The most recent International Standards Meeting occurred in San Diego, CA,on March 12, 1999, the Friday after SEMI-THERM XV. The meeting started at 8:30 AM and was well attended (click here to view the attendance list). Tom Tarter (tom.tarter@amd.com) and Bernie Siegal (bsiegal@thermengr.com) co-chaired the meeting.
The purpose of the International Standards Meeting is to provide a forum for discussion of semiconductor thermal issues, at the component level and at the component-in-a-system level, without ties to any regional or industry segement groups or organization. Emphassis is placed on reviews of activities of standards creation orgainizations in Japan, Europe and the United States.
After Tom Tarter's welcoming remarks, the meeting was turned over to Dr. Bruce Guenin, the new chairman of the EIA/JEDEC JC15.1 subcommittee, for a review of his committee's recent activities. He presented the committee's roadmap, showing published standards, work-in-progress (i.e., in the ballot process), and future activity plans. Dr. Guenin invited members of the international thermal community to get involved in the ballot process, even if informally, and to consider taking on some of the tasks within the future roadmap objectives.
Dr. Arnold Free then made a presentation intitled "Driving PCA Thermal Analysis Using IDF 4.0 Data Exchange". According to Dr. Free, the ".idf" file format bridges the gap between various computer-assisted design activities (i.e., ECAD, MCAD and CAE) with a rich definition of PCA (Printed Circuit Assembly) design. The ".idf" format, originally developed by Mentor Graphics and SDRC in 1992 but is now managed by Intermedius, is currently supported by over 20 MCAD and ECAD vendors. Prior versions of this file format had limited support for thermal information. The new version, referred to as IDF 4.0, was releaded in 1998, will support material properties and compact thermal models.
Clemens Lasance provided a status report on European activites resulting from the original DELPHI and SEED projects. A major portion of his discussed centered on compact models and when it appropriate to use them. Specifically, he suggested that the compact model used be used if the detailed model exists; otherwise use the JEDEC measurements for different thermal metrics and construct a star-model from the measured values. The latter remark suggests a division of responsibilities between component manufacturers (supplying the measured values using standard boundary configurations) and customers (using these values to create the simplified model). Clemens then went on to suggest a list of modeling issues, indicating which ones could be subject to a standardization effort; this list was formalized agter the meeting and submitted subsequently.
Sarang Shidore presented An Update on Component Modeling Activities in Industry. His slide presentation covered the traditional approach to component thermal characterization for package comparisons, the development of CFD detailed models to predict package thermal behavior, and the DELPHI methodology that lead to the development of Compact Models. He pointed out that one major advantage of Compact Models supplied by component vendors is that the vendor does not have to reveal the internal details of the component, thus keeping internal component designs proprietary. Examples of how the subsequent SEED project used the detailed models to generate Compact Models for vendor internal and external consumption.
Bernie Siegal then lead a short question and answer session covering the information presented and a discussion of future activities for the thermal community. The next meeting may be held in conjunction with SEMICON/West in July 1999 in San Jose if suitable arrangements can be made in time.
Return to TEAmain
page