Search

The Giant Magellan Telescope (GMT) is an advanced Extremely Large Telescope and is being designed for broad wavelength coverage and maximum sensitivity in pursuit of several high priority cosmology science objectives. The GMT will use large primary mirror segments to provide well corrected wavefronts and a very fast optical system with a 20-arcminute diameter field of view. It will have a direct Gregorian optical prescription and deformable secondary mirrors to implement adaptive optics. The project timeline leads to initial commissioning with early instruments by 2023. The Systems Engineering group, along with the chief engineer, is developing and flowing down science and operational requirements and is the guardian of the performance budget. An integrated performance model of the telescope is being developed to support requirements derivation, performance assessment and trade studies that drive telescope architecture decisions. This model has four key elements; a structural plant model which exhibits the complex dynamics of the telescope; a controls model that models both controller dynamics and noise; an optics model to accurately demonstrate performance and optical feedback; and loads models to characterize various environmental loads. This paper will introduce the GMT integrated modeling architecture, but will focus on the specific tools and process developed to build the critical structural model. An efficient process was required to support the variety of required analyses, update frequency of underlying finite element models, and various project critical trade studies. The end-to-end process, from subsystem finite element models to a system level plant and Simulink block diagram has been achieved through development of a new MATLAB class with many useful functions. Functions are covered that combine the subsystem models, manipulate those models, and help them run efficiently in discrete and continuous time. Verification and version control are also discussed. Case study of wind is presented with preliminary performance results.

The Thirty Meter Telescope (TMT) needs to maintain precise positioning of the optical elements to deliver unprecedented image quality. Any vibration from observatory sources must therefore be minimized; model-based analysis leads to maximum allowable forces from any individual source typically of order a Newton or less in the most sensitive frequency band. Careful attention throughout the design process is needed to ensure that these challenging requirements are satisfied. We focus here in particular on cogging forces from the azimuth motor drive. Each motor yields periodic tangential and vertical forces of order 100's of Newtons, with higher harmonics of the waveform potentially exciting telescope structural resonances. There are 56 drive motors, and appropriate phasing between them can ideally cancel most of the net torque or net vertical force. However, the moment created between non-collocated forces still produces image motion even if the net force cancels, and further, small errors in forcer positioning result in imperfect cancellation. We provide a general methodology for estimating total cogging forces and then present an example of the expected TMT performance impact from motor cogging placed in context with the larger challenge of demanding vibration requirement.

Program CC is a computer aided control system design package which operates on IBM-PC's and compatible personal computers. It is command driven and works with a wide variety of classical, sampled-data, modern, and optimal control systems. Systems Technology, Inc. is releasing a new version, Version 4, and this note highlights new features which are of interest to industrial engineers and academic researchers.