PCB design improvement in the circuit between the north bridge and SDRAM through an integrated procedure

Authors:
Chih-Ming Hsu
Affiliations:
Department of Business Administration, Minghsin University of Science and Technology, 1 Hsin-Hsing Road, Hsin-Fong, Hsinchu 304, Taiwan, ROC
Venue:
Expert Systems with Applications: An International Journal
Year:
2010

Citing 3
Cited 1

Genetic Algorithms in Search, Optimization and Machine Learning

Genetic Algorithms in Search, Optimization and Machine Learning
Back-propagation neural network for the prediction of the short-term storm surge in Taichung harbor, Taiwan

Engineering Applications of Artificial Intelligence
Optimization of a multi-response problem in Taguchi's dynamic system

Computers and Industrial Engineering

Improving the lighting performance of a 3535 packaged hi-power LED using genetic programming, quality loss functions and particle swarm optimization

Applied Soft Computing

Quantified Score

Hi-index	12.05

Visualization

Abstract

Printed circuit boards (PCBs) are the primary base of electronic components, and they exist in almost all electronic devices. The PCB stackup design, physical rule settings, and electrical rule settings are the three main factors influencing the signal quality in a PCB. An improper PCB stackup design and physical and electrical rule settings will cause a high-speed signal out of the defined specification, and ultimately contribute to an unreliable device. Experienced engineers usually refer to the PCB design guides and chipset specifications recommended by the integrated circuit (IC) supplier, and determine the optimal PCB design parameter settings through trial and error. However, this approach cannot guarantee that parameter settings are truly optimal. To solve PCB design problems, this study proposes an integrated procedure based on a neural network, desirability functions, and genetic algorithms. A case involving PCB design improvements in the circuit from the north bridge (NB) to SDRAM in a personal computer (PC) is used to demonstrate the feasibility and effectiveness of the proposed optimization procedure. Results show that each quality characteristic very nearly approaches its perfect state. Furthermore, all of the crucial quality characteristics fulfill the required specifications and achieve a 100% yield from the six sigma quality viewpoint. Consequently, the proposed procedure can be considered an effective method of resolving multi-response parameter design problems.