Single-Axis and Multi-Mass Drag-Free SatellitesMore Complex Drag-Free DesignsThe definition given on the Home page is for a full three-axis Drag-Free Satellite where the proof mass floats freely in the outer satellite. This is the simplest and most reliable implementation, but there are other Drag-Free Satellite configurations:
Single-Axis Drag-Free SatellitesIt is possible to construct a Drag-Free Satellite which is only drag free in one axis. For example if the satellite attitude control holds it stable with respect to a locally-level reference frame as would be the case with Gravity-Gradient stabilization, the primary disturbance to be compensated for (air drag) mostly lies along only one axis of the satellite. In this case much of the disturbance can be eliminated by only making the axis parallel to the satellite velocity vector drag free. The first single-axis Drag-Free satellite was proposed, analysed, and built by Bill Davis in 1966 at the old Lockheed Missiles and Space Company in Sunnyvale. Unfortunately, there was a launch failure; and it never flew. There were, however, 5 single-axes Drag-Free satellites which actually flew on the TRANSIT Improvement Program (TIP). Single axis is more complex than three axis, and it took three flights (not counting Lockheed's) before it finally worked correctly. The Two-Mass Free-Falling Drag-Free SatelliteAn Equivalence-Principle (EP) test must compare two masses. Since only one mass can truly be drag free, the question arises how two drag-free masses can be used in a free-floating, free-falling experiment. The answer is straightforward. If one mass is a sphere and one is a spherical shell surrounding the first, the inner sphere can be the drag-free proof mass; and the outer shell can be constrained by an electric support until the experiment begins when it would be set free to co-fall with the sphere. Both would be inside of the cavity of the outer satellite, so the system is a nested set of masses: a sphere surrounded by the spherical shell which is in turn surrounded by the outer satellite. The shell must, of course, be transparent so that the position of the inner sphere can be measured optically. In a typical experiment the centers of the sphere and the shell never separate more than 10-8 meters, so the two never collide. This is the configuration used in the DC Cancellation EP Experiment (Free-Fall STEP). This the best design for an EP experiment since both proof masses are monopoles, a perfect shell has no interaction with the central sphere, they fall almost together, and with DC Cancellation an EP violation causes them to separate as t2 not t and there is almost no confusion with semimajor-axis errors.Drag-Free Satellites Which Change the Orbit Dynamics of the SatelliteIt is possible to use a Drag-Free Satellite to change the orbit dynamics of the main satellite. The orbit dynamics of any satellite depend on the gravity-gradient field in which the satellite finds itself. By using local masses in the satellite, the gravity-gradient field seen by the proof mass can be changed, and thus its dynamics will be altered. Since the outer satellite follows the proof mass, its dynamics are the same as the altered dynamics of the proof mass. This is very useful in the case of an Equivalence-Principle experiment. With normal satellite dynamics a difference in the semimajor axes of the two proof masses looks exactly like an EP violation. The tolerance on the semimajor-axis error is so small that a high-accuracy EP test is essentially impossible without some kind of counter measure. In addition, an EP violation causes the two proof masses to only separate as t not as t2. By adding two large masses fixed in the satellite on either side of the two proof masses, the dynamics of the proof masses can be changed to solve both of these problems. See Two Major Problems With a Satellite EP Test or the DC Cancellation paper.The Minimum-Force Drag-Free SatelliteThere may be reasons not to have the proof mass free floating in the outer satellite cavity; but rather to keep it centered with a support system, typically electric. In this case the outer satellite propulsion control system can be designed to minimize the support forces on the proof mass. In the case of a spherical proof mass, minimizing the forces requires that only three degrees-of-freedom be controlled, and the propulsion control system design is fairly simple. For other proof-mass shapes, however, the proof-mass attitude with respect to the satellite cavity must be constrained, and it becomes necessary to control a 6 degree-of-freedom system. This requires that six forces (or equivalently 3 forces and 3 torques) be applied to the proof mass. The Multi-Mass Single-Axis and/or Minimum-Force Drag-Free SatelliteSome experiments propose multi-proof-mass systems where the proof masses are nonspherical. Two important examples are two single-axis cylinders sliding inside of one another (STEP) or multi-mass proof masses constrained in attitude (LISA). In these cases, however, it may still be desired to considerably reduce the disturbances acting on the proof masses. STEP, for example, uses two cylinders supported by superconducting magnetic bearings sliding inside of one another along a single rotating axis. By designing the Drag-Free control system to be drag free along the differential axis and to minimize the constrain forces along the perpendicular axis, the experiment disturbances can be considerably reduced. One design of LISA uses to cubical proof masses as mirrors for the laser beams (one for each of the other satellites in the triangle). In the case of two single-axis proof masses in one Drag-Free satellite, the two axes must be oriented 90 degrees with respect to each other for each axis to be truly drag free. |
|
Copyright (c) 2001, Benjamin Lange, All rights reserved.
Benjamin Lange 1922 Page Street San Francisco, CA 94117 415-221-6600, Extension 310 email: blange(at sign)virtualpbx.com |