Title :
PCB Testing Using Infrared Thermal Signatures
Author :
Moldovan, Horatiu ; Marcu, Marius ; Vladutiu, Mircea
Author_Institution :
Dept. of Comput. & Software Eng., "Politehnica" Univ. of Timisoara, Parvan Timisoara
Abstract :
A printed circuit board (PCB) testing method using infrared thermal signatures is presented. The concept of thermal signature for PCBs is introduced. Based on this concept, the testing method is able to classify the integrated circuits (ICs) on a PCB into a number of classes (e.g. functional - fault free, non-functional - faulty and less reliable - functional circuits with high current consumption). According with thermal signature of each IC on the PCB, the PCBs can be also classified in the same number of classes. The classification system is a feed-forward neural network that learns and classifies the information achieved from the infrared image
Keywords :
circuit reliability; fault trees; feedforward neural nets; infrared imaging; printed circuit testing; PCB testing; classification system; fault free; feed-forward neural network; functional circuits; infrared imaging; infrared thermal signatures; integrated circuits; Circuit faults; Circuit testing; Infrared imaging; Inspection; Integrated circuit modeling; Integrated circuit testing; Logic circuits; Logic testing; Neural networks; Printed circuits; neural network; printed circuit board; thermal image; thermal signature; thermal testing;
Conference_Titel :
Instrumentation and Measurement Technology Conference, 2005. IMTC 2005. Proceedings of the IEEE
Conference_Location :
Ottawa, Ont.
Print_ISBN :
0-7803-8879-8
DOI :
10.1109/IMTC.2005.1604516
