Occam Files No. 9
Thermal Management Options with Occam
Reliability experts are fully aware and often warn of the negative effect of thermal exposure and excursions on electronic products. It can be said with a high degree of confidence that the majority of electronic failures can be traced in one way, shape or form to the effects of thermal energy on an electronic assembly. It is also abundantly clear how concerned reliability experts are from the numerous types and wide variety of thermal testing protocols that are used to assert the reliability of electronic products in field use.
However, thermal energy has also long been an important and vital part of nearly all electronics fabrication. For example, high temperatures are used, along with pressure, to make the laminates for printed circuit substrates and to join inner layers to make multilayer PCBs. In addition, high temperatures have long been required as a part of the HASL (hot air solder leveling) process which is used to apply a thin uniform layer of solder to the terminations on a PCB for later assembly. This is actually one of the first places that temperature can effect a PCB quality potentially delaminating the PCB or causing defects in plated through hole and plated micro vias.
Of course, the most obvious use of thermal energy is seen in the component assembly process which is accomplished using molten solder. The temperature required depends on the solder alloy used. For decades, tin-lead eutectic solder was the “go to” alloy. It’s eutectic point was 183C. However with the advent of lead-free solder tin-silver-copper (SAC) alloy became more common and their use temperatures were in the 260C range, a significant increase which proved problematic and it remains bothersome to this day. One interesting fact of interest is that a new term was added to laminate specification, Td, which is the temperature of decomposition of the polymer used in the construction of laminates. It was a “wake up call” for many in the industry.
Heat is nevertheless a useful “tool” in a great many manufacturing processes and a natural byproduct of electronic assemblies in use, the question is how to manage the heat both in manufacturing electronics and in their operation. This is where the Occam Process excels.
Most of the processes used in making an electronics assembly (depending on approach) can be performed at temperatures lower that 150C. Moreover because solder is not prescribed for assembly it is possible to make assemblies almost exclusively of metal. (This was discussed in Occam Files No. 7.) An predominantly metal structure is ideal for thermal management, especially so if the metal is a good conductor of thermal energy such as aluminum is. As mentioned before and repeated here for emphasis, aluminum is also relatively light weight and is dimensionally stable. When aluminum or other metal or combinations of metal (keep in mind that aluminum itself can be plated with a metal such as copper) is used as either the base or the core of an assembly, the entire assembly is a thermal spreader. Moreover, though not yet demonstrated, it is believed possible that a heat pipe could potentially be used either at the center or as a part of a stack of interconnected Occam assemblies to create what might be viewed as a thermal spreading brick of components embedded in or on a metal carrier, interconnecting circuits with insulator layers. It is also worth remembering that the metal cores can serve electrical function as well, such as a ground and/or a power layer and required by the circuit. This is rife with potential benefit.
Because the assembly will not see the temperatures required for soldering, a battery could also be potentially embedded in the assembly and its heat extracted as well. The reader should know that this would be virtually impossible with traditional solder assembly methods.
Finally, the reader should bear in mind that the entire assembly could be plated with a metal jacket, leaving open to the environment only those locations required for powering up and accessing terminations required to use the electronics within. There are some quite interesting possibilities with such structures. I have shown some concept drawings in some of my presentations. If interested, you can contact me directly and I can arrange to get them to you.
The integration of all of these various prospective approaches just described and the intrinsic value that can be accessed with relative ease seems a compelling argument in favor of such structures. These are not stand alone benefits nor are the afterthoughts. Thinking of all of the potential problem ahead of their encounter is vital to harnessing the prospective benefits of integrating the thermal solution into the assembly design. As thermal guru Bernie Seigal a 53 year veteran of thermal measurement and management and author put to me, “This idea allows electronic system designers the opportunity to address thermal energy concerns on the front end of design rather than after the fact”
Presently, and really since the beginning, thermal management has been ignored until the problem is recognized. Then there is a scramble to fix. The heat generated by high end processors today is mind numbing. Thermal densities approach and can even exceed those on an electric stove cook top. Heat pipes and heat spreaders are typically required because water cooling is not an option. However, heat is a problem that can be dealt with in a number of ways on one of them is to reduce the speed of the processor or to reconsider the processor all together including the programming language chose. This has been accomplished by going to multicore processing pioneered by Chuck Moore who also developed Forth Language. This will be discussed in more detail in the next article in the series.
Next time: 144 full-fledged computers on a one chip…
evident from the amount
Next time: