Sophisticated thermal management solutions cool hi-rel systems – Part 1
Today’s demanding military applications require components and systems to push the limits of bandwidth and performance while enduring intense environmental conditions. Survivability is critical, as sophisticated features are only as good as their ability to operate without fail. The rigors of battle in remote locations continue to present an evolving range of unique challenges to military system designers. Everything from severe temperatures and shock and vibration, to explosive decompression, immersion, and exposure to sand and dust are variables that must be considered when building rugged, high performance systems for the armed forces. Innovative thermal management techniques have become essential to meeting these requirements for fault-free performance, and military designers are now making it a priority to solve cooling challenges early in the design phase.
By integrating cooling capabilities into the size, weight, and power (SWaP) protocol, packaging engineers have created SWaP-C (size, weight, power, and cooling) as a focus for next-generation solutions. Using commercial office-the-shelf (COTS) solutions as the foundation for semi- and full-custom thermal management, designers can increase the capabilities of subsystems and enable new designs that meet or exceed current mil/aero thermal requirements. Thermal options will vary accordingly depending on application, expertise, cost, and development time. As a result, engineers need to have a thorough understanding of how designs generate heat and how design choices reliably dissipate that heat. Knowledge of the primary thermal management methods illustrated in this paper will help designers determine the most appropriate path for their specific design.
Thermal demands
The modern military’s more extreme applications and increased performance needs mandate rugged design. The most critical and unique design requirements encompass a range of characteristics grounded in ever-increasing computational power and communication bandwidths. Many airborne and ground mobile applications, for instance, operate high-definition vision systems with real-time processing. These systems drive the need for significant power densities at the board, chassis, and platform levels. The growing deployment of unmanned aerial vehicles (UAVs) combined with current initiatives such as Brigade Combat Team (BCT) modernization has designers dealing with system aggregate power needs of tens of kilowatts. At the same time, system developers are required to deliver smaller, lighter, faster solutions with effective thermal management ensuring extreme reliability.
Higher temperature decreases overall reliability, and more transistors mean more heat is generated within these systems. Faster, higher-density chips equate to higher power densities overall and increased thermal cycling can lead to fatigue failures. Improved cooling has become a priority, as illustrated by designers embracing the concept of SWaP-C.
Thermal management options
Cooling challenges have increased in step with higher performing processors, smaller system footprints, and evolving rugged environments. The overall goal for cooling electronic military equipment is to maximize the flow of thermal energy from heat-generating equipment to a local heat sink. The heat sink could be ambient air, a cold plate, or a liquid exchange system. However, for all the demands of military applications—bandwidth, performance, form factor, and more—designers are always bound by the laws of classical physics.
Newton’s three laws of motion form the foundation of classical physics, which through the application of statistical methods are used to derive the basic laws governing thermodynamics. These laws establish that heat will always move from warmer areas to cooler areas through the action of one or more of the following principal modes of heat transfer (see figure 1):
1. Convection: Energy transfer by mixing action of fluids (gas or liquid)
2. Conduction: Energy transfer from one molecule to another
3. Radiation: Energy transfer by electromagnetic waves
Over time, real-world requirements have shaped these basic tenets into several principal and proven cooling methodologies:
- Forced convection
- Conduction paired with forced air or liquid
- Conduction paired with passive convection
- Conduction paired with a cold plate
The various cooling methods dissipate heat to varying degrees (see table). It’s important to note that there may be considerable variation in these values depending on environmental conditions.
All of these methods benefit from the ever-present effects of cooling by radiation; however, the contribution from radiative cooling is usually ignored unless the overall power dissipation is low. In these low-power scenarios, the fractional contribution of radiation can be significant and should not be ignored.
Military applications can vary greatly in their mission objectives. For example the thermal design goals for short- and long-range UAVs, man-wearable computers, and environments sealed to avoid any number of airborne contaminants may be quite different. A given thermal solution might need to tolerate the corrosive aspects of marine environments, or airborne dust may limit the design’s ability to force ambient air across electronic components. To determine an optimal cooling solution, packaging engineers must fully understand the boundary conditions of the system, its form factors, and its component-level attributes. By analyzing thermal demands of the end-use application, as well as understanding the unique requirements of individual devices designed into the system, designers can determine an optimal cooling scenario.
Forced convection cooling
Forced convection cooling applies air in direct contact with boards and system components via the use of chassis fans or a host platform environmental control unit (ECU). Forced air impinging directly on system boards, power supplies and other components, absorbs heat and then exits the system via exhaust vents.
Forced convection cooling is functionally acceptable if the air is clean and dry, and is ideal for slightly more benign military environments (see figure 2). Assuming a maximum air temperature of 55°C, power dissipation can be as high as 100 W per board.
To increase thermal dissipation, designers can simply use more or bigger fans to move air through the system enclosure, assuming space, cost and noise are not design issues. It is important to note however, that raising the number of fans increases cost and weight while adding points of failure. For more sophisticated applications such as unmanned aerial systems, SWaP constraints force the designer to explore other means for getting rid of heat.
Conduction cooling
Unlike forced convection systems, conduction systems cool by transferring heat energy through direct contact of the heat-generating components to a heatsink such as the system enclosure. Thermal energy then transfers to either a moving air stream or liquid inside hollow side walls, or to external fins for passive convection. In contrast with forced-convection solutions, air moving through a conduction design never physically touches system components.
In a typical conductively cooled system, boards, power supplies, and other components are sealed inside an air tight enclosure. Wedge locks or other mechanical elements clamp the edges of each component to the enclosure’s structure. As the wedge locks expand when tightened, they create a primary cooling path from the heat-producing elements to the chassis. The wedge locks assure the boards are mechanically secure and offer excellent resistance to shock and vibration. The drawback to conduction cooling is that designs tend to be more costly than their forced convection-based counterparts.
Pairing conduction with passive convection
Conduction cooling paired with passive convection offers military designers an alternative for applications in which fan-cooling is impractical. These types of environments may also demand higher MTBF, due to mission critical aspects and/or limited accessibility for maintenance. As a result, a thermal solution with no moving parts may be required.
Conduction-based passive convection solutions do not use a fan in the system. The boards are confined and completely isolated from the ambient environment. The designs remove heat through conduction to a passive cold wall that then convects or radiates the heat away. As the box itself physically heats up, it heats up the air immediately surrounding it. The air’s reduced density causes it to rise and pull in cooler air from beneath, resulting in passive convection. Conduction-based passive systems are available in either standard ARINC 404A style form factors or custom enclosures (see figure 3).
A common myth in systems packaging is that radiation plays a marginal role in cooling electronic equipment. That holds true for higher-power systems with greater levels of power to be dissipated. Engineers should pay attention to the effects of radiative cooling in passively cooled convection systems that operate at low power, however. These smaller, lower-power systems are carving a path in military electronics and radiation can play a significant role in their growing applicability. A generic 8 in. x 12 in. x 7 in. metallic box in an environment at 100°F and 0 Pa air pressure (i.e., a perfect vacuum), for example, can dissipate more than 27 W by radiation effects alone. This is significant for designers, providing additional cooling for enclosures in unmanned aerial vehicles flying at very high altitudes where the air is thin or where there is a lack of infrastructure to support liquid or air cooling.
A passive convection approach can offer scalability and good power dissipation in a sealed system (see figure 4). Efficient thermal design can support fanless operation in severe environments, making the approach compatible with a full range of ground vehicle, UAV, manned airborne or shipborne requirements.
Conduction paired with forced air
In using conduction cooling methods with a fan, the boards are again sealed off from environmental elements. The types of seals used in conduction cooling offer not only protection from environmental contaminants but also protection from EMI effects for conducted, radiated, and emitted electric fields. A rear-mounted fan pulls air through hollow side walls and exhausts it out of the enclosure. The reason air is pulled rather than pushed is to prevent fan-generated heat from being introduced into the air stream and reducing the system’s overall thermal efficiency. Similar to the passive convection enclosures discussed earlier, these conduction-cooled forced-air systems are available in either standard ARINC 404A styles or custom form factors (see figure 5).
Some applications cannot afford access to a replenishable air supply for either forced convection or passive convection cooling. In these cases, a conduction chassis paired with a cold plate may be the optimal thermal configuration (see figure 6). As with any conduction-cooled chassis, heat is conducted to the chassis side walls, but in this case the heat is directed to a bottom- or side-mounted cold plate. The cold plate itself can be either actively or passively cooled.
Another example of liquid cooling is the use of liquid flow-through modules. These designs provide superior thermal performance by delivering coolant closer to the sources of heat within the system, in turn enabling higher power densities. In these designs, liquid, in addition to flowing through the chassis side walls, also flows through special channels on the system boards themselves (see figure 8).
To adequately cover the range of functionality in today’s military mobile device applications, designs must protect systems against heat, dust, and other airborne contaminants. Some of these added requirements can further constrict system airflow, and demand customized attention to thermal management.
Due to the diversity of requirements, optimal designs address performance, cost, reliability needs, and development timelines. Thermal management, in particular, must be dealt with head-on. Packaging engineers first need to calculate a system’s thermal requirements, then make design decisions accordingly. Selection of the appropriate thermal solution also hinges on selecting a manufacturing partner that is experienced across the range of platforms and form factors who can serve as advisor. Ideally this type of manufacturing partnership provides a significant competitive advantage to designers, with all critical levels of expertise combined into a single engineering resource.
Increasingly complex military systems are driving increased challenges in integrating effective thermal management methodologies. Making early choices about power dissipation, design layouts, paths for air flow and overall thermal performance has become essential to developing rugged systems suitable for mission-critical military environments. On the positive side, advancements and the broad scope of thermal methodologies benefit packaging engineers who have a deep understanding of each type of available solution.
Rugged military applications demand it all: higher performance with higher speed and density components, coupled with smaller form factor boards and reduced system footprints. As designers continue to push the envelope with increased functionality, new thermal management options will continue to evolve to satisfy requirements for more efficient cooling solutions to match new standards specifications or increased ruggedness. Managing the complex cooling issues associated with many of these unique or extreme environments requires broad thermal expertise, and a thorough understanding of supporting design choices including working with a strong partner that can provide extensive expertise in how and when to provide proven semi- and full-custom solutions.
In part 2, we will review examples, illustrating how thermal design choices meet the computing challenges of specific military program applications.
David O’Mara is product manager at Kontron. David has a diverse engineering background that includes extensive experience with military and aerospace electronics packaging, pressure and accelerometer sensor design, and high pressure solid state physics. He graduated from UCLA in 1985 with a Bachelor of Science degree in Physics.
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
