A series of experiments designed to study reacting nitrogen flow over double-wedge geometries was conducted in the T5 shock tunnel at the California Institute of Technology. These experiments were designed using computational fluid dynamics to test nonequilibrium chemistry models. Surface heat transfer rate measurements were made, and holographic Mach-Zehnder interferometry was used to visualize the flow. Analysis of the data shows that computations using standard thermochemical models cannot reproduce the experimental results. The computed separation zones are smaller than the experiments indicate. However, the computed heat transfer values match the experimental data in the separation zone, and on the second wedge the computed heat transfer distribution matches the shape and heights of the experimental distribution but is shifted due to the difference in the size of the separation zones. The most likely reasons for failure of the computations to reproduce the experimental data are uncertainties in the equilibrium and nonequilibrium nitrogen dissociation rates, non-Boltzmann vibrational energy distributions in the freestream, and possible noncontinuum effects at the model leading edge and in the shock interaction region.