Page 1 of 1
Despite their technical sophistication, defense electronics systems must deliver uncompromised performance under difficult environmental conditions, including excessive heat, humidity, poor air quality, high altitude, shock and vibration. Embedded computers must be able to keep their electronics from overheating, even when temperatures range up to 55°C and the air is too thin to be used for cooling. At the same time, they must possess the enhanced mechanical integrity to withstand high shock and vibration forces at various frequencies.
In the recent past, requirements for rugged defense electronics systems were commonly met by ruggedizing various individual slot cards (often VME modules), then configuring them in a ruggedized chassis. In many situations, a standard, air-cooled board was given enhanced mechanical integrity to withstand high shock and vibration forces at various frequencies. This usually took the form of metal stiffening, often down the center of the module to control response to lower frequency vibration, though sometimes it was applied around the edges or across the surface. Further modifications were made to support conduction-cooling if the system was to be deployed at high altitudes or in confined spaces.
The Individual Card Approach
This individual card approach offered the advantage of flexibility. An integration team, usually an application-focused team from a prime contractor, could work with slot cards from multiple vendors, putting them together into a chassis-level system designed to meet their program-specific needs—the integration team could select from among technologies from many card vendors. Then application software would be developed, de-bugged and tested on lab environment development systems.
At some point a deployable version of the system would be developed by ruggedizing the slot cards; often the ruggedization was done by the card vendors specifically to meet specifications established by the integration team. The integration team would also select a rugged chassis, again enjoying flexibility in selecting or modifying from a variety of market choices.
While selection flexibility is a positive characteristic, there are two main disadvantages to the individual card approach. The first is a negative effect on time-to-deployment. After the huge integration challenge of getting application software to run on multiple modules, the engineering team then faced the challenge of ruggedizing a collection of disparate boards.
Until the recent advent of the VITA 47 REDI standard, there was no formal, industry-wide method for defining levels of ruggedization. This lack of rugged standards, combined with multiple engineering approaches by vendors at the slot-card level, made integration for rugged systems very time-consuming. However, for many defense programs, an extended period of deployable systems integration was not significant because overall development time was measured in multiple years.
The second disadvantage to working with individual cards is that it rarely results in an optimal thermal and mechanical design at the system level. Some components may be overdesigned for a given application; some characteristics, like heat dissipation, may be widely non-uniform. But this disadvantage was also acceptable as long as the functional requirements of the system were met
Conditions are changing, however, and a new set of challenges faced by defense programs have driven the rugged requirements for defense electronics to new levels. First, defense forces are meeting the need for more intelligence-gathering assets by placing sensors on unmanned vehicles (UVs)—which are airborne (UAVs), ground-based, or undersea. Early implementations of these UVs, such as the Global Hawk and Predator UAVs (Figure 1), are fairly large platforms, but each succeeding generation is smaller. The resulting challenge is to make the sensor-supporting processing power fit into a smaller Size, Weight and Power (SWaP) budget. These smaller computing packages must also be rugged enough to withstand operation within deployed UVs.
Early UAV implementations like the Predator MQ-1B UAV are fairly large platforms, but each succeeding generation is smaller. The challenge is to make the sensor-supporting processing power fit into a smaller Size, Weight and Power (SWaP) budget.
A second significant challenge is driven by changes within computing technology. New components, especially processors, are orders of magnitude faster, but they are also much, much hotter, magnifying the cooling challenge. In 1992, a 66 MHz CPU consumed about 7W of power. In an office environment it would not even need a cooling fan. Now processors will often draw over 50W, sometimes over 150W, depending upon clock speed, core type and processing load.
Taken together, defense program teams have a situation requiring that much hotter, albeit faster, components must operate with increasingly difficult SWaP restrictions. Meeting this new level of challenge requires a systems-level approach to solutions design. These system-level approaches can be divided into two tiers. First, a broad range of applications can be addressed by a more rigorous, standards-based approach. The evolution of the OpenVPX standard is proving to be a great benefit to this systems-level approach. Created to improve interoperability of COTS modules, OpenVPX achieves this goal by implementation of predefined system topologies that simplify integration of components while retaining a significant range of configuration flexibility. The result is reduced program risk and faster development.
When programs reach the stage for integrating a deployable system, OpenVPX leverages the VITA 47 and VITA 48 standards. Designs can be optimized at the systems level for ruggedness and cooling, while the use of standards-based components—modules and chassis—reduces the integration effort and speeds time-to-deployment. Adhering to a systems level approach based on open industry standards gains these advantages while retaining a large degree of design flexibility.
A range of OpenVPX modules and chassis is currently available for this style of systems design. Mercury Computer Systems offers a broad choice of both 3U and 6U options, including development, rugged air-cooled and rugged conduction-
cooled modules. The value of these modules is enhanced by systems integration expertise, available to support program teams in implementing standards-based, system-level rugged solutions.
However, for defense application with the most stringent SWaP constraints, a stand-alone rugged-box approach has taken shape, with computing, I/O and chassis enclosure all designed together as a single, preconfigured solution. This approach trades off some flexibility for maximum ruggedness, cooling efficiency, performance per watt and performance per cubic centimeter of space. Figure 2 shows a sample set of system specifications that help to give shape to this design avenue.
Meeting today’s tighter smaller Size, Weight and Power (SWaP) requirements calls for a rethink in rugged box-level system implementations.
Designing that style of ultra-compact processing platform is possible if rugged characteristics are targeted from the very beginning, including interdependencies between modules and enclosures. For example, in a system constrained to such a small size, conduction-cooling provides an efficient means to draw heat outward to the walls of the enclosure. Furthermore, if each module is designed with a custom heat sink, then using a wedge-locking mechanism between that heat sink and the enclosure walls will ensure maximal heat conductivity with the positive side effect of increasing reliability across a range of environmental conditions (including shock, vibration and temperature changes).
Removing the Heat
There is still the final step in dealing with heat dissipation, removing the heat generated by 100 Gflops of processing from the enclosure. One design option is to use flow-through liquid sidewalls in the enclosure, exploiting the fact that the thermal capacity of a liquid is much greater than that of air. In addition, unlike air, the cooling capacity of a liquid is unaffected by altitude.
This approach puts the final cooling burden on the platform to supply the liquid flow to take away the heat, but some platforms already support other liquid-cooled electronics. Almost any liquid can be used, provided the liquid temperature and flow rate are sufficient. Platform cooling strategies can also be very creative, such as platforms that use their own fuel, moving from storage tank to engine to cool electronics.
Ribbed Chassis for Heat Flow
Another design option for dissipating heat from the enclosure is simple conduction to the outside environment. In this technique, a ribbed chassis maximizes heat transfer to surrounding air while the metal base conducts heat directly to any supporting platform. Although this approach is much simpler to implement than liquid flow-through, heat removal efficiency is highly dependent on the surrounding temperature and altitude during operation.
Mercury offers a choice of fully integrated ultra-compact embedded computers in the Ensemble 1000 family. The PowerBlock 50 (Figure 3) and the PowerBlock 15 meet the stringent demands of sensor imaging for next-generation unmanned platforms, featuring a modular architecture that allows for flexible configurations of multiple processors, including PowerQUICC III processors, Intel x86 processors, FPGAs and graphics processing units (GPUs).
The PowerBlock 50 System provides a modular architecture that allows for flexible configurations of multiple processors and delivers over 100 Gflops of processing power in a small, lightweight package weighing less than 10 pounds.
Mercury Computer Systems