There are many acceptable ways to construct an imaging Polarimeter, each with its own benefits and drawbacks. The most common systems involve rotating elements, but use of these systems puts limitations on the dynamic nature of the scene. Division of Aperture (DoAP) and Division of Amplitude Polarimeters (DoAmP) solve the temporal synchronization issue by using multiple light paths, each of which has its own set of polarization optics. These systems can provide real-time imagery, but there are significant challenges surrounding optomechanical alignment and sensitivity to vibrations. Division of Focal Plane devices (DoFP) use an integrated array of micropolarizers to solve the temporal and mechanical alignment issues, but suffer from exactly 1 pixel of IFOV error that cannot be compensated for at the full resolution of the system. Recently we presented a concept that creates a highly parallel array of non-imaging DoAP devices. The design uses two microlens arrays to relay the image at an intermediate focal plane through a microgrid polarizer. The microlenses are configured such that each lens in the first array feeds four lenses in the second array, so as to create a non-imaging DoAP Polarimeter. Our previous work was only a conceptual design. In this paper, we will present a design and ray-tracing analysis of a proposed system. We quantify the principal drawbacks of vignetting and crosstalk, and give expected performance parameters of a final device.