1. The part that was never designed for this

Andy Weir’s The Martian returns as the recalled work because Mark Watney survives by repurposing equipment that was never meant for the exact situation it faces. A LEO desktop will do the same: many of its most capable components will be commercial parts that were not designed for space, and the design must compensate for that fact.

This entry reads about the three categories of parts available to spacecraft designers.

2. Radiation-hardened

A radiation-hardened component is designed and processed specifically to withstand radiation. The manufacturer specifies TID and SEE performance as part of the datasheet, and the design uses techniques like:

  • enclosed-layout transistors to reduce charge collection;
  • guard rings to suppress latchup paths;
  • insulating substrates such as silicon-on-sapphire or silicon-on-insulator;
  • wider feature sizes that are inherently less sensitive to single events.

Rad-hard parts are expensive, have long lead times, and lag commercial performance by years or decades. They are the default for long-life, high-value missions where failure is not an option.

3. Radiation-tolerant

A radiation-tolerant component is typically a commercial or automotive design that has been screened, tested, and sometimes repackaged to meet a known radiation level. As HASCO’s COTS in space article explains, radiation tolerance is established by testing, not by redesigning the device from the transistor level.

Microchip’s COTS-to-radiation-tolerant and radiation-hardened Arm MCUs illustrate the scaling approach: start development on a cheap COTS SAMV71, then move to the radiation-tolerant SAMV71Q21RT or the radiation-hardened SAMRH71 without rewriting software. This reduces schedule risk and lets teams prove algorithms before buying space-grade parts.

4. Commercial-off-the-shelf

COTS components are the cheapest, fastest, and most capable option. They are also the least characterized for radiation. A Tech Briefs article on LEO constellations describes the modern shift toward hybrid architectures: COTS for performance and cost, with rad-tolerant or rad-hard parts only where the mission analysis says they are necessary.

COTS in LEO is not blind optimism. It works when:

  • the orbit is low enough that total dose accumulates slowly;
  • the mission life is short enough that cumulative damage does not dominate;
  • SEEs are detected and corrected at the system level;
  • destructive events are prevented by fast current limiting and power cycling;
  • parts are lot-tested or screened to catch outliers.

An MDPI assessment of COTS in the true LEO radiation environment emphasizes that orbit, lifetime, and solar activity together determine whether a COTS part is reasonable.

5. The selection logic

The choice is not rad-hard or nothing. It is a spectrum:

  • Use rad-hard parts where destructive failure or cumulative dose would end the mission.
  • Use rad-tolerant parts where performance and heritage matter but cost and lead time must be controlled.
  • Use COTS where system-level mitigation can handle the residual risk, or where obsolescence and performance dominate.

6. What this changes

  • The desktop can use a hybrid strategy rather than defaulting to rad-hard everything.
  • The decision should be driven by orbit, lifetime, consequence of failure, and the availability of system-level mitigation.
  • COTS FPGAs and processors are viable if paired with error correction, redundancy, and scrubbing.
  • The next entry will read about those system-level mitigation techniques.