April 5-7, 2022|Santa Clara Convention Center| Santa Clara, CA
Speakers:
Yuanzhuo Liu (PhD student, Missouri University S&T)
DongHyun Kim (Assistant Professor, Missouri S&T)
Authors:
Yuandong Guo (Ph.D Student, EMC Laboratory, Missouri University of Science and Technology)
Chaofeng Li (Ph.D Student, Missouri University S&T)
Xiaoning Ye (Principal Engineer, Intel)
Location: Ballroom D
Date: Wednesday, April 6
Time: 9:00 am - 9:45 am
Track: 04. Advances in Materials & Processing for PCBs, Modules & Packages
Format: Technical Session
Theme : Data Centers, High-speed Communications
Education Level: All
Pass Type: 2-Day Pass, All Access Pass
Vault Recording: TBD
Audience Level: All
Reducing conductor losses due to surface roughness is a critical challenge for signal integrity in high-speed system design. A rough surface results in additional conductor loss and requires accurate modeling. The Hammerstad–Jensen model, Huray model, and a model based on hexagonal close-packing of equal spheres (HCPES) model are widely used for modeling surface roughness. These models rely on cross-sectional profile images to determine parameter values. However, the models use only limited information from cross-sectional images and assume a repetitive surface roughness profile across the entire surface of the copper foil. To utilize all available information within the cross-sectional image, an improved modeling approach using integration is proposed in this paper. Instead of assuming a constant total roughness level along the surface, a surface roughness correction factor is generated and modified by an integration method to account for all available copper foil profiles. This paper demonstrates that the proposed integration-based surface roughness modeling method can be combined with previous surface roughness models to improve the accuracy of conductor loss modeling.
Integral surface roughness modeling method for copper foils, accurate and time efficient surface roughness model