Control challenges for extremely large telescopes

Abstract

The next generation of large ground-based optical telescopes are likely to involve a highly segmented primary mirror that must be controlled in the presence of wind and other disturbances, resulting in a new set of challenges for control. The current design concept for the California Extremely Large Telescope (CELT) includes 1080 segments in the primary mirror, with the out-of-plane degrees of freedom actively controlled. In addition to the 3240 primary mirror actuators, the secondary mirror of the telescope will also require at least 5 degree of freedom control. The bandwidth of both control systems will be limited by coupling to structural modes. I discuss three control issues for extremely large telescopes in the context of the CELT design, describing both the status and remaining challenges. First, with many actuators and sensors, the cost and reliability of the control hardware is critical; the hardware requirements and current actuator design are discussed. Second, wind buffeting due to turbulence inside the telescope enclosure is likely to drive the control bandwidth higher, and hence limitations resulting from control-structure-interaction must be understood. Finally, the impact on the control architecture is briefly discussed.

Wind loads on ground-based telescopes

“…Measurements taken at the Gemini observatory with the vents closed and the telescope facing upwind showed, on average, that the effective velocity u M1,eff near M1 was 7% of the external wind speed 19 (Fig. 7), and that this was a good predictor of both the rms pressure ͑C p ϳ 1͒ on the primary mirror and the outer-scale frequency.…”

“…While the turbulence inside the telescope enclosure is clearly not isotropic, observations at Gemini 5,19 (e.g., Figs. 9 and 10) and other telescopes (Keck, VLT, MMT) as well as measurements in wind-tunnel testing (see Fig.…”

Parametric modeling and control of telescope wind-induced vibration

“…• a 1 (broadband on M1 due to dome opening): Measurements taken at Gemini observatory with the vents closed and the telescope facing upwind gave that on average the rms velocity at the primary mirror was 6% of the external wind speed, 5 and that this was a good predictor of both the rms pressure on the primary mirror and the outer scale frequency f 1 = D s /U 1s . The internal wind speeds are assumed to scale with the ratio of dome opening size to dome diameter,…”

“…The outer scale L s associated with the turbulence observed near the secondary mirror, and the turbulence near the primary mirror when the vents are closed is roughly equal to the diameter D s of the dome opening. 5,10 The outer scale L v associated with the turbulence over M1 that is caused by flow through open vents is related to the height D v of the vent gates. 10 The spatial correlation length of pressure on a surface resulting from turbulence is related to the outer scale by = L/ √ 70.8.…”

“…Several studies have analyzed the wind-induced buffeting of the telescope structure that results from turbulence inside the telescope enclosure [3][4][5] and concluded that this vibration must be taken into account in the design process, affecting the design of the telescope enclosure, the telescope structure and the active control system. Estimating the degradation in image quality that results from the wind-induced vibration requires understanding the wind inside the enclosure, coupling these forces to a structural model, estimating the achievable reduction in vibration using active control, and computing the optical consequences of the residual vibration.…”