Rubin Observatory primary tertiary mirror cell assembly: mirror support system modeling and tuning

The Vera C. Rubin Observatory1i is now under construction, on Cerro Pachón, in Chile. Its telescope has an 8.4 m primary tertiary mirror (M1M3), and has been designed to conduct a 10 years survey in which it will map the entire night sky every three nights. The support systems is composed by six Hardpoints and 156 pneumatic actuators. The Hardpoints define the mirror’s position in the mirror cell, and hold that position while observing in order to maintain the alignment of the telescope. The pneumatic actuators provides mirror support and figure correction, from optimizations provided by the Active Optics System (AOS). In order to complete the sky mapping in three nights the telescope mount has been designed to develop high accelerations that will allow it to change the observing field in just 2 seconds. The rapid telescope slews will apply high inertial forces on the mirror that have to be offloaded from the Hardpoints by the mirror's pneumatic actuators. This keeps the mirror safe during operations. For this reason, the support system has been optimized introducing innovative control concepts in both the Outer and Inner Loop, and the individual component. Extensive modeling of the control system dynamics has been developed to simulate and test different control architectures. Inner loop control systems were developed and tested using hardware-in-the-loop simulations which take advantage of the power of the modeling, using the real hardware components. This technique is especially suitable for complex real-time systems and allows for faster development time, with more precise results. The Outer Loop model has been used to tune the Hardpoints off-loading control system. The resulting parameters were used directly in the real system, once integration was completed, and worked as the simulations predicted. Gravity lookup- tables and inertial compensation forces, to counteract the high acceleration from the telescope mount, have also been included in the model. Simulations show the Hardpoints forces are far below their limits. This is particularly important, as we will not have the possibility to test the inertial compensation until the mirror cell is actually installed on the telescope mount. The simulations have provided encouraging results.