Radiation Design Margin (RDM) for TID is one of those topics where everyone has a number, but the justification for that number varies widely. The traditional approach — “test to 2× the requirement” or “test to 100 krad if the requirement is 50 krad” — is deeply embedded in space program culture. But is it technically justified, and does one size fit all?
Factors that should influence RDM:
-
Environment uncertainty. How well do you know your actual dose? Trapped proton models have known uncertainties. Solar particle contributions are probabilistic. Shielding analysis introduces additional uncertainty through geometry simplifications and material property assumptions.
-
Part-to-part variability. You tested 5 parts from one lot. What about the other 500 in your flight build? Lot-to-lot variability for some technologies can be a factor of 2-3× in TID response.
-
Dose rate effects. If your TID requirement is based on total mission dose but your testing was at high dose rate, ELDRS-susceptible parts may fail at lower cumulative doses than your test results suggest.
-
Annealing. Co-60 testing at high dose rates may include significant room-temperature annealing during irradiation steps. Depending on mission duration and dose rate profile, the in-flight annealing behavior may be different.
What do different standards say?
- ECSS-Q-ST-60-15C recommends RDM ≥ 2 for components tested at lot level.
- Many NASA programs use RDM = 2 as a starting point, with justification for lower values case-by-case.
- Some commercial constellation programs accept RDM as low as 1.2-1.5 for parts with extensive heritage data.
What RDM philosophy does your program use? Is it a blanket factor or does it vary by part criticality, failure mode, and data quality? How do you justify your margin to a review board?