In this paper, we present a statistical-neural network modeling approach to process optimization of fine pitch stencil printing for solder paste deposition on pads of printed circuit boards (PCB). The overall objective was to determine the optimum settings of the design parameters that would result in minimum solder paste height variation for the new board designs with 20-mil, 25-mil, and 50-mil pitch pad patterns. As a first step, a Taguchi orthogonal array, L27, was designed to capture the main effects of the six important printing machinery parameters and the PCBs pad conditions. Some of their interactions were also included. Fifty-four experimental runs (two per setting) were conducted. These data were then used to construct neural network models relating the desired quality characteristics to the input design parameters. Our modular approach was used to select the appropriate architecture for these models. These models in conjunction with the gradient descent algorithm enabled us to determine the optimum settings for minimum solder paste height variation. Confirming experiments on the production line validated the optimum settings predicted by the model. In addition to the combination of all the three pad patterns, i.e., 20, 25, and 50 mil pitch pads, we also built neural network models for individual and dual combinations of the three pad patterns. The simulations indicate different optimum settings for different pad pattern combinations.

1.
Hwang, J. S., 1994, “Solder/Screen Printing,” Surface Mount Technology, Mar., pp. 44–50.
2.
Morris
J. R.
, and
Wojcik
T.
,
1991
, “
Stencil Printing of Solder Paste for Fine-Pitch Surface Mount Assembly
,”
IEEE Trans. CHMT
, Vol.
14
, No.
3
, pp.
560
566
.
3.
Markstein, H. W., 1993, “SMT—Meeting Requirements of Fine Pitch Printing,” EP&P, Mar., pp. 34–39.
4.
Erdmann, G., 1991, “Improved Solder Paste Stenciling Technique,” Circuits Assembly, Feb., pp. 66–73.
5.
Xiao, M. C., Lawless, K. J., and Lee, N. C., 1994, “Prospects of Solder Paste Application in the Ultra Fine Pitch Era,” Journal of SMT, Apr., pp. 10–24.
6.
Anderson, M., Kirchner, K., and Moltz, E., 1993, “Development of X-Ray Laminography as a Process Control Tool,” Journal of SMT, Apr., pp. 2–16.
7.
Peace, G. S., 1992, Taguchi Methods: A Hands On Approach, Addison Wesley.
8.
Marwah, M., Li, Y., and Mahajan, R. L., 1996, “An Integrated Neural Network Modeling Approach and Its Applications in Electronic Manufacturing,” To appear, Journal of Electronic Manufacturing.
9.
Samuel
S. H.
, and
Zhang
H.
,
1994
, “
Artificial Neural Networks in Manufacturing: Concepts, Applications, and Prespectives
,”
IEEE Trans. CMPT
, Vol.
17
, No.
2
, pp.
212
228
.
10.
Li
Y.
,
Mahajan
R. L.
, and
Tong
J.
,
1994
, “
Design Factors and Their Effect on PCB assembly Yield-Statistical and Neural Network Predictive Models
,”
IEEE Trans. CPMT
, Vol.
17
, No.
2
, pp.
183
191
.
11.
Neubauer, C., and Hanke, R., 1993, “Improving X-Ray Inspection of Printed Circuit Board by Integration of Neural Network Classifiers,” Proc. IEEE Int. Electronics Manufacturing Technology Symp., pp. 14–19.
12.
Smith, M., 1993, Neural Networks for Statistical Modeling, Van Nostrand Reinhold, New York.
13.
Wasserman, P. D., 1989, Neural Computing: Theory and Practice, Van Nostrand Reinhold, New York.
14.
Nagata, S., Sekiguchi, M., and Asakawa, K., 1990, “Mobile Robot Control by a Structured Hierarchical Neural Network,” IEEE Control Systems Magazine, pp. 69–76.
This content is only available via PDF.
You do not currently have access to this content.