The importance of thinking thermal early in the design process

5 mins read

Thermal management is often left until pretty near the end of the design engineer's checklist of elements to cover off on a project. There are often reasons offered for this, some being more logical and valid than others. But by building an understanding of heat management techniques and working with experts in what still remains a somewhat mysterious aspect of electronic product design, there are significant performance, cost and reliability benefits to be had. These can prove to be crucial differentiators in the final design specification and be the difference between a compelling, high selling idea and a 'me too' product that may be late to market and taken an inefficient and torturous route to get there.

In most sectors of the electronics industry, technology advancements continue at pace. Smaller, faster, increased functionality, better durability and reliability, are all demands that get fed to design engineers from the sales and marketing teams in sectors ranging from automotive and aerospace to mobile communications and consumer. In most applications some if not all of these can be influenced positively through the provision of optimised thermal management of the electronics contained within a product enclosure. As a bare minimum, the strategies and materials employed under the heading 'thermal management' need to ensure that all electronic components in a product operate within their rated temperature range. This might be -20°C to +85°C for consumer applications and in the region of -40°C to +125°C for challenging military, aerospace or under bonnet automotive designs. For companies who specialise not only in supplying state of the art heat dissipating materials, but also in offering thermal management design support and consultancy, being called in to address thermal problems early in the design process allows proper, optimised solutions to be developed rather than merely 'fixing' a thermal problem given a fait-accompli in terms of the product's electronic design and packaging. When looking at thermal material choices, the obvious option – from an individual part cost perspective – is not always the most economical or effective from an overall product point-of-view. Often, by choosing something other than the obvious or the previous design iteration default material or solution can yield major benefits. These could be overall bill of material savings, weight reductions and better reliability. Perhaps a good way to illustrate the point is to consider a theoretical scenario in a particularly challenging industry sector. The automotive sector is perhaps one of the best to demonstrate the benefits of dealing with thermal management expertly and as early as possible in the design process – as an intrinsic part of the design, not in a linear fashion as one of the last steps. Electronics content value per vehicle continues in what seems an almost relentless surge. The effect on demand for electronic component suppliers is magnified by increasing global light vehicle sales which were 84.3 million in 2013, up 3.8% year-on-year. 2014 is expected to see that figure reach 87 million and the number seems likely to top the magic 100 million in 2018. European numbers have declined for six consecutive years, but the net growth has been driven by demand from China and other emerging markets where car ownership is coming into reach for what are huge populations. More and more vehicle electronic systems in established areas such as convenience, safety, body electronics and security are being supplemented by an explosion in advanced driver assistance systems (ADAS) and driver information. Hybrid electric and pure electric vehicles have even greater demand for electronic systems, especially in the area of powertrain. All of this electronic content translates into a huge number of thermal management challenges that must be addressed effectively by the automotive designers. Ask an automotive designer what is most the important aspect of his electronic system design whether it be a park assist module or adaptive front lighting motor control, and you might get the not so helpful answer 'everything'. That is: performance, weight, size, cost, reliability just to name the main few. A tough challenge indeed, but a good thermal management strategy can help all of the above. Performance – an automotive designer will want his module to maintain its stated level of high performance over the life of the vehicle. A thermal management solution that ensures all the components within both the power and signal circuits in a module sit well within their individual max. and min. operating temperatures will go a long way to ensuring this. If you were to look in detail at the performance graphs of most electronic devices you will see that just within – not just beyond – the margins, outputs, stability and other parameters start to drift. The compound effect of this happening for a lot of components in a circuit can be significant and the difference between a module working and not working. Weight and size – These two parameters are usually closely linked and of crucial importance in a passenger car. Just a few grams of weight saving in many different areas of the vehicle can excite the automotive designer as he sees the cumulative effect translated into improved vehicle economy, lower emissions and even the chance to use a smaller power unit, whether it is combustion, hybrid or pure electric. As the demand for space – both in the passenger cabin and under bonnet – increases with the growing amount of electronic content on the typical vehicle, so the pressure for smaller, lighter modules grows too. Advances in semiconductor and power electronics design mean smaller topologies, and clever integration of many functions into a single chip or device shrink the total board space needed to effect a working design. The resultant space and weight savings can all be undone by a poor thermal management approach. For example, choosing the right material can allow the use of an enclosure or chassis as a heat spreader or dissipater as opposed to needing to have a dedicated, heavy and space consuming heatsink. Also self-adhesive interfaces or materials that form a permanent bond when first heated by components in a powered-up circuit, can negate the need for weighty, expensive and space consuming screws and other fixings. A further potential benefit here could be the simplification and speeding of the module assembly procedure. Cost – Whilst selecting a more expensive, higher performance thermal interface material may seem a little counter-intuitive, if chosen for the right reasons, it can lead to more than compensatory savings elsewhere in the vehicle electronics module. For example and as already mentioned, a thermal material that forms a structural bond can allow the removal of fixing screws and related hardware from the bill of materials. Also on the hardware side, a high-performing and therefore more expensive material (thermal performance is linked to cost due to the use of more exotic ingredients in the compound) can permit the use of a smaller, less complex profile and therefore less costly heatsink. It can also allow components to be driven harder and therefore may allow a reduction in the overall number required. Again, a significant cost advantage can be realised which is far above the increased cost of a higher performing thermal material. Reliability – In automotive applications the reliability of on-board electronics can be tested by both long and short bursts of 'activity' (i.e. car journeys). Different but stringent demands need to be endured and electronics modules can be required to operate in extremely cold environments that rapidly become very hot environments as the vehicle and its engine warm up. A good thermal solution will allow an electronic component to reach its optimum operating temperature but not drift upwards beyond that to a point where life expectancy, stability and sustained performance are jeopardised. Thermal interface materials are passive materials in that they do not rely on other factors to make them work; this is crucial to their use as a dependable thermal solution. For example a cooling fan may provide a way of effectively cooling a component or module, but if its power supply were interrupted, its bearing were to fail or a rogue object became jammed in its blades, then very quickly the components it was cooling would overheat and fail. In some instances, particular thermal interface materials may also perform a secondary function that can also enhance reliability. For example, a gap filling thermal material fitted between a component needing heat management and a heat spreader or heatsink, may also dampen potentially harmful mechanical vibration. Or a material that also bonds a component to a heat dissipating surface will alleviate the need for fixings that may work loose over time and lead to vibration and even thermal runaway for the component. Summary So the message is clear: it is very important to consider and deal with thermal challenges early in the design process and with assistance of specialists if you don't have them in house. Doing this can lead to a better, more reliable end product with predictable performance that avoids the risk of components running at excessive temperatures that shorten their life expectancy and risk erratic performance. Highly desirable associated benefits can be bill of material plus weight and cost savings, the ability to use more standard components as opposed to wide temperature range niche products, greater on-PCB component densities and an all-round more elegant and ergonomic design. James Stratford is the technical director of Universal Science Ltd.