RDM can be specified at system level down to subsystem, board or component level, depending upon the local radiation environment specification at different components, and the effects analysis methodology adopted for the equipment.
Requiring the RDM to exceed a minimum value ensures that allowance is made for the uncertainties in the prediction of the radiation environment and damage effects, these arising from:
• Uncertainties in the models and data used to predict the environment;
• The potential for stochastic enhancements over the average environment (such as enhancements of the outer electron radiation belt);
• Systematic and statistical errors in models used to assess the influence of shielding, and determine radiation parameters (e.g. TID, TNID, particle fluence) at components’ locations;
• Uncertainties in the radiation tolerance of components, established by irradiation tests, due to systematic testing errors;
• Uncertainties as a result of relating test data to the actual parts procured, and variability of measured radiation tolerance within the population of parts.
An appropriate selection of the radiation design margin takes into account:
• the criticality of the component, subsystem or system to the success of the mission, imposed through equipment reliability and availability requirements, and
• the type of mission (e.g. scientific, commercial, “low-cost”, an optional mission extension).
Margins are also achieved by application of worst-case analyses. The quantification of the margins achieved is a good engineering practice. However, it is recognized that such a quantification is sometimes difficult or impossible.