Multiresolution sensor fusion approach to PCB fault detection and isolation

This paper describes a novel approach to printed circuit board (PCB) testing that fuses the products of individual, non-traditional sensors to draw conclusions regarding overall PCB health and performance. This approach supplements existing parametric test capabilities with the inclusion of sensors for electromagnetic emissions, laser Doppler vibrometry, off-gassing and material parameters, and X-ray and Terahertz spectral images of the PCB. This approach lends itself to the detection and prediction of entire classes of anomalies, degraded performance, and failures that are not detectable using current automatic test equipment (ATE) or other test devices performing end-to-end diagnostic testing of individual signal parameters. This greater performance comes with a smaller price tag in terms of non-recurring development and recurring maintenance costs over currently existing test program sets. The complexities of interfacing diverse and unique sensor technologies with the PCB are discussed from both the hardware and software perspective. Issues pertaining to creating a whole-PCB interface, not just at the card-edge connectors, are addressed. In addition, we discuss methods of integrating and interpreting the unique software inputs obtained from the various sensors to determine the existence of anomalies that may be indicative of existing or pending failures within the PCB. Indications of how these new sensor technologies may comprise future test systems, as well as their retrofit into existing test systems, will also be provided.