No description available.

No description available.

The problem of using a computer to control the figuring of an optical surface is investigated. By assuming a linear, shift invariant figuring process, the amount of material removed during each figuring run can be computed. This is done by convolving a tool removal profile with a dwell function that describes the amount of time the figuring tool spends in each area element on the surface. Four methods of computing a dwell function that will best remove the figure errors are described. The advantages of making surface figure measurements using direct wavefront measurement techniques over the interferogram analysis methods used in previous computer controlled figuring machines are also discussed. The design and construction of a computer controlled optical figuring machine is then reviewed. The machine uses a computer controlled heterodyne interferometer to provide optical testing data on the surface being polished. Two microcomputers are used to analyze the test data and run the machine. Optical figuring is performed by scanning a polishing head with a known removal function over the surface at a rate derived from the surface errors. The operation of the software that computes the run path data and controls the machine hardware is outlined. The performance of each of the machine components is evaluated by comparing the behavior predicted by theory to the measured behavior. Initially, the accuracy of the interferometer is measured. The interferometer is then used to determine the performance of the polishing head by measuring the tool removal function. It is then shown that the machine can be run so that a predictable amount of material is removed from the surface. Finally, the feedback loop is closed and surface figure data from the interferometer is used to correctly polish the central region of a 16 inch diameter mirror. It is shown that the surface figure can be predicted with good accuracy over the entire surface. This work concludes with recommendations for improving the machine hardware and for improving the figuring performance near the edge of the surface being polished.

No description available.

The subject of this thesis is lens design using a system of functional differential equations arising from Fermat's Principle in geometrical optics. The emphasis is primarily on existence, uniqueness, and analyticity, properties of solutions to these equations, but some asymptotic methods are developed for special cases. Three specific lens problems are considered in detail: the first is an axial lens having two pairs of foci on the optical axis, the second is an axial lens which focuses light at two different frequencies to two distinct points, the third is a lens symmetric about an axis having foci not on said axis.

Contact lens production requires high accuracy and good surface integrity. Surface roughness is generally used to measure the index quality of a turning process. It has been an important response because it has direct influence toward the part performance and the production cost. Hence, choosing optimal cutting parameters will not only improve the quality measure but also the productivity. In this study, an ONSI-56 (Onsifocon A) contact lens buttons were used to investigate the triboelectric phenomena and the effects of turning parameters on surface finish of the lens materials. ONSI-56 specimens are machined by Precitech Nanoform Ultra-grind 250 precision machine and the roughness values of the diamond turned surfaces are measured by Taylor Hopson PGI Profilometer. Electrostatics values were measured using electrostatic voltmeter. An artificial neural network (ANN) and response surface (RS) model were developed to predict surface roughness and electrostatic discharge (ESD) on the turned ONSI-56. In the development of predictive models, turning parameters of cutting speed, feed rate and depth of cut were considered as model variables. The required data for predictive models were obtained by conducting a series of turning test and measuring the surface roughness and ESD data. Good agreement is observed between the predictive models results and the experimental measurements. The ANN and RSM models for ONSI-56 are compared with each other using mean absolute percentage error (MAPE) for accuracy and computational cost.