Lab inventor puts new twist on ball bearings

Livermore physicist and inventor Dick Post has an alternative to the ball bearing, a part that is found in nearly every moving machine. Post's design, called the passive magnetic bearing, will eliminate the friction that is created when metal bearings roll against each other and will let them last longer, with fewer breakdowns. Such ball bearings would be particularly useful for long-running, low-maintenance machines located in remote or not easily accessible places such as space.
Magnetic bearings have been around for some 15 years. In these bearings, two powerful, repelling magnets are attached to the two bearing parts-one part turning and the other remaining in place-to prevent them from touching and creating friction. But the magnets need help to stay centered to each other. In existing magnetic bearings (called active magnetic bearings), a series of electronic sensors pushes the turning magnet back in place when it starts to slip to one side or the other of the in-place magnet.
Post's passive device uses no electronics and instead is based on a fundamental rule of physics: when a wire moves through a magnetic field, it creates an electrical current, and conversely, an electrical current moving through a wire creates a magnetic field around the wire. Post's device uses a magnet on the unmoving part of the bearing and a bundle of wires on the turning part. In the vicinity of the magnet, the turning motion creates an electrical current, which in turns creates a magnetic field. The induced magnetic field both repels the magnet on the static bearing part and pushes the wire bundle toward the center to keep the magnet arrangement equalized.
Post is cooperating with Trinity Flywheel Power, a company in Livermore, to translate his design into practice. Trinity's machines store energy; the new bearing would make it possible for the machines to run continuously for some 20 years, virtually maintenance-free.
Contact: Richard Post (925) 422-9853 (

Sensing perfection in paper

A system for monitoring moisture in paper has been developed by two Lawrence Livermore engineers to help detect moisture irregularities during the manufacturing process and correct them immediately, before large quantities of off-quality paper are produced.
The system devised by Jose E. Hernandez and Jackson Koo uses two linescan cameras mounted directly on a paper machine to continuously monitor the full width of a sheet of paper as it is being made. The camera technology is new and is based on an indium-gallium-arsenide linear array that measures near-infrared light. Hernandez and Koo have selected specific wavelengths sensitive to the higher moisture levels of paper early in the manufacturing process, so monitoring can begin earlier. Their initial results are promising and are catching the attention of paper manufacturers.
Hernandez and Koo started their project in 1996 as part of the Industries of the Future program sponsored by the DOE Office of Industrial Technology. The program creates partnerships between industry, government, and laboratories to identify technologies that can improve the energy efficiencies of nine energy- and waste-intensive industries.
The moisture monitoring system will help paper manufacturers reduce energy by not overdrying paper and minimize waste generation by detecting and correcting moisture problems before large quantities of defective paper are produced.
Hernandez and Koo built a prototype benchtop system in 1997, and last year they teamed up with ABB Industrial Systems in Ohio to test their new system on moving paper. "We wanted to compare our measurements with that of a commercial scanning unit," said Hernandez. "The folks at ABB were very impressed, considering that our sensor technology is substantially different from theirs," he added. This fall, Hernandez and Koo will test two new cameras and conduct more controlled experiments to quantify how well their system can measure moisture in different types of paper.
Contact: Jose E. Hernandez (925) 423-2160 (

Many asteroids are rubble piles

Mark Hammergren, a planetary scientist at Lawrence Livermore, has studied nearly 850 asteroid observations by astronomers and come to this conclusion: elongated or stretched asteroids are apparently weaker than spherical ones. They are never seen to be rotating faster than once every 4 hours, while the more spherical asteroids can rotate as fast as once every 2.3 hours. Hammergren said that this observation supports the theory that most asteroids are not solid chunks of rock tightly bound together but rather are loose aggregates of materials called "rubble piles." The elongated asteroids, with weaker gravity at their ends, cannot rotate faster or their piles of rubble would break up.
Hammergren theorizes that rubble-pile asteroids are governed by the same processes that lend stability to piles of sand on Earth. The rubble piles have the ability to support large surface features on asteroids, just as loose sand and weak dirt can support mountains on Earth. Hammergren also thinks that changes in a rubble-pile asteroid's shape would happen cataclysmically as a series of massive landslides and that if such landslides were to occur on the surfaces of rapidly rotating asteroids, fragments of the asteroid's surfaces may be thrown off into space to form asteroid moons.
Hammergren presented his findings at the centennial meeting of the American Astronomical Society in Chicago earlier this year.
Contact: Mark Hammergren (925) 423-0737.
Back to October 1999