Optimal low-frequency noise criteria used as a reliability test for BJTs and experimental results

Two basic problems in the application of noise criteria to reliability testing are considered. First, the need to look at all noise mechanisms rather than noise at a specific frequency is emphasized, and noise criteria using both 1 f noise and g− r noise are established. Second, the physical mechani...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Microelectronics and reliability 1991, Vol.31 (1), p.75-78
1. Verfasser: Dai, Yisong
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Two basic problems in the application of noise criteria to reliability testing are considered. First, the need to look at all noise mechanisms rather than noise at a specific frequency is emphasized, and noise criteria using both 1 f noise and g− r noise are established. Second, the physical mechanisms which explain the low noise level of some failed devices are discussed, and the optimal noise threshold levels are found for minimum error during reliability testing. Reliability testing results for 1000 BJTs (3DG6) show that the failure rate of devices which initially exhibit high noise is about 2–3 times higher than for devices which initially have low noise. This means that the life expectancy of BJTs with a low noise level (using optimal noise criteria) is increased by a factor of 1.5–2.5; thus, noise measurements could be a useful reliability screen classification method for BJTs.
ISSN:0026-2714
1872-941X
DOI:10.1016/0026-2714(91)90350-G