Segment Alignment Maintenance System (SAMS) for the Hobby-Eberly Telescope

McDonald Observatory’s Hobby-Eberly Telescope (HET) saw first light in December 1996 with seven segments. By summer 1998, the HET was fully populated with 91 spherical mirror segments. The HET’s 91 primary mirror segments (1 meter flat-to-flat hexagons) are aligned upon commencing an evening’s observation. Originally, the intention was that the structural stability of the telescope support truss, mirror segment supports and mirrors would ensure that no further alignments or figure maintenance would be required throughout the night. However, thermoelastic effects likely coupled with nonlinearities in truss joints and mirror support interfaces have caused unacceptable degradation of telescope performance throughout the night. To remedy the situation, McDonald Observatory decided to implement a Segment Alignment Maintenance System (SAMS) utilizing segment edge sensors.

McDonald Observatory approached MSFC because, other than the Keck telescopes, the PAMELA is the only other segmented primary mirror telescope in the world that has successfully used edge sensors to maintain the primary mirror figure. MSFC and McDonald Observatory signed a mutual Memorandum of Understanding in early 1999 and developed a good working relationship in defining the SAMS concept. MSFC personnel visited the HET and conducted some structural dynamics testing on the HET truss. The testing confirmed loose joints in the truss. MSFC performed simulations of a SAMS control system concept in order to derive SAMS requirements. MSFC also assisted in developing a draft SAMS specification.

In September 1999, the Marshall Space Flight Center, teaming with Blue Line Engineering, was awarded a contract to design, develop and install a Segment Alignment Maintenance System (SAMS) for the HET. The SAMS utilizes inductive edge sensors and high-fidelity electrolytic tilt sensors to sense mirror deviations from the initial aligned figure. The sensing system can observe all system degrees of freedom, including global radius of curvature, to acceptable accuracy. Optimal control algorithms use the sensor outputs to generate segment actuator commands, which minimize the global edge misalignment and deviation from global radius of curvature.