Estimation of Micro-drilled Hole Wall of PWBs Using Data-mining

Abstract

Recently, electricity and an electronic equipment which centers on portable terminals such as cellular phones and notebook computers have been rapidly widespread. The through hole processing by micro diameter drill to connect the conductor layer of the printed wiring board is an especially important process for forming the circuit. After the drilling, plating is done on each through hole wall to connect conductor layers. The plating thickness becomes not uniform due to the through hole wall roughness, and so there is a possibility of disconnection of plating by thermal shock. Therefore, to improve the reliability of an electric connection, the improvement of the through hole wall quality is desired. However, it has been considered that a lot of factors influence on the quality of the through hole wall. Then, the purpose of this report is to apply data mining that is the method of statistically analyzing the factor to the image processing data, and to elucidata the factors which influence on the deterioration in the through hole wall roughness. In addition, we take the Motif method developed by Keogh in order to express through hole wall quality more accurately. Following conclusions were obtained : (1) The cutting distance, drill temperature, and width of the fiber bundle of yean influence on the deterioration in the thorough hole wall roughness. (2) Number of “rough” Motif and the size of Motif influence on the stability of thickness of plating.