Overcoming the radiation challenges of space electronics

By Paul Leys   |   May 2, 2022

Through low-cost launches designed for nanosatellites, the possibility to put technology into space has opened up a slew of applications. Although there are many possibilities, space environments (even in orbits close to Earth) are particularly harsh. To deploy systems into orbit that will be reliable requires a focus on radiation hardening of electronics during the design process.

Radiation represents the biggest reliability challenge for electronic hardware in space and the small satellites in low-Earth orbits (LEOs) can be particularly under threat. The radiation environment close to Earth is divided into two main categories.

Sources of radiation

One important source of radiation comes from energy-charged particles trapped by the Earth’s magnetic field, which form the Van Allen belts. These particles are most problematic for satellites in the higher, slower orbits found in the  upper edge of the LEO region. Higher densities of charged electrons appear close to the poles, with protons concentrated more in the region above the southern Atlantic Ocean. These trapped particles are supplemented by high-energy particles produced by the Sun, and which are most troublesome during periods of high solar activity.

The other prominent source of radiation that affects electronics comes from highly energetic cosmic rays produced by countless stars outside our solar system. These particles are among the most destructive that a system can encounter, with energies reaching as high as 1020eV, meaning that they are not readily deflected by the Earth’s magnetic field.

Impact of radiation on electronics

Electronics can be affected by single occurrences or as the result of an accumulation of damage from lower-energy radiation over time. The former is generally classified as a single-event effect (SEE), generally caused by a solitary high-energy particle passing through a device. Its passage injects additional charge into neighbouring regions. This charge often accumulates in storage cells such as registers and static or dynamic memory cells. If the charge accumulation is sufficient, the outcome may be a flipped bit that if left undetected and uncorrected will go on to cause an error in downstream logic. Many SEEs only have a temporary influence and are known as soft errors. Often, the accumulated charge is quickly dispersed by further logic switching.

If it causes a change in register or memory state, this will lead to a single-event upset (SEU) - and some types of SEU can be serious. A single-event functional interrupt (SEFI), for example, is often associated with a control bit or register unexpectedly changing state. This change can cause the component to reset or lock-up, requiring reset by a watchdog circuit.

The damage caused by just one particle may be serious enough to cause permanent and irreversible damage - referred to as a hard error. Such an error will not only lead to lost or altered data, but the affected device will no longer function properly even after a full reset. One example of such a hard error is a single-event latch-up (SEL). This is caused by the formation of a parasitic bipolar transistor in the wells of a CMOS transistor, creating a temporary low resistance path between power and ground. High currents passing through this path will result in permanent damage. 

Failure can also result from accumulation of damage over a period of time. Total ionising dose (TID) represents the failure of an electronic device based on its susceptibility to this kind of damage. Often the TID damage is due to the passage of multiple charged particles through the device, each of which generates pairs of electrons and holes. If this damage occurs at the interfaces of semiconductors or the gate oxide, leakage current can be increased and shifts in threshold voltage leading to an inability to switch or respond accurately to incoming signals may result.

It is worth noting that device shielding can be used to effectively reduce the accumulation of TID radiation. However, some of the effects of TID can be counter-intuitive which may affect design choices. For example, bipolar electronics can suffer from a phenomenon where very low doses can cause failures even if they often pass tests based on higher TID levels. As a result, bipolar devices may need to be tested under an enhanced low-dose rate sensitivity (ELDRS) regime to ensure they are not susceptible to this issue.

Dealing with atomic oxygen

There are easily neglected aspects of radiation damage that can cause issues with device packages and connectors. In orbits between 200km and 700km above the Earth’s surface, one cause for concern is damage from atomic oxygen. This stems from the absorption of energy from photons in the ultraviolet range.

Material erosion caused by atomic oxygen is particularly problematic for plastics, and contamination from the package has been implicated in effects such as ELDRS. Depending on the expected operational lifetime of the satellite, it may be important to investigate oxygen-resistant coatings for boards and protective films or coverings for exposed sensors.

Issues relating to vacuum and temperature exposure

Aside from radiation, the hard-vacuum conditions of space also cause issues for electrical parts, as tin whiskers form more easily in the vacuum of space. If these form between the pins of a device, they can easily cause short circuits that will stop the device from functioning correctly. Tin whiskers have been implicated in the complete failure of four commercial satellites since 1998. 

Outgassing is another way in which the vacuum of space can affect device operation. Molecules from packaging materials or coatings can be released that may then impair the performance of neighbouring devices by becoming deposited upon them - such as sensors or imaging arrays.     

There is also the issue of thermal cycling. This can accelerate the degradation of electronic components, though the relatively short mission lifetime of most LEO satellite deployments mitigate this issue.

Strategies for addressing the effects of space on electronic hardware

There are a number of strategies that design teams can use to prevent or alleviate the possibility of failures caused by radiation and related problems while in orbit. It is important to understand the risks faced by the overall mission if a particular subsystem fails temporarily or permanently, and the level of mitigation that will be needed to ensure mission success. In the context of a satellite constellation, the choice may be taken to accept that several systems will fail in orbit but can easily be replaced. If the satellite is expected to operate on its own, it will require a design that can survive a wider range of problems.

It should be noted that the effects of radiation are not distributed evenly across the different components present in electronic systems. Devices that store or accumulate charge are often more prone to radiation effects. Registers and memory cells can flip, leading to software errors. CMOS image sensors are also prone to damage from energetic particles. Though much depends on the importance of each subsystem to the mission, focusing attention on memory-intensive electronics and sensors is often important when it comes to radiation hardening.

One method for improving overall reliability is to employ redundancy where possible, particularly in critical parts of the system. Dual- or triple-modular redundancy helps ensure that if one subsystem fails other copies are available to support normal operation. This can work for both soft and hard errors. In a triple-modular design, the likelihood is that only one subsystem will be affected by an SEU in any given clock cycle. Comparison logic will show that the other two systems are in agreement and provide the output for that cycle, before performing a recovery operation on the failed subsystem.

Error checking and correction (ECC), which takes advantage of finer-grained redundancy, is an important technique for ensuring that memory arrays affected by SEEs do not cause SEUs. Many memory controllers support ECC with most able to correct single bit errors. Additional redundancy may be required depending on the nature of the mission and how critical to that mission the target subsystem is.

Redundancy can only go so far in handling radiation and other effects of the space environment though. It is entirely possible that multiple systems will fail, due to an excessive TID or SEUs causing hard errors.

Rad-hard space solutions

Radiation-hardened (rad-hard) devices are designed to be able to survive encounters with highly energetic protons and electrons. Some process technologies offer greater protection against damage caused to the crystal lattice of semiconductors. Silicon germanium (SiGe) bipolar devices, for example, are known to survive an accumulated dose of up to a million rads. In contrast, a CMOS device built on an advanced small node size process can suffer permanent damage with a TID of just 5krad.

Devices supplied as rad-hard will have been tested to high TID and SEE levels and are often supplied in ceramic packages that are more resistant to the harsh environment of space. They are far less prone to effects such as outgassing than the plastic packages employed for commercial-grade products. Rad-hard products are normally tested and approved under Class Y or Class V of the MIL-PRF-38535 standard. However, it is important to understand that this level of protection against environmental effects was designed for comparatively long-lifetime satellites and probes that are launched into high orbit or beyond.

Radiation-tolerant alternatives

Because of the effects of atmospheric drag, the typical lifetime of low-cost LEO deployed nanosatellites is only around two to four years. The shorter lifetime may make it worthwhile to investigate components that do not have full rad-hard accreditation, but which provide a greater level of protection for use in space. A risk analysis based on expected radiation levels in the orbit planned for a satellite will indicate the TID and the probability of encountering highly energetic particles over its lifetime. From that calculation it should be possible to determine if use of radiation-tolerant components will be acceptable. These are devices that, after tests have been conducted, are shown to have sufficiently better resistance to higher TID and single-event impacts than commercial ones, but this is not as high as for rad-hard parts. Such radiation-tolerant parts are often available at lower cost. Because they do not need to go through the stringent Class Y/V process, designers may also find they have a wider range of choices when it comes to component selection.

Similar risk analyses will also help inform component and design choices for issues such as thermal cycling and resistance to the formation of tin whiskers. For example, any sensitive subsystems should be designed to not require components with tin leads or be treated to avoid whiskering in vacuum conditions.

Though innovations such as CubeSats have made it easier than ever to take a satellite project to launch and operation, the harsh environment of space needs careful consideration when it comes to implementation of the electronic systems.  In addition to a portfolio of rad-hard and radiation-tolerant components including FPGAs, microprocessors, SRAM memories, ASICs, and DC/DC converters, support is available to designers to help them with their selection decisions.

Follow Avnet Silica on LinkedIn

About Author

Paul Leys

Paul Leys is the Market Segment Manager for the Aerospace and Commercial Avionics business at Avnet ...

View Bio