Feasibility of focusing optics for the use of RELIEF in engine test facilities

Trial lens systems of focusing optics were designed and analyzed for their effectiveness in focusing and co-locating two tagging laser beams to be used in the technique of Raman Excitation and Laser Induced Electronic Fluorescence (RELIEF). The RELIEF technique is being developed to provide a measurement of local flow velocities of engine-inlet flows in engine test facilities. The engine test cells at Arnold Engineering Development Center (AEDC) served as a model around which to design the systems. The two laser beams, each containing both wavelengths of light required for Raman excitation, are equally displaced about the optic axis. The lens systems were designed to bring these two off-axis beams to a common focus at the points along a radial line extending from the center of a given engine-inlet air supply duct to near the inner surface of a cylindrical duct access window, the curvature of which matched that of the interior of the given duct. The optic axis of the trial lens systems is oriented perpendicular to, and intersects, the longitudinal axis of the duct. The two beams are contained in the plane defined by the two axes and, thus, cross within this plane. The trial lens system designs were obtained using the Zemax^TM optical design program. This program provides a platform to simulate the representative duct sizes in the engine test facilities at AEDC and to design scaled optics for each. Though Zemax^TM offers the capability to optimize and analyze lens systems for use by Gaussian beams, it is limited to a rotationally symmetric case. The off-axis nature of the two laser beams and the cylindrical window used for optical access into the engine-inlet duct clearly violate this restriction. Thus, the systems in Zemax^TM were designed, optimized and toleranced strictly based on ray approximations of the off-axis beams. To analyze the Gaussian beam focusing properties of the trial lens systems, it was necessary to develop a FORTRAN code as part of this thesis. The FORTRAN code uses a basic ray-tracing algorithm for the chief ray of each laser beam combined with a method for determining the ABCD matrices encountered by that off-axis beam. These ABCD matrix values are then used to transform the input beam using the ABCD transformation law. A limited validation of the FORTRAN code was accomplished by comparing output beam spot sizes for rotationally symmetric systems to those predicted by Zemax^TM. This validation showed a maximum difference of 5% between the FORTRAN code and Zemax^TM. Variants of the trial lens system designs that modeled manufactured optics were simulated using a Monte Carlo technique in Zemax^TM. These models, along with the optimized designs for each of the trial lens systems, were analyzed using the FORTRAN code. The results showed that the designs were capable of focusing the two off-axis laser beams to the spot radii necessary to achieve RELIEF tagging (50 μm). Thus, the feasibility of these optics was quantitatively demonstrated for the optimized systems and for the models of manufactured systems.