Here is the abstract you requested from the IMAPS_2011 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.
|AUTHOR WITHDREW 8-10-11: Stripline Surface Roughness Correction Factor Extraction|
|Keywords: Modeling and Characterization , Surface Roughness, Signal Integrity|
|AUTHOR WITHDREW 8-10-11: In order to have good quality of digital signals in a computer system, wires with low resistance are needed on the motherboard. Signal quality is more difficult to maintain for high-speed digital signals, because of the increase in the wire resistance proportional to the square root of frequency – a phenomenon called the skin effect. If signal quality is poor, the logic circuits will not be able to distinguish a binary ‘1' signal from a binary ‘0', causing malfunctioning of the computer system, such as the notorious blue screen in personal computers. For high-speed signals, the resistance of wires increases even faster than the square root of frequency. The increased skin-effect resistance is because of the surface roughness of the wires. This subtle effect can even double the loss. Current state of the art, however, is at the beginning stages of understanding the effects of surface roughness on wire resistance. Empirically derived formulas, such as the Hammerstad formula, are being used to approximately model the surface roughness. Simulation results also need to be verified against measurements of high-frequency behavior of wires. In this paper, a new methodology will be demonstrated to characterize the loss mechanism due to copper surface roughness based on scattering parameter measurements. Existing extraction methodologies are not accurate, because they are not capable of extracting the dielectric loss from conductor loss. The proposed methodology can for the first time completely eliminate the contribution of dielectric losses even for highly lossy dielectrics. Extraction procedure is based on measurements of two striplines with different widths.|
|Femi Akinwale, Graduate Student
San Diego State University
San Diego, CA