The Web Newsletter Front Page Hanford Livingston Caltech


Stiff Vibration Isolation at MIT

Stiff Vibration Isolation at MIT

- Contributed by Sam Richman

Summer is the vacation season, a time when many people elect to "get away from it all" and try to find a little peace and quiet in remote, unpopulated areas.

Gravitational wave experiments, too, are best conducted in remote, unpopulated areas so as to get away from all the artificial ground vibration that dense concentrations of humans tend to produce. Both LIGO sites are located in areas seismically much quieter than the typical urban environment. Even so, the first generation detector (LIGO I) must go to some lengths to shield its delicate optics from small ambient ground vibrations, and still the seismic motion will obscure gravitational wave signals at frequencies below 40 cycles/second. However, one of the goals of the next generation detector (LIGO II) is to see gravitational waves with frequencies as low as 10 cycles/second. This is a goal substantially more difficult to achieve.  Not only is the ground motion rather larger at lower frequencies, but traditional methods of isolation like those used in LIGO I are much less effective.  A new approach is called for.

There are many ways to achieve low-frequency vibration isolation, most of which involve the use of clever mechanical suspensions to cradle payloads softly.  LIGO II has adopted an alternative approach using a technique commonplace in practically all other aspects of gravitational wave detection feedback.  Each table from which mirrors are suspended will be instrumented with a complement of seismometers to measure its vibrations in various directions. The signals from these seismometers will then be fed back to an array of electromagnetic forcers that push on the table and stabilize it while the surrounding lab vibrates.  The table itself must be mounted on springs, but these can be relatively stiff, making the system fairly easy to work with and a robust platform from which to hang the exquisite optics.  The combination of two such tables stacked together will reduce ground vibrations at 10 cycles/second by a factor of a few thousand, fulfilling one of many conditions to allow LIGO II to detect gravitational waves at that frequency.

A group of researchers called the "stiffs" (the moniker derives from the aforementioned stiff mounting springs--any resemblance to corpses, tramps, flops, etc., is purely coincidental) from JILA, Louisiana State University, MIT, and Stanford have been developing this particular isolation strategy for LIGO II.  One major focus of the work to date has been the construction and testing of a two-stage prototype system.  This device went from a sketch on a white board in December 1999 to installation in a vacuum tank at LIGO-MIT in March 2000.  The detailed design and fabrication were performed at High Precision Devices in Boulder, Colorado, in close coordination with the JILA people. The components were shipped to MIT where, in the course of a whirlwind few days, they were assembled on the high bay floor and the completed prototype was hoisted into one of the chambers formerly occupied by part of the Phase Noise Interferometer.

Jamie Rollins with the Stiff Isolation Prototype Jamie Rollins, a graduate student at MIT, is shown in the photo at left guiding the prototype into place. The support structure, enveloping the two isolation tables, is being carried by the yellow lifting straps. The rectangular and disc shapes attached to the sides of the tables are ballast mass, used for simulation of a payload and adjustment of spring positions and level. For reliability and ease of use, the prototype is instrumented with commercial forcers and sensors. The dull green domed cylinder on the corner of the upper stage is one of three Streckeisen STS-2 seismometers, widely used by geophysicists and recognized as being one of the best instruments for measuring low-frequency vibrations.  Just to the left of the STS-2 is a smaller cylinder, a so-called geophone, whose signal is used to detect motions at higher frequencies (greater than about 20 cycles/second).  Not shown in the picture is another important part of the feedback system--the computer that filters the seismometer signals, then feeds them back in appropriate amounts to the appropriate forcers.

Tests on the prototype so far have shown that it is possible to apply feedback to reduce all motions of both tables together for long periods of time.  Significantly, there has been no need for any mechanical adjustments since the time of the initial installation and alignment. All diagnostic tests and tuning of the system can be done by an operator sitting in front of the electronics rack and computer.  The most important goal for the near future of the prototype is demonstrating that it can give significant isolation at frequencies as low as 0.2 cycles/second, where ground motion is largest.  It will also be programmed to have different operating modes, for example a feedback scheme that produces drag forces on the tables as if they were moving through syrup--useful when earthquakes hit or when the payload is being adjusted.

Experience gained with this prototype will be applied to the mechanical and electronic design of the next generations of isolation systems.  Though LIGO II may still feel distant, in fact there is already a challenging schedule of development work in place to achieve the milestones.  Look for full-scale, fully functional isolation systems for both types of LIGO vacuum chambers to be installed in the MIT advanced systems testbed (LASTI) by the second quarter of 2002!