Thermal management critical to military electronics upgrades
With new platforms scarce in today’s budget-constrained defense environment, more funding emphasis has been placed on upgrading the existing electronics suites in avionics, radar, shipboard, and other military applications, which means solving problems such as reducing the heat generated by modern processors in size, weight, and power (SWaP)-constrained environments and managing long-term life cycle costs. In this Q & A with Andy Mason, Head of the Technology Platforms Group at Kontron’s Avionics, Transportation, and Defense division, he discusses the thermal management challenge in embedded computing upgrades and how open architectures and a building block approach that leverages commercial-off-the-shelf (COTS) can help manage it across multiple platforms while keeping development costs down. Edited excerpts follow.
MCHALE REPORT: Please provide a brief description of your responsibility within Kontron and your group’s role within the company.
MASON: My current role at Kontron is Head of the Technology Platforms group, which aims to leverage common hardware and software building blocks into system platform solutions across multiple vertical markets. The primary focus of the Kontron team here in San Diego is avionics, transportation and defense, while industrial automation, medical, and communication verticals are mainly handled by our groups in Europe and Canada.
MCHALE REPORT: Your career in this industry dates back to well before the 1994 memo from then Defense Secretary Bill Perry directing the Department of Defense (DoD) to use COTS technology wherever and whenever possible. What impact did that have on the defense electronics industry and how would you define COTS today at Kontron?
MASON: As I recall, the DoD COTS directive was formalizing a trend that was already well underway within the industry by the early 90s. The “design from scratch” approach previously used on many defense programs was coming under increasing budgetary pressure around that time. VME and CompactPCI started gaining traction in real-time data acquisition, simulation, and command/control systems as these architectures gained wide spread acceptance. Standards groups such as VITA and PICMG were essential to the overall proliferation of COTS within the defense market.
MCHALE REPORT: Are you also seeing an increased demand for COTS technology and commonality from your defense customers? If so, what military applications are the best bets for embedded computing suppliers?
MASON: In the U.S., the past several years have seen constraints on the DoD budget due to sequestration and other political considerations. This has caused the defense prime contractors to look for “best value” COTS solutions to reduce development costs and risk. So the COTS strategy becomes even more important when development budgets are tight.
In the current U.S. defense environment, I don’t foresee significant new aircraft, ships, ground vehicles, or UAV [unmanned aerial vehicle] platforms being initiated in the near term. So the best bet from my perspective is to focus on technology insertions on existing platforms. Using COTS buildings blocks and COTS systems based on latest embedded technologies can provide additional capabilities and extend lifetime of these legacy platforms.
MCHALE REPORT: Kontron is a global company, based out of Germany. Is there a similar approach internationally toward COTS and open architectures that we see here in the United States?
MASON: We see similar COTS trends in the overseas defense market – Europe and elsewhere. The main difference is that the size of the programs is generally smaller than in the U.S.
MCHALE REPORT: COTS use and open architectures also means more leveraging of commercial processing technology by military integrators for radar, electronic warfare, and similar applications. However, the heat generated by these components can be counterproductive to keeping costs down for the end-user. How do you balance that along with the obsolescence challenges of commercial technology?
MASON: You’ve identified two key points that are critical to our defense customers – thermal management and obsolescence management.
On the thermal side, we first consider the platform installation constraints when proposing a thermal solution. A military ground vehicle may have the plumbing and weight budget for a liquid cooled system, while a small UAV probably does not. There are also cost impacts that go along with the different thermal solutions, which may cause us to recommend one option over another. The COTS approach we discussed earlier applies to thermal management as well as boards and systems – it is best to have a selection of proven cooling solutions that can be adapted to different platform requirements.
Obsolescence management, especially for defense and aviation customers, needs to be planned for at the earliest stages of a program. Customers understand the benefits of COTS, but they also understand the penalties if a product goes end-of-life unexpectedly. Our recommendation is that experienced COTS suppliers, such as Kontron, are best suited to manage obsolescence and maximize product lifetime.
MCHALE REPORT: It seems every piece of electronic equipment today is getting smaller – GPS systems, radios, etc. How have reduced SWaP requirements affected ruggedization and thermal management challenges in your military designs? What are the tradeoffs with smaller tech?
MASON: Small form factor, ruggedized systems have been a differentiator for Kontron in the defense, avionics, and transportation sectors over the past few years. We expect this trend to continue. Most of these systems are module based, where Kontron has a strong product portfolio of processor and switch building blocks. COM Express and SMARC are the common module standards that we use to build system-level products such as COBALT and AIRES for the defense market.
MCHALE REPORT. What other factors are driving thermal-management and rugged solutions today in military embedded computing systems?
MASON: Aside from SWaP and SWaP-C [SWaP-cost] considerations, we are constantly pushing the envelope on High Performance Embedded Computing (HPEC), which involves both processing and data interconnects. Regarding interconnects our Telecom group usually is a step or two ahead of defense market, with products based on 10 gigabit Ethernet and 40 gigabit Ethernet, with 100- gigabit Ethernet coming soon. We will then leverage that know-how for defense applications. Regarding processing trends for HPEC, there has been a lot of interest for next generation PCI Express solutions.
MCHALE REPORT: Are you and your team exploring any new types of cooling/thermal management techniques for future systems?
MASON: One area of focus is optimizing proven thermal/cooling solutions for new applications. In other words “take what we have already done, and improve on it”. This may involve looking at new materials or design approaches for heat sinking, assessing the latest high-speed fan solutions on the market, or developing cost effective approaches for liquid cooling at the system level.
We also allocate a certain amount of our R&D budget for more fundamental investigations and proof of concepts.
MCHALE REPORT: Do other markets/applications Kontron plays in have similar ruggedization and thermal management requirements to the military such as industrial and commercial avionics applications?
MASON: One of the benefits of the common platform approach used by Kontron is that we can leverage ruggedized products developed for one vertical market into another. For example, the COBALT small form factor rugged computer family meets the ruggedization requirements of military aircraft, UAVs and ground vehicles as well as those of business aviation applications. The thermal and ruggedization requirements for many of the target environments are very similar and we can use the same design techniques to develop a reliable solution.
MCHALE REPORT: Looking forward, what disruptive technology or innovation will be a game-changer for military embedded computing applications? Predict the future.
MASON: We see continued proliferation of autonomous and unmanned vehicles, across many vertical markets, including defense. Computer vision, machine learning, time sensitive networks, and platform connectivity are some of the interesting application areas.