We describe a process for increasing power efficiency of concentrator photovoltaic systems by optimizing the lens-to-cell spacing. We find that there is an optimum defocus position with improved power output and reduced sensitivity to pointing errors, which in combination can result in a more than 10% enhancement. The improvement can be realized by minor changes to module cases which should not require changes to other manufacturing, installation, or component costs. In fact optimizing the defocus position allows for lower costs per unit power due to increased power and relaxed system tolerances. The paper focuses on detailed data illuminating the behavior of ultra high concentration photovoltaic modules. While one can look forward to optimizing defocus through sufficiently detailed simulation, at present, we find that an empirical determination of optimum defocus is necessary. The data reveals that even without design parameters changing, supply chain changes can have a significant impact on the optimum defocus - data from five different module configurations with components from different manufacturing lots are presented. These different configurations serve to illustrate the consequences of component changes and the importance of verifying the optimum defocus. A discussion of the effects that are important to determining the optimum defocus and which underlie these differences is included.