficient and inexpensive high-speed bearings.

Figure 1. Rotor of the Bearing


Figure 2. Bearing Prototype

and Test Rig

Progress in the Past Year:

We have developed a structure of the bearing, theory of its operation and built a bearing prototype in which non-contact stable suspension of a 3-kg rotor was realized when the rotor rotated above a critical speed of approximately 1200 rpm.

The idea of this project may sound incredible; it is about realizing non-contact suspension of a rotating body in the presence of external forces such as gravity forces without using control systems.  In other words, pretty heavy objects can levitate in the air or vacuum without touching anything and, moreover, almost no energy is required to achieve this effect.  Strictly speaking, the phenomenon of stable robust levitation without energy consumption and utilizing control systems was known for a long time.  It can be realized with application of superconducting materials at cryogenic temperatures.  The big advantage of the approach pursued in this project, however, is that superconductors are not required, and the system is operable at room temperature.  This makes it much more practical.  Moreover, the goal of the project is to develop a system which would be not only of theoretical interest, but whose parameters such as load capacity, stiffness, rotation losses, size, weight and price would be acceptable for applications.

One such application is in flywheel energy storage systems, which allow storage of energy in the form of kinetic energy of a rotating body(rotor) and which are considered to be an attractive alternative to conventional electro-chemical batteries.  Due to the non-contact nature of suspension offered by the proposed system, if it is used in a flywheel system to support the rotor, the rotor could be speeded up to a limit where it breaks into small pieces by centrifugal forces.  One can easily estimate that energy sufficient to tear a steel or a fiber-glass rotor into pieces is much, much bigger than energy which can be stored in an electrochemical battery of the size of the rotor.

Apparently, the system used to suspend a rotor must exhibit an extreme reliability and the proposed system does so indeed, due to its intrinsic stability, in contrast to active magnetic bearings it can be said that this system cannot fail.  The fact that its rotor levitates is ensured by physics laws with pretty much the same certainty as the fact that Newton's apple falls down.  Besides, the proposed system promises to be much easier to manufacture and much cheaper than active magnetic bearings.  The validity of the proposed theory has been recently demonstrated by building a prototype in which non-contact passive suspension of a 3.2kg rotor is realized when it rotates above 20Hz.

Please visit our website:

American Flywheel Project

Students: Bin Huang, Guoxin Li, Yingjie Liu, Brian Overby, Robert Rockwell,

Hai Zhang

Advisor: Paul Allaire

Research Staff: Michael Baloh, Wei Jiang, Jun-Ho Lee

Project Start Date: June 2000

Expected Completion Date: Sept. 2002

Sponsor: AFS Trinity Power

Project Objectives:

I want to describe an interesting energy storage flywheel project that we have been involved with for the last three years. The project is supporting a research sponsor, American Flywheel Systems (AFS), who is focused on the application of the high specific energy flywheels for both commercial and space applications. The commercial applications include uninterrupted power supplies (UPS), load-leveling energy storage equipment for residential buildings and industrial plants, and potentially electric automobiles at some point in the future. All of these applications can be supported while reducing the emission of green house gasses through improved electrical energy efficiency and eliminating environmental contamination from conventional battery chemicals. Space application is aimed at replacing chemical batteries and reaction control wheels with an integrated component that both stores energy and controls the attitude of the spacecraft.

The energy storage flywheel consist of high strength and lightweight composite material, a high efficiency motor/generator, and a low loss magnetic bearing utilizing advanced control theories implemented with real-time control and state-of-the-art electronic technology. In May of this year, the project successfully completed the second phase of a NASA sponsored Small Business Innovative Research (SBIR) contract. UVA was a key member of the AFS research team. The team received the Tibbetts Award from the Small Business Administration for the quality of its research. Several important goals have been achieved:

First, to simulate the energy storage flywheel, we built a flexible rotor test fixture on a flexible base structure and spun it up to 15,000rpm. The rotor successfully passed three rotor critical speeds, including two rigid body modes, one flexible mode, and several substructure modes.

Secondly, different control algorithms, including PID, H-infinity, Mu synthesis and Linear Parameter Varying control, were tested and compared based on their performance. A precise model of the test rig was developed using analytical (finite elements) and experimental methods.

Lastly, a real time control and monitoring computer system was developed and tested that achieved sufficient reliability and throughput to implement advanced state space controllers.

Progress in the Past Year:

The project is now moving into Phase III (commercialization) under AFS sponsorship and will develop both passive and active low loss magnetic bearing. In the next two years the project will implement the Phase II technology in a working energy storage flywheel unit. This will include both the design and test of flywheels, additional control algorithm development for linear parameter varying system, and sophisticated computer software and hardware development to incorporate advanced control algorithms. A new spin pit facility is being constructed at UVA by AFS to test the integrated technology in prototype flywheel systems. Once this research project is completed, the spin pit facility will be available to support other types of high-energy rotor testing at UVA. The efforts of UVA, supporting the work of AFS, will expedite the commercialization of high performance energy storage flywheels for the benefit our power thirsty society.


University of Virginia, Mechanical and Aerospace Engineering

122 Engineer's Way, P.O. Box 400746

Charlottesville, VA  22904

(804) 924-3292 Phone

(804) 982-2246 FAX