RHA Methodology Helps Improve Mission Duration Timelines

RHA Methodology Helps Improve Mission Duration Timelines

3-minute read

A new Radiation Hardness Assurance (RHA) methodology could lead to shorter, but more frequent missions at NASA. Mike Xapsos, a research physicist in the Radiation Effects and Analysis Group and member of the NASA Electronic Parts and Packaging Program (NEPP) at NASA Goddard Space Flight Center, has developed an improved method for evaluating the performance of a microelectronic device in space, which closes the gap between design specifications and on-orbit performance.

Xapsos looked at NASA’s approach to radiation environment evaluation and developed a methodology that should allow engineers and designers to have better views of risk assumption based on ionizing dose damage levels for a mission. The new method is a confidence-based approach; it uses current radiation environment models more consistently and replaces the concept of radiation design margin with failure probability.

Historically, engineers would take a mission parameter space (i.e., when the project is flying, where it’s going and for how long) and determine the expected radiation environment the spacecraft is likely to encounter during the mission. Then they would assign design margins on those numbers depending on the mission risk profile, covering both part-to-part variants in the electronic parts being used as well as variability of the environment. Engineers tended to be very conservative with their estimates, putting arbitrarily high margins on the specifications resulting in electronic designs that would often last significantly longer than the mission requirement.

Although the longer lifespan can lead to additional scientific findings, it also can take funding, personnel and time away from potential new missions. This new approach allows engineers to quantify mission requirements to be more consistent with the required timeline outlined for the mission; a three year mission would last five years maximum rather than 20, for example.

“The problem with a 20-year lifespan is the ongoing operational costs. The science team is still working, the operations team is still working, so you have an extra 15 years of costs,” said Ken LaBel, co-manager of NEPP. “The modern idea is that we can refresh using smaller missions on a more regular basis each with improved instrumentation and means of measuring, making for improved science collection.”

In his new methodology, Xapsos uses a probabilistic calculation involving confidence levels. The confidence levels are how engineers view both the prediction of the environment and the radiation tolerance of electronics used in the project.

Using the new methodology, engineers can evaluate a radiation environment as a function of the confidence level and test the device for total dose radiation in a laboratory. If they test 10 devices, they might receive 10 different failure levels. The devices themselves have a probability distribution of failure (i.e. mean failure levels with a distribution tail), and with this new approach, they can combine the device failure level with the probability of confidence level for the environment and calculate the failure probability of this device on a mission.

“If you use the more recent models of the radiation environment, you can predict that the radiation levels will not exceed a given level at 90 percent confidence or won’t exceed a higher level at 99 percent confidence,” said Xapsos. “By doing these probabilistic calculations, it offers new capabilities and flexibilities for the design engineers.”

Using this method, engineers will be able to make more trades to optimize design. For example, engineers can trade off the failure probability of a device for more speed or use a faster device if they are willing to accept a higher failure probability due to total dose. Or, if they want to decrease failure probability, they may need to use a slower device.

The next phase of Xapsos’ work is to take these types of confidence levels and turn them into Reliability-type numbers. Members of NEPP are building a toolset around this approach to define requirements and goals that help validation processes within the Model-Based Mission Assurance framework.

“If we do our job right in NEPP, mission and project managers won’t know we exist because we’re putting infrastructure in place that makes their lives easier,” said LaBel.

NEPP co-sponsored a research paper on this methodology that won the Best Paper Award as well as the Best Radiation Effects Data Workshop Presentation Award at the 2016 IEEE [Institute of Electrical and Electronics Engineers] Nuclear and Space Radiation Effects Conference, a premiere meeting for radiation effects on electronics.