Our goal was to determine the feasability of replacing a refractive collimation lens with a diffractive collimation HOE. Two aspects of the design were especially critical to feasability. The first was dispersion and compensation for dispersion, the second was blocking the zero order in an on axis configuration. Other details to be studied were aberrations and the means by which they could be controlled and corrected. The work that was performed included fabrication of several prototype single and sandwich HOE devices in sizes from 3.8 cm to 20 cm diameter. Angular bandwidth and effeciency both had to be as high as possible while meeting recording conditions that would minimize both aberrations and Bragg plane distortions. Results of our fabrication efforts were in close agreement with prior work except that we did acheive higher efficiencies at broader angular bandwidths than were previously shown. The sandwich efficiencies were high enough to effectively block out the zero order light in an on axis configuration. No light control film or air space was needed to acheive good viewability. The prototype shows bad blurring for 50 nm bandwidths but almost none for 15 nm bandwidths. The various prototypes each had some imperfections in design or fabrication but they demonstrate that efficiencies and field of view are acheivable to make a useful HUD collimator. The horizontal field of view can easily exceed 25 degrees with a near 10 degree vertical field and efficiencies approached 98% with some areas blocking 99% of the zero order light in both polarizations.