Materials for Automotive Applications: Thermal Management Issues
For me, the highlight of the recent HDP User Group Automotive Technology Webinar was Alun Morgan’s presentation on materials for automotive applications. Introduced by HDPUG marketing director Larry Marcanti and moderated by TechSearch International President Jan Vardeman, this forward-looking informational session covered the latest developments in automotive standards and automotive electronic packaging, with expert speakers from ZVEI, Infineon, UL, and TTM.
Morgan, technology ambassador for Ventec International Group and president of EIPC, reflected on the evolution of electronics in vehicles since the first introduction of the valve radio in the early 1930s and its rapid advancement over the last two decades. In the year 2000, it was estimated that 18% of the value of the car lay in its electronics. At present, the figure is 40% and heading toward 50%—a massive change in the configuration of the vehicle and an illustration of the significance of its electronic systems. Besides their essential functionality, most of these systems were potential hotspots requiring thermal management. “A vehicle is effectively a power device.” Automotive lighting was the particularly critical area on which he chose to focus his discussion.
LED lighting technology was no longer confined to premium vehicles. It was becoming universal, even for headlights where the high production costs that had previously been a barrier were becoming outweighed by benefits. Compared with halogen and xenon lighting, LEDs offered brighter light with lower energy consumption and a lifetime longer than the vehicle itself so that they could be considered permanent elements of the design rather than consumable components.
Reviewing the history of LED development, Morgan referred to Haitz's law, the LED counterpart to Moore’s law. In 2000, Dr. Roland Haitz forecasted that for a given wavelength of light, for every decade, the cost per lumen would fall by a factor of 10, and the amount of light generated per LED package would increase by a factor of 20. In the 18 years since 2002, efficiency has increased eight-fold, cost per lumen has fallen by a factor of more than 100, and expected lifetime has increased to more than 100,000 hours, which represented 12 years of continuous use—effectively outlasting the useful life of the car and reducing the need for switches and connectors.
But notwithstanding the efficiency with which LEDs could convert electrical power to light, the fact remained that a large proportion of their electrical power input was output as heat. The management of that heat was critical in determining both the life expectancy and the quality of light emitted. Morgan demonstrated the dependence of light output and colour on the junction temperature of the LED before explaining the science behind the thermal management of electronic devices with a lesson in the basics of thermodynamics—the branch of physics concerned with the relationships between heat and other forms of energy.
Thankfully, he didn’t delve too deeply into the mathematics. He used straightforward, practical examples to illustrate the main points. The first law stated that in a closed system, energy could neither be created nor destroyed; it could only be converted into different forms. The second law was that heat would always flow from a region of high temperature to one of lower temperature.
Morgan introduced the concept of entropy—everything going from an ordered state to a less-ordered state, and the entropy of a system increases with time. He used a teenager’s bedroom as one example and even predicted the heat-death of the universe, although that was not likely to happen until very many years from now.
The third law stated that the entropy of a system approached a constant value as the temperature approached absolute zero. An additional or “zeroth” law stated that if two thermodynamic systems were each in thermal equilibrium with a third one, then they were in thermal equilibrium with each other. The second law was the key to thermal management—order always going to disorder, with no way back.
Thermal transfer could be accomplished by three principal mechanisms: radiation, conduction, and convection. Morgan’s example of all three was a saucepan being held above an electric hotplate. The hotplate radiated heat towards the pan; the handle conducted heat toward the hand, and convection moved heat within the fluid. In the electronics context, the primary mechanism was thermal conduction.
The conventional method of managing the heat emitted by LEDs was to use a discrete heat sink coupled to the device with a thermal interface material. His example showed equivalent thermal resistors in series, representing individual values for components, substrate technologies, and cooling systems. But there had been major developments in materials designed for removing heat: metal-in-board constructions, insulated metal substrates, thermally conductive prepregs, and thermally conductive coatings.
Themally conductive PCBs were extensively used to keep heat-generating components cooler and reduce system costs. In Morgan’s example, an insulated metal substrate (IMS) offered a cost-effective alternative to conventional methods by simplifying designs, reducing weight, eliminating the need for thermal interface materials, and using standard PCB manufacturing techniques.
In terms of thermal dissipation, whereas standard FR-4 laminate had a typical value of 0.25 watts per metre Kelvin (W/mK), glass-reinforced IMS laminates were available with values between 1 W/mK and 3 W/mK, and non-reinforced IMS laminates with values of 3W/mK to 8W/mK and higher. These were bendable and capable of being mechanically formed. Morgan explained the structure of an insulated metal substrate: a metal base layer of aluminium or copper, a dielectric layer, and a copper foil layer capable of being photoimaged and etched to form a circuit. A current innovation was a coating for the aluminium base layer, which helped in distributing the heat uniformly and also improved its emissivity—a measure of how well a surface would radiate or emit heat.
Back to thermodynamics—“A fun topic but full of maths”—Morgan gave an impressive demonstration of his technical numeracy in deriving equations for convective and radiative heat transfer in terms of the coefficient of heat transfer of the process, the temperature differential, the Stefan-Boltzmann constant, and the emissivity. He described how a change in emissivity could make a substantial difference to the performance of an IMS system and showed that Ventec’s newly developed surface coating could double the amount of heat removed compared to uncoated aluminium.
Listing the thermal conductivities of a wide range of materials—including air at 0.024W/mK, epoxy at 0.35 W/mK, glass at 1.05 W/mK, silicon at 149 W/mK, aluminium at 250 W/mK, and copper at 400 W/mK—Morgan gave a simple example of how even a basic 1W/mK IMS material dissipated hotspots and distributed heat far more uniformly that an equivalent assembly on a standard FR-4 substrate.
At a more advanced level, Ventec continued to work closely with their customers’ engineers to analyse and model specific thermal management challenges and either select appropriate IMS materials from their product range or quickly develop bespoke IMS solutions with the exact thermal properties needed by the OEM. Morgan showed examples of modelling calculations for real applications, illustrating how Ventec’s agility had enabled OEMs to bring new designs to market with reduced development timescales.
Failure analysis was a related topic for discussion. Morgan reviewed two case histories, identified the failure mechanisms, and described how Ventec had provided technical solutions. The first was an automotive headlamp assembly that had failed electrical function test on thermal cycling, exhibiting conductor fractures and solder-joint cracking. The second was a ceramic packaged LED unit that showed similar failures, although under different test conditions. In both case studies, thermal expansion mismatch proved to be the root cause. In the first example, the problem had been resolved by selecting a low-expansion aluminium alloy; in the second, by reducing the elastic modulus of the IMS dielectric.
On the subject of test methods, there was consensus among automotive suppliers that there should be standardised test methods and conditions. The problem was that every supplier had their own methods and agreed that theirs should be the ones to be used!
Although Alun Morgan focused this presentation on materials for thermal management, he made it clear that these only represented one specialist category of Ventec’s extensive range of base material solutions for automotive electronics. And he commented that the industry, in general, was ready for a change from categorising materials based on their structure and chemistry on definitions developed in the 1960s.
Ventec was working toward grouping materials appropriate to particular industry sectors and introduced autolam—a base material solutions set specifically curated for the diverse and unique requirements of automotive applications, or in Morgan’s words, “To relieve the designer of the laborious task of scrutinising 130 IPC slash sheets to find the product they need.”
Editor’s note: To read more about thermal management, download “The Printed circuit Designer’s Guide to Thermal Mangement With Insulated Metal Substrates” by Didier Mauve and Ian Mayoh from Ventec International Group.