High Res TIFF (9.1 MB)

DYNAMX Vibration Isolation System Shake Test

This is a vibration isolation system designed to support a low temperature physics experiment while reducing vibrations from the Space Shuttle when in orbit. The project has been delayed and is now planned for flight on the International Space Station.

The test article failed during a test simulating the violent shaking expected during launch. This photograph illustrates the importance of testing new hardware before committing to use it during a space mission. Although such testing can be quite expensive, it is much less expensive than the cost of a mission. When designing a new system to perform under very harsh conditions, it is not possible to anticipate all possible failures. Testing is a cost-effective tool to identify such failure modes so the design can be modified to succeed.

This system was designed to operate at about 2 kelvin, very close to absolute zero temperature. This is colder than the temperature at which helium becomes a liquid at normal atmospheric pressure. The vibration isolation system was designed to support an experiment called DYNAMX, which will measure certain properties of superfluid liquid helium. The measurements will improve our understanding of quantum mechanics. The measurements are very sensitive to minuscule accelerations. That is why it is planned to perform the experiment in orbit. On either the Shuttle or the Space Station, there are vibrations from the crew moving about and from operating equipment. These vibrations, or accelerations, needed to be reduced by a factor of 10 to 20 below the typical vibration levels of these platforms. But during launch, the vibrations are about a quarter of a million times greater than the required operating point for the experiment. The vibration isolation system was designed to cover this entire range without any kind of launch latches that would normally be used with such a sensitive device. It would be difficult to use latches in the very cold, very small volume available. It is perhaps not too surprising that the first attempt to do this was not completely successful. We did learn a lot from the test and improved the design.

What Are We Looking At?
The cylindrical object in the middle, sitting somewhat tilted, is a dummy mass representing the sensitive experiment. The larger hollowed-out metal housing is a titanium support structure. Connecting the support structure to the aluminum dummy mass is a network of titanium flexures -- paper-thin foils of titanium -- many of which are broken in the photograph. The ends of these flexure ribbons were bonded using a structural adhesive appropriate for the low temperature. A careful analysis revealed that the failure was caused by peeling of one of these bonds. The design was modified to prevent this occurrence in future. The brownish rings around the base of the structure and the brownish cover in the background formed an insulated shroud around the test article. Liquid nitrogen was poured through a hole in the top of this shroud and onto the test article to cool it to cryogenic temperatures before shaking it. The pale green base-plate is a stiff mounting structure, but a poor heat conductor, to prevent cooling the shaker plate on which the entire assembly is mounted by the large Allen screws. The silvery objects with orange wires connected to them are accelerometers. These measure the acceleration of the base and of the suspended dummy mass. This particular model of accelerometer is qualified to operate at cryogenic temperatures. This test was arranged to shake the test article from left to right.

How Does the Vibration Isolation Work?
As designed, the ribbon flexures were arranged to be slightly wavy so the experiment would be supported very gently - floating an several very weak springs - to provide the vibration isolation needed in orbit. As the surroundings vibrate back and forth, the ribbons bend slightly, and the suspended apparatus stays almost stationary in the middle. But during launch, the vibrations are so much greater that the ribbons alternately snap straight and go loose. At all times, about half the ribbons support the mass. In this mode, the ribbons (being pulled in tension) act like very stiff springs with a small amount of dead-band. It was this aspect that was being tested in this particular test.