Reliability Prediction Models for Microcircuits Boeing Company

Format:

Conference/Event

Author/Creator:

Porter, Porter, author.

Contributor:

Finke, W. A.

Conference Name:

9th Reliability and Maintainability Conference (1970-07-20 : Detroit, Michigan, United States)

Language:

English

Physical Description:

1 online resource

Place of Publication:

Warrendale, PA SAE International 1970

Summary:

Some of the common methods of reliability prediction utilizingcalendar or lot size dependence and various "k" factorsfor complexity, end use, product family, and environment arebriefly reviewed. These techniques, although sometimes accurate andsimple, do not provide adequate reliability tradeoff informationand do not fully treat the effects of vendor variability andimprovements in the state-of-the-art. From these techniques it isfrequently difficult to understand causes of observed failure ratesand to determine what can be done to achieve cost-effectivereliability.One method of alleviating these objections is to derive a modelbased on microcircuit failure mode and mechanism knowledge. Thecurrent status of microcircuit failure knowledge is reviewed.Considerable use is made of data which has been acquired at veryhigh stress levels and the relationship between this data and enduse data is discussed.Microcircuit failures can be grouped into four essentiallyindependent categories: failures due to time degradation, failuresdue to mechanical stress, serial failures due to two or more eventsoccurring in succession, and parts which never worked. A failureprediction model based on failure modes and mechanism knowledgewould contain each of these elements. The resulting model mightprove unwieldy and of dubious value because of its size. However, asimplified model can be employed. If the user knows how it wasderived, it can be a useful and flexible tool. One form that such amodel might take is presented and discussed

Notes:

Vendor supplied data

Publisher Number:

700645

Access Restriction:

Restricted for use by site license