| 摘要 |
Detecting failed integrated circuit among integrated circuits, by (a) assuming that all integrated circuits under test define under-test set, and testing each one of the integrated circuits in the under-test set, (b) removing integrated circuits judged to be in failure in step (a) from the under-test set, (c) measuring spectrum of a current supplied from a power source into each one of integrated circuits in under-test set, (d) calculating both mean value and standard deviation of spectrum for under-test set, (e) judging whether an integrated circuit is in failure or not, based on both the mean value and the standard deviation of the spectrum, (f) removing integrated circuits having been judged to be in failure in step (e), from the under-test set, and (g) judging under-test set to be in no failure. Thus, it possible to find failed integrated circuits without preparing data of integrated circuit in no failure, as a reference.
|
