Here is the abstract you requested from the Thermal_2012 technical program page. This is the original abstract submitted by the author. Any changes to the technical content of the final manuscript published by IMAPS or the presentation that is given during the event is done by the author, not IMAPS.
|Component Compact Modeling in High-power Applications|
|Keywords: compact, model, resistance|
|There has been a lot of effort and debate on compact modeling approaches with respect to their accuracies. For instance, the Delphi model approach from the European consortium derived a decade ago is an attempt to improve the accuracy over the JEDEC’s 2-resistor approach. However, all these methods fall short as the micro-processing components increased in power and power density drastically in the past decade. To exacerbate the issue, the packages have grown to enormous size- 35mm body size was considered relative large body and today it is not uncommon for components to have footprint in the range of 45 to 55 mm. The issue is not with how good or how accurate is the compact model, but it is a nodal mismatch- single node to multi-node system. As long as these compact models are using the package single case node approach, the underlying physics will not be modeled properly for the temperature responses between the component through TIM2 and heat spreading in the heat sink base as well as through the PCB. In this work, these effects are studied for the compact models and compared to the detailed model in an attempt to demonstrate these short falls.|
|Herman Chu, Pricipal Engineer
Cisco Systems, Inc
San Jose, CA