Close Advertisement

TECH RECON

Space Poses Unique Challenges for Commercial Components

The idea of ruggedizing commercial components for space use is appealing. But there are numerous issues to consider from the transistor level all the way up to the system-level.

DOUG PATTERSON, VP, MILITARY AND AEROSPACE AITECH DEFENSE SYSTEMS

Keywords in this Article:

No Keywords

  • Page 1 of 1
    Bookmark and Share

Article Media

 

With the recent shift in space electronics to look towards using ruggedizing commercial components as a cost-efficient alternative to 'space-qualified' devices, the question of reliability has been asked repeatedly. But the answer isn't simply "yes" or "no" as there are just too many variables to consider when factoring in the effects of space on non-qualified components. Depending on the program's mission requirements, some limits may be OK to push, while others may yield catastrophic failures.

Another, and equally as important, question to ask is, why did the space industry look towards commercial components in the first place? The answer is two-fold. First, designers were looking to emulate the functionality they can employ on their desktop PCs in the space environment, but more often than not, space-qualified components were limited in certain abilities. Only so much could be done with the traditional set of components earmarked for in-orbit applications. Second-as its moniker suggests-a space-qualified component has undergone rigorous, and therefore costly, testing and validation. With research and development budgets being cut tremendously across several areas of defense and military programs, ways to reduce overhead and extend the value of each dollar was high on the designers' priority list. And not having to wait for testing results would certainly speed up time to market.

Ruggedizing less expensive, yet more feature-rich, commercial components for space seemed like the Holy Grail. And commercial components themselves had advanced in form and function, making them potentially viable alternatives to qualified components. So theoretically, the answer is yes-commercial components should be able to be used in space applications, but let's actually look at this concept in application.

Why Does Qualification Exist?

Space is not only one of the harshest environments that electronics need to operate in, but it's also one of the hardest to replicate. Testing and validation of mission critical systems used in space exist for a reason. Success in a space system is defined by its continued reliability, autonomous operation and unwavering communication within its network. (Figure 1)

Figure 1
Space electronics endure the harshest environmental conditions from temperature extremes to high doses of radiation, while still maintaining reliable operation.

The system essentially needs to survive wholly on its own, without manual intervention. When a problem arises, tech support can't exactly take the unit out of orbit and into the shop for testing and repair. And launching a satellite or other space system has its own associated costs and timeframe, so it just isn't feasible to send a system into space to 'test the waters.' Once it's deployed, it needs to function as designed. Period. So, although commercial components may not undergo the rigors of space-qualification, there is still a litany of aspects to consider before specifying something into a system destined for orbit. Screening is still paramount to ensure the components meet a minimum level of quality and reliability.

Commercial Components in Space

One of the biggest attributes of components that can reliably operate in space is their resistance to radiation. Fortunately, some components are inherently radiation-resistant, which is extremely critical when evaluating commercial components for space applications. Most diodes and crystal oscillators as well as GaAs technologies and bipolar devices with low dose characterization offer properties of radiation resistance that can give a certain edge to components being validated for space use.

Take, for example, Silicon on insulator (SOI) technology. The inherent transistor construction of SOI-based microprocessors eliminates the parasitic SCR and therefore the effects of total dose radiation and single event upsets. This is extended during full cache utilization, since the radiation tolerance applies to the L1 instruction and data caches as well as the L2 cache on the die. And the cache arrays feature error correction mechanisms that minimize data corruption from single event upsets. Risk associated with these types of components is further reduced through derating and screening processes that tighten the design margins.

Of course, all of the components discussed still need to be verified and tested to ensure proper operation, but this is typically done to a far lesser (and less costly) degree than components that are truly space-qualified. Commercial components not only save costs in testing and validation, but also in implementation, since commercially-available software tools and real-time operating systems can then be employed. This lowers NRE, while providing more advanced design functionality. Designers are provided with more cost-effective tools and components that allow them to build complex designs in a much shorter timeframe and with far reduced expenses.

Assessing the Challenges

So if commercial components seem to have the capabilities to withstand the harsh environment of space, why is there this big question about their use in space? It's because along with these benefits, there are challenges that need to be considered as well. Once a space program has been defined (e.g. Earth Orbital, Deep Space, etc.), its specification are pretty well set in stone. And many times there are extremely unique circumstances that have been considered and evaluated to ensure the system can properly accommodate them.

Space, weight, functionality, extended operating temperature, EMI/EMC and radiation shielding and a host of other attributes are all taken into account when specifying components for a space system. One small shift in a component's internal functional thermal envelop and the entire system architecture can be sent out of whack.

Because they are used in the rapidly changing world of consumer electronics, commercial components are much more dynamic by nature. Frequently, technology changes happen with barely enough time for designers to learn the full part numbers of the parts they are using. And the sheer quantity of components produced precludes many manufacturers from employing a decent traceability system. Are the parts being produced from the same facility/fabrication line or are there multiple fab sources of the same part located in different countries from the same company? If so, how can you be sure of the integrity of one batch of products from the next? And with product development and release happening at a breakneck pace, obsolescence is the norm for many a component in the commercial world.

Space programs are designed for far longer life spans than today's smart phones. How can a system that defines itself on long-term availability, backwards compatibility and reliable operation manage these risks associated with commercial components?

Radiation's Impact on Electronics

Back to the initial question-can commercial components successfully be deployed in space systems? We'll hedge a tentative yes, with a few caveats. But important to note is that radiation tolerance in space components defines the success or failure of a mission, so accounting for the proper levels before launch is critical.

Radiation has one of the largest impacts on Earth orbital systems, and two of the most important aspects to consider are the length of the mission and which orbit will the system operate in, since these determine the type and quantity of radiation that a system will be exposed to. The criticality of the mission is also a key area to consider (Figure 2).

Figure 2
Electronics used in space applications need to accommodate the specific orbit, mission length and critical nature of the flight.

The majority of CMOS-based processors available today have shown dismal results in TID (total ionizing dose) that severely limits their viability in long-term space missions. With most coming in with a TID of only 400 Rads(Si) when tested at heavy ion cyclotrons, these components might last a month in a low earth orbit platform. A system designer needs to weigh this information against the known radiation resistance of the SOI microprocessor noted earlier.

And as more processor cores are added, the radiation tolerance numbers may get even worse. A heavy ion can do far worse damage to a smaller transistor or gate since even in space, F=ma. On the brighter side, since the transistors and gates are smaller, the probability of them taking a hit by the same ion is lower, so there are always trade-offs to be made.

Cumulative Radiation Effects

Another area to consider is overall radiation exposure and its cumulative effect-how much can a component 'soak in' before it just fails. The typical range that radiation tolerant electronics need to meet is 15 to 50K Rads(Si), with rad hard reaching over 100K Rads(Si) and more. Yet most unqualified commercial components top out at maybe 1K Rads(Si)-again a number that will be reached before the system has been in orbit for one month.

If the needed time in flight will be short, say less than 3 years, and in low earth orbit, some commercial components will reasonably withstand radiation effects for a decent amount of time if properly shielded. Of course, orbital altitude and angle of inclination are always factors to consider when designing a system for space, as is looking for new ways to ensure reliable operation incorporating redundancy.

For example, the trend to launch satellite clusters, instead of one large unit, helps to spread the burden of reliability across multiple units. If one satellite in the cluster fails, and if the system is designed as such, another can pick up where the failed unit left off, so there is no loss in communication and the 'constellation' stays operational. This is a shift from the traditional method of launching one super rad-hard mega-satellite solely responsible for the mission's entire operation.

Most clusters are launched in low earth orbit, so commercial components may make sense in this environment. But for longer missions, and deeper space applications, these components have not been proven to possess the needed endurance to withstand repeated and prolonged exposure to harmful radiation.

Proper Implementation

Techniques employed to adequately prepare electronics for space include modified circuit designs, supporting real-time software and cache validation as well as scrubbing techniques, qualified EEE Grade 2 or Grade 3 component selection and appropriate testing and certification. These methods are applied to COTS single board computers, mezzanine cards, and other subsystem solutions that can then deliver cost-effective performance with the functionality demanded by manned and unmanned/robotic space applications (Figure 3).

Figure 3
This space-qualified, radiation-hardened SBC uses a standard 3U conduction-cooled CompactPCI form factor with a conduction-cooled PMC I/O expansion slot to withstand both the thermal and radiation effects of space environments.

The key comes down to proper EEE component selection and appropriate validation for the space mission at hand. It won't be worth the short term cost savings if your system doesn't hold up for the long term, due to a premature component failure.

Aitech Defense Systems
Chatsworth, CA
(888) 248-3248
www.rugged.com

 

LEAVE A COMMENT