1997-2002 Solar Array String Failures Revisited

In 1997, two commercial geostationary satellites experienced a new phenomenon: sustained solar array arcing. Although arcing on solar arrays in space had been expected from ground tests and space flight experiments, it was heretofore unknown that arcs into the space plasma could turn into arcs between adjacent solar array strings at high interstring voltages. Experiments validated the concept, and design changes were made to succeeding satellites that have prevented similar occurrences. Here, the dates and times of 32 sustained arcing string failures from 1997 to 2002 are reported, and NASCAP-2k charging simulations are used to try to determine thresholds necessary for the arcs to occur. Surprisingly, in some cases, conditions for sustained arcs last for a longer time when the environment is relatively benign. A plausible scenario for primary arcs in sustained arcing conditions is presented that is based on charging time histories from NASCAP-2K simulations. The charging scenario involves solar array voltage turn-on after eclipse or upon unshunting the array, and it hypothesizes lower than usual arc voltage thresholds immediately after differential string voltages are activated. Laboratory testing under both geosynchronous-Earth-orbit and low-Earth-orbit conditions is cited to show plausibility of this argument. Improved design and testing rules for spacecraft to avoid sustained arcing with a minimum of overdesign are presented.