Modeling & Monitoring of Product DPPM with Multiple Fail Modes

Author

Anderson, Thomas J. ; Carulli, John M., Jr.

Author_Institution

Texas Instrum., Inc., Dallas, TX

fYear

2006

fDate

26-30 March 2006

Firstpage

545

Lastpage

551

Abstract

A method to model and de-rate product DPPM using existing production EFR (early failure rate) and burn-in data is proposed. This approach incorporates multiple classes of reliability fail modes. Changes in process, test, or design may then be employed to optimize product outgoing defect levels. These changes will be based on the defect acceleration kinetic parameters, instead of just the reliability fail-fraction or wafer sort yield. Monitoring of low part-per-million defects is also critical in maintaining general quality and a new method is discussed

Keywords

Weibull distribution; failure analysis; production testing; remaining life assessment; Weibull distribution; acceleration parameters; burn-in data; conditional reliability; de-rate product DPPM; defect acceleration kinetic parameters; early failure rate; existing production EFR; multiple defects; multiple fail modes; optimize product; product defective part per million modeling; product defective part per million monitoring; reliability fail modes; reliability fail-fraction; wafer sort yield; Acceleration; Condition monitoring; Geophysical measurement techniques; Ground penetrating radar; Instruments; Kinetic theory; Production; Read-write memory; Stress; Testing; Burn-in; Weibull; acceleration parameters; conditional reliability; de-rated; extrinsic; multiple defects;

fLanguage

English

Publisher

ieee

Conference_Titel

Reliability Physics Symposium Proceedings, 2006. 44th Annual., IEEE International

Conference_Location

San Jose, CA

Print_ISBN

0-7803-9498-4

Electronic_ISBN

0-7803-9499-2

Type

conf

DOI

10.1109/RELPHY.2006.251277

Filename

4017218