The accurate prediction o f the operatkg temperatures of critical electronic parts at the component-, board-and system-level is seriously hampered by the lack of reliable, standardised input data. This paper reports on the status of work-in-progress in the 3-year European collaborative project, named DELPHI, whose goal is to solve the aforementioned problem. The project concerns the creation and experimental validation of thermal models (detailed and compact) of a range of electronic parts including mono-chip packages, heat sinks, electrolytic capacitors, transformers and interfacing materials. The purpose of this paper is to give an account of the principal project results. A bibliography of published DELPHI technical papers is provided
Project Funding and PartnersThe DELPHI project which commenced in November 1993, involves six companies from five European member states. The industrial Partners are Thomson CSF, Philips CFT, Alcatel BELL and Alcatel Espace, who respectively manufacture: electronic equipment in the areas of civil and military radar; lighting and consumer electronics; telecommunications equipment; and satellite electronics. The remaining two Partners are: the Project Coordinator Flomerics (a developer of a thermal CAD tool in wide use in the electronic industry) ; and the National Microelectronics Research Centre in Ireland. DELPHI is a three-year project due to end in November 1996.Compact modelfor a mono-chip package a compact model is a thermal resistor network containing no more than 10 (say) nodes. This terminology extends to recogize that detailed models may contain compact models within them: an example of this is the representation in finite-volume software of the lead frame of a PQFP by a layer of uniform property material (Rosten, et al. [5]), an assumption separately validated by a detailed model of the lead Graine within finite-element software.Computational Fluid anamics (CFn)the method for 173 1996 ~~e c~~o n i~ Components and Technology Conference