Optomechanical tolerancing of drop-in optical systems

Optical systems are commonly toleranced using a statistical Monte Carlo analysis that includes three main operations: the perturbation of the nominal system parameters through the application of the manufacturing and assembly tolerances, the optimization of the compensator positions, and the evaluation of the system performance. These perturbation, optimization, and evaluation operations are usually performed using the built-in tools that commercial optical software packages provide to streamline the tolerancing process. However, these tools are not well-suited to faithfully model the complex interactions between optical and mechanical components that limit the alignment tolerances in drop-in optical systems. For this reason, we propose a novel approach that divides the tolerancing process into a perturbation stage that is performed in Matlab and an optimization and evaluation stage that is performed in optical software. Implementing the perturbation stage in Matlab allowed us to model the mechanical components (barrel, spacers, etc.) and the way in which these components constrain the positions and orientations of each lens in a drop-in optical system. Using Matlab we first calculated the static tolerance band, which we defined as the perturbation range (decenter, tilt, roll) of each lens when all the previous components are assumed to be perfect. Following this step, we calculated the dynamic tolerance band, which we defined as the build-specific perturbation range that accounts for the stack up of tolerances prior to each lens. We then applied random perturbations within these bands to each lens, optimized the compensator positions, and evaluated the optical performance. Using this process, we calculated the sensitivity and the expected MTF yield of a Ruda Optical drop-in lens under static, dynamic, and floating glass tolerance scenarios. The outputs of these methods were then compared to assess the importance of considering the mechanical components and the stacking order of operations in the assembly process to more accurately model the expected yield when tolerancing drop-in optical systems.