This paper describes a neural network approach to detecting solder defects on printed circuit boards when using thermal signatures. Solder defects such as open and insufficient solder was investigated. A multi-layer neural network with multiple inputs and a single output was utilized. A back-propagation algorithm was utilized within the network. Computer mouse printed circuit boards with known introduced solder defects and amounts of solder were used for experiments. Thermal images were acquired as the boards were powered up. A Visual Basic program was written to retrieve temperature data from an encoded image file format. Afterwards, MATLAB neural network routines were applied to analyze the data. The neural network was able to diagnose solder defects on two of five resistors with 91.1% accuracy, and on three of five resistors with 61.1% accuracy.

The use of the photoacoustic technique to monitor the thermal properties of materials that can be obtained only as parts of multicomponent samples is illustrated by performing the thermal characterization of two porous materials: porous silicon obtained from n-type crystalline silicon through the spark process and that obtained through the electrochemical etching method. This nonseparative, and hence nondestructive, approach makes use of an effective thermal diffusivity treatment based on the analogy between thermal and electrical resistances, in combination with simplified compositional models for the corresponding multicomponent systems. The thermal parameters obtained are in agreement with existent studies concerning the composition of these materials. This approach offers the possibility of performing the thermal characterization of other porous semiconductors and analogous materials. © 1997 Society of Photo-Optical Instrumentation Engineers.Subject terms: photoacoustic and photothermal science and engineering; photoacoustics; porous materials; thermal properties.Paper PPS-01