Threats to Resiliency of Redundant Systems Due to Destructive SEE

Destructive single-event effects (SEEs) pose serious challenges for the reliable use of commercial off-the-shelf (COTS) devices in space systems. We modeled a simplified, redundant system to determine single-event latchup (SEL) rates that would likely compromise system reliability, resilience, and capabilities. We then assembled a representative data set of COTS Complementary Metal Oxide Semiconductor (CMOS) parts susceptible to SELs. After investigations of these parts revealed no compact distributions or other relationships useful for $a$ priori estimation of SEL rates and showed minimal correlation between Weibull fit parameters for the cross section versus linear energy transfer (LET) data, we used nonparametric statistical techniques to generate quantile plots that can be used to bound part SEL rates for a given confidence. We used these distributions to assess the threat posed to redundant systems by destructive SEEs.