Absolute calibration of three reference flats based on an iterative algorithm: study and implementation

摘要

LMJ and LIL are two French high power lasers dedicated to fusion and plasma experiments. These laser beams involve hundreds of large optical components, the clear aperture of the beams being 400x400 mm2. In order to control the flatness requirements of its optics, the CEA has an 800 mm diameter Fizeau interferometer. Determining if optical components fulfil the very strict wavefront specifications can be difficult because these specifications can be equivalent to the defects of the reference flats of the interferometer. That is why we want to calibrate our reference flats in order to subtract their defects from the performed measurements. This absolute calibration is based on an iterative algorithm requiring three reference flats. In addition to the three basic combinations of the three flats, this method uses rotations and translations of one flat with respect to the others. First, we shall present a characterization of this method. The choice of different parameters, as the operations of translations and rotations required, will be discussed. Moreover, experimental errors have been introduced in the simulations and their limit values have been studied with regard to the other parameters. An application of this method on our three reference flats has been implemented over a 600 mm diameter. An absolute calibration requires a very precise implementation of the measurements and then we will see why data processing is necessary to reduce the residual shifts in translation but also in rotation and in zoom between the different measurements. Lastly, first uses of the absolute maps show a factor 5 to 10 improvement on the final accuracy.