A brief history of electronic reliability in space — including today’s risks and how to mitigate them

By Mark Toth, marketing manager, HiRel, Texas Instruments

Most people are aware of some of the most famous milestones in early spaceflight, such as the orbit of Sputnik in 1957, Yuri Gagarin and Alan Shepard’s flights into space in 1961, and Neil Armstrong setting foot on the moon in 1969. But you may not know that the first man-made objects flying into space (as defined by the 100-km Kármán line) go back about 75 years to the first V-2 rockets launched from Germany during World War II. These first suborbital rockets spent only a few minutes in space and used simple analog computers for onboard guidance.

By 1957, when Sputnik first orbited Earth for roughly four months at a height of about 250 km, the electronic content of spacecraft had increased significantly, along with the mission length. The following year (1958), the Explorer 1, 3, and 4 missions carried radiation detectors farther into space and led to the discovery of the Van Allen belts, regions of trapped, highly energetic, charged particles around the earth. Early space radiation research was highly focused on protecting the intrepid astronauts (and cosmonauts), who ventured into this new frontier, from the effects of radiation.

In July 1962, the Telstar 1 satellite launched, which successfully transmitted the first television pictures and telephone calls through space. Where would technology be today without this pioneering spacecraft? But only four months later, Telstar unexpectedly went out of service. The cause was eventually traced back to an on-board transistor failure caused by the total radiation dose from a high-altitude nuclear weapons test that occurred the day before Telstar’s launch. The Telstar engineering team was able to recover operation in January 1963 before a permanent loss of service occurred one month later. Telstar had become the first “victim” of radiation-induced electronic failure. Six other satellites were lost within the next seven months, and the earnest study of radiation effects in electronics became a top priority for engineers and scientists the world over.

For the next 40-plus years, the consequences of losing any spacecraft to electronics failure were generally catastrophic. It goes without saying that manned space missions have always carried the highest priority for safety to protect their priceless passengers. But the costs of building unmanned satellites soared into the hundreds of millions of dollars, even exceeding the billion-dollar level for advanced government satellite programs. Satellites became critical to many facets of everyday life: for television viewing, weather forecasting, navigation, and, of course, military uses. The cost of launching was so high that only a few countries could afford the endeavor.

With such astronomical (pun intended) consequences of failure, government spaceflight organizations and spacecraft manufacturers created requirements and standards to ensure the robustness of electronic components in space. You may hear spacecraft engineers refer to such components as triple-E parts, which is shorthand for electrical, electronic, and electromechanical components. Manufacturers typically implemented specifications by creating device-specific source control drawings (SCDs), which focused on testing integrated circuits (ICs) for radiation performance, operating life, thermal performance, and mechanical resistance to forces endured during launch.

A common mission profile for SCDs was based on 15 years of operation in geosynchronous Earth orbit (GEO), approximately 36,000 km above Earth. That’s quite a change from those first rockets, which flew just above 100 km for only a few minutes!

Mil standards

By 1995, the U.S. government (namely, the Department of Defense) had released the MIL-PRF-38535 standard, which created consistent qualification, testing, and reliability standards for military and space ICs. MIL-PRF-38535 defines requirements for IC manufacturers if they wish to be listed on a qualified manufacturer list (QML), along with requirements for ICs to qualify to specific classes. Classes M, N, and Q are intended for terrestrial applications, while Class V is defined for space applications, requiring hermetically sealed packaging. Class V devices are typically tested for radiation tolerance during development and are not radiation-tested in production.

MIL-PRF-38535 defines an additional quality level — radiation hardness assured (RHA) — that requires the screening of each production lot of an IC to meet a specified radiation performance level (more on radiation performance in a minute). Radiation-tolerant ICs that meet QML Class V requirements are often referred to as QMLV devices, while devices that also meet the RHA specifications are called QMLV-RHA or, simply, RHA. In 2012, MIL-PRF-38535 was revised to add a Class Y, covering non-hermetically sealed, high-pin-count land-grid-array– and column-grid-array–packaged devices.

Since the creation of the QMLV designation, designers of high-value, mission-critical spacecraft have come to rely on QMLV and RHA ICs to ensure mission success. The high level of quality and reliability of these components results in a higher cost, but in the intended application, the cost can be justified due to the level of risk mitigation needed.

Cost factors

In recent years, changes in a number of factors have combined to reduce the cost of access to space. Both scientific and commercial organizations have shown that lower-cost small satellites (smallsats) can also be of value alongside the traditional “big birds” deployed for long-life missions. Smallsats in low Earth orbit (LEO) have provided useful data on a variety of scientific Earth observation missions, with mission lives as short as a week. It is now possible for researchers, entrepreneurs, and students to get a smallsat into LEO at a relatively low cost.

Commercial launch operators also offer “ride-sharing” for smallsats on launches in which the primary mission is for a larger satellite. NASA and its commercial partners deliver smallsats to the International Space Station (ISS) aboard re-supply missions, wherein the satellites are then deployed into space by astronauts living onboard the ISS. Take a look at these examples of how this easier access to space has been utilized by different groups:

• A team of middle-school students from Florida designed, built, tested, and launched a 1-kg CubeSat in December 2018 as part of a NASA educational initiative.
• Commercial companies have launched small LEO satellites to provide high-resolution, frequently updated images of Earth to their customers.
• Telecommunications companies have announced plans to launch “constellations” of LEO smallsats to provide internet and other telecom services to customers.

This new array of satellite missions has given rise to a much wider variation in risk tolerance and cost sensitivity compared to traditional missions.

Engineers designing smallsats for LEO missions with shorter mission durations and lower-cost targets have begun to use commercial-off-the-shelf (COTS) ICs, which receive no special screening from manufacturers to address the hazards of spaceflight. These COTS ICs are encapsulated in plastic packaging, which reduces their size compared to the hermetically sealed ceramic packages of space-grade components, and have electrical performance benefits. However, the use of COTS components can expose a mission to significant risks, and selecting, screening, and testing ICs to mitigate these risks can end up costing many times more than the ICs themselves.

Another term used to refer to plastic-packaged ICs used in space is plastic encapsulated microcircuits (PEMs). PEMs may be COTS or specially screened or tested. There has also been interest in using automotive-qualified, or AEC-Q100 (“Q1” for short), grade parts in space. While these devices receive additional testing and qualification compared to commercial ICs, these tests are focused more on the terrestrial environment than the extreme environment of space.

The use of COTS and Q1 ICs can also require compromises in system design and operational readiness when the radiation hardness and other reliability factors of the components are not well-controlled. Due to the desire for ICs to deliver the necessary quality and reliability for shorter, lower-altitude missions at a lower cost point compared to QMLV devices, a number of IC suppliers have introduced new grades of IC, which fit between COTS and space grade (QMLV) from a cost perspective. These “new space” or “space plastic” ICs use plastic packaging but undergo significant additional testing and qualification compared to COTS and Q1 devices. Fig. 1 compares COTS and Q1 ICs to high-reliability parts.

Fig. 1: Comparison of different reliability grades from Texas Instruments.

Next, I’ll discuss the risks of COTS and Q1 ICs in space and how the enhanced quality and reliability of space plastic ICs can enable success for shorter LEO smallsat missions.

Radiation

The tragic story of Telsat underscored the need to address radiation performance in space-borne electronics. If you don’t have (or plan to get) a Ph.D. in nuclear physics, you can use a couple of analogies to understand radiation effects on electronics. I’ll categorize effects into two primary categories: total dose effects and single-event effects (SEEs). Consider total dose effects analogous to sunburn: The longer a spacecraft is in space and the farther it is from Earth, the higher the total dose received and the resulting sunburn. Engineers quantify radiation effects like this as total ionizing dose (TID), measured in units of kilorads (krad).

SEEs are more like a lightning strike: The probability of an occurrence at any moment is low, but the longer a spacecraft is out in the “thunderstorm” of space, the more likely it is to be struck by a highly energetic particle that can cause failure. The sensitivity of an electronic component to SEEs is measured as a quantity called linear energy transfer, which is the energy transferred from a particle to a substance measured in units of MeV∙cm² /mg.

A particular type of SEE that designers are particularly concerned with is single-event latchup (SEL), which is a high-current condition that can destroy an IC quickly. We can “thank our lucky stars” that Earth’s magnetic field protects our earthbound electronics and delicate bodies from the harsh radiation environment of space.

Traditional space-grade (QMLV and RHA) ICs have been designed, qualified, and tested to meet all specifications to TID levels as high as 1 Megarad (Mrad). For the sake of comparison, an unshielded satellite in GEO could absorb a TID of over 20 Mrad in just one year. In practice, shielding satellites can drastically reduce the dosage, but long-life GEO missions typically look for ICs with ratings of at least 100 krad. In contrast, the same unshielded satellite in LEO would absorb only about 100 krad in a year. While smallsats in LEO cannot typically support heavy shielding, the shorter mission profiles mean that TID requirements of 10–30 krad are often sufficient for these applications.

Beyond defining the necessary radiation tolerance level, there are two primary challenges when trying to use a COTS IC for spaceflight: measuring the device’s radiation tolerance and potential variation in that tolerance. Testing ICs for radiation tolerance can be an expensive and time-consuming process due to the specialized radiation test equipment required. For SEE testing, there are only two cyclotrons in the entire U.S. with the heavy-ion beam capability to test devices for spaceflight cost-effectively.

Sometimes, engineers can find radiation test data from academic studies or third-party testing. Such test data may be less accurate than testing by an IC manufacturer. Plus, researchers and third parties typically don’t have access to hidden test modes and may not fully understand the proper operation, specifications, and testing of the device.

Second, the lot-to-lot variation of the radiation tolerance of an IC can be huge. It has been shown numerous times that the radiation performance of a particular IC can vary greatly from one wafer fabrication lot to another. Today’s high-volume semiconductor wafer fabs are great at controlling electrical parameters but do not control for radiation performance.

For example, National Semiconductor’s space-grade LM108 operational amplifier could produce a lot rated to 100 krad one month and have another lot fail at 30 krad the next month. For this reason, radiation lot acceptance testing (RLAT) has been a fundamental pillar of manufacturing ICs for space since the days of SCDs. COTS ICs, even within a single reel, can come from different wafer lots, even different wafer fab factories, yielding drastic swings in radiation tolerance. Flexibility in wafer supply is a benefit for commercial applications but a risk for space missions.

Space-plastic–grade components will go through at least one-time radiation testing by the manufacturer to ensure the validity of the radiation performance and eliminate ICs unsuitable for spaceflight. The TID specification varies by manufacturer but typically targets a TID of 30 krad and single-event latchup immunity to 40 MeV∙cm² /mg, in line with radiation levels within LEO.

Texas Instruments’ space-enhanced product line goes the extra mile by undergoing RLAT to 20-krad TID in every production lot, in addition to a one-time characterization to 30 krad. In addition, space plastic products are manufactured within a single controlled baseline, which means one wafer fab, one assembly site, and one material set, and greatly reduces sources of variability.

Mechanical and thermal concerns

In the extreme environment of space, electronic systems are subjected to extreme temperatures and associated thermal cycles, which cause mechanical stress on ICs and their packages. In LEO, satellites orbit the earth at least 12 times a day, passing from the extreme heat of direct sunlight to the extreme cold of darkness in space. This can result in thermal gradients of over 3°C (5.5°F) per minute, even inside the relatively temperature-controlled internal volume of a spacecraft.

Components can be subjected to temperatures as low as –55°C (–67°F) and as high as 125°C (257°F). These thermal conditions can induce a number of different failure modes, including package and die cracking, bond-wire breakage, moisture ingress, die delamination, tin whisker growth, and solder-joint failure. Low-cost packaging materials used in COTS components are susceptible to all of these failures, while space plastic ICs utilize enhanced-reliability material sets and undergo specialized testing to ensure robustness.

Because of the reliability concerns associated with lead (Pb)-free solder, the military and aerospace industries have been allowed to continue using tin-lead (SnPb) solder. When a typical commercial Pb-free solder ball is combined with SnPb solder, reflow temperature mismatch and insufficient mixing of the two types of solder can create voiding and a solder joint too fragile to withstand board- and system-level temperature cycling and vibration. For ball-grid-array (BGA) packages in space-plastic–grade ICs, SnPb solder balls are used to eliminate the risk of incompatibility within the SnPb solder profile.

Typical commercial ICs are tested only under thermal/moisture testing such as temperature cycling and highly accelerated stress testing (HAST) at the time of development, whereas space plastic ICs typically undergo temperature cycle and HAST lot acceptance continuously throughout production. This practice ensures that any anomalies in package performance can be contained before reaching the customer.

It is possible to select packaging materials like die attach epoxy, mold compound, leadframe, and bond wire for robustness. Typically, formulations of die attach and mold compounds are proprietary to the manufacturer, but there are clear variations across the industry. Space plastic IC makers should select the highest-performing compounds regardless of cost.

A failure mode of particular concern in harsh environments is delamination, wherein the silicon die separates from the package leadframe, and can lead to performance degradation or, in extreme cases, all-out failure of the die. IC package leadframes can have varying performance with respect to cracking and die delamination.

One well-studied technique to improve resistance to die delamination is roughening. This increases the adhesion of the die to the leadframe in conjunction with the correct die attach compound. Roughening adds cost to leadframe manufacturing and, therefore, is not commonly used in the manufacturing of COTS ICs, but it is standard practice for space plastic ICs.

Bond wires can also be a source of failure for ICs in spaceflight. While copper (Cu) bond wire is well-proven in multiple terrestrial applications, and ongoing process improvements have resulted in increased uniformity and improved overall field defect parts per million, spacecraft manufacturers and government organizations still have concerns about the use of Cu wire in space components.

A few key issues identified with Cu wire bonds are:

• Bond integrity issues with Cu wire bonding to aluminum bond pads
• Corrosion due to mold compound interaction or moisture ingress
• Bond-wire neck breaks due to coefficient of thermal expansion mismatch during temperature cycling

While many COTS ICs are manufactured with low-cost Cu wire, space plastic components are manufactured with gold wire, which has been proven in high-reliability applications.

Tin (Sn) whisker growth is a widely studied phenomenon in the aerospace industry and can be observed in ICs in leaded packages that have an Sn-based lead finish. Space plastic IC makers ensure that matte-Sn and other high-Sn content lead finishes are not used on such components. For BGA packages, many commercial IC manufacturers moved to high-Sn content solder in order to eliminate Pb-based solder balls, which is a government requirement for commercial applications.

Outgassing

Outgassing is the release of gas from any material and has been demonstrated in numerous mold compounds used in plastic IC packaging. While this phenomenon has not typically been an issue for terrestrial applications, it can be of particular concern in space. The vacuum environment of space can increase the amount of outgassing compared to atmospheric conditions on Earth. Furthermore, in spacecraft, gases released from IC packages and other sources can condense during extreme cold, creating reduced image quality on optical sensors or other moisture-related failures.

NASA began testing materials for outgassing in 1967 and developed the American Society for Testing and Materials E595 specification, which requires that collected volatile condensable materials are less than 0.1% and the total mass loss is less than 1% after vacuum testing. In order to eliminate risk from outgassing in space applications, space plastic ICs use low-outgassing mold compounds that meet NASA requirements.

Summary

Efforts to ensure the quality and reliability of electronics in spacecraft have continued to evolve since the initial failure of Telstar. The evolution of the spaceflight industry has recently created an opportunity to use electronic components without the full testing and qualification of traditional QMLV components. However, components for these new space systems must still meet mission requirements for radiation tolerance and electrical, thermal, and mechanical reliability. The new class of space-plastic–grade ICs can provide cost and size benefits compared to traditional “full space” components while still contributing to the success of shorter, lower-altitude missions.

References:

LaBel, Kenneth A. “ Radiation Effects on Electronics 101: Simple Concepts and New Challenges.” NASA Electronic Parts and Packaging (NEPP) Webex presentation, April 21, 2004.

“ Space Applications: Radiation-Induced Effects.” Lawrence Berkeley National Laboratory wall chart.

Telstar Wikipedia entry.

TID-depth curves for various orbits around Earth for one-year mission length.

For more in-depth insight into a new era in space electronics, check out these other articles in this AspenCore Special Project.

Space Is Opening Up for Electronics Design

This Special Project on Space Electronics seeks to help make the opportunities in space electronics more accessible by introducing these technologies and design experiences. It starts by discussing the trends and opportunities from this new era in space electronics. It then goes on to provide insights into the design challenges, best practices, tools, and components of electronics design for a space environment.

A new era in Space Electronics

The development of reusable rockets is lowering barriers to both scientific and commercial exploration of space, stimulating increasing interest and investment in space electronics. This Special Project provides designers with a look at the technologies and design practices needed for creating space-worthy electronic designs. These may include, but not be limited to ICs, ASICs, flex-cable, connectors, thermal management for electronics, Rad hard techniques, space-related testing methods, Apollo 1960s electronics and more.

EMC in space: The James Webb space telescope

Testing for EMC on a spacecraft isn’t unlike testing for terrestrial equipment, except for the fact that spacecrafts live in a different electromagnetic environment.

The convergence of traditional and new space electronics solutions

A hybrid approach allows a design team to implement a rad-hard functional base in conjunction with the latest state-of-the-art COTS technologies.

AI and Machine Learning: Shaking up the space industry

Today’s increasingly ambitious mission requirements are demanding higher levels of autonomy and greater navigational precision from spacecraft, requiring more than logic-based AI. High-precision space navigation to small comets and asteroids; entry, descent and landing (EDL) on moons and planets; and, rendezvous and proximity operations (RPO) with both cooperative and uncooperative targets all need sensing and perception capabilities provided by vision-based systems.

Changing trends in designing space electronics

Space electronics has moved-on considerably from the bespoke, radiation-hardened guidance computer and hand-woven memories used on the Apollo missions. Today’s avionics are millions of times more powerful than that used to land man on the moon!

Safe and Affordable Space Travel Starts With Sourcing

Commercial space tourism is on the rise, at a fraction of its predecessor’s cost, creating a cause for passenger safety concerns. One crucial aspect of space flight safety is radiation shielding and mitigation, which New Space ventures have not thoroughly addressed in the past. Traditional space radiation mitigation strategies are costly, yet New Space mitigation techniques may not be suitable for safety critical missions. A solution to this cost-safety tradeoff is Sub-QML hardware electronic devices, which are radiation-hardened by design, yet available at a lower price point due to reduced screening levels.

Mission Critical Space Flight Systems Stay Rad Hard

To understand the effects of radiation on electronic systems and components, we explore the effects of various space radiation environments and their impact on commercial “New Space” and NASA mission profiles.