Physics of Failure Approach to Accelerated Testing of Assemblies

Objectives

 Illustrate a physics-of-failure (pof) method for conducting accelerated stress tests cost-effectively, to obtain maximum test-time compression, using simultaneous application of different stresses. Further, illustrate quantitative approaches for assessing damage metrics, acceleration transforms and stress margins from accelerated life testing data.

Background

Accelerated stress testing is used by industry for both life testing as well as screening. Unfortunately, there is a lot of confusion in the literature as to the most effective way to conduct accelerated tests with combination of different stresses to obtain the most test-time compression. Further, there appears to be a lack of scientific methods to extrapolate quantitatively from the test results to field durability estimates. In an effort to provide a pof-based solution to these issues, last year CALCE EPRC initiated a comprehensive multi-task, multi-year effort to explore the fundamental issues and arrive at practically useable answers. Three specific issues were identified, based on the interests of IAB members, for investigation during this phase:

Synergy between vibrational stresses and temperature cycling stresses for test time compression
Effectiveness of repetitive shock (RS) random vibration stresses vs. electrodynamic(ED) random vibration stresses for test time compression of fatigue failures.
Physics-of-failure approach for deriving quantitative damage metrics and acceleration transforms for these stresses.

 Year 1: Experimental phase to qualitatively compare the effectiveness of
 
 

Simultaneous vs. sequential application of combined temperature and vibration stresses, and
Repetitive shock (RS) vs. electro-dynamic (ED) random vibration test machines, for accelerated stress application

Year 3: Documentation phase to incorporate the findings of the first two phases into standards, software tools, and guidelines for design and testing.
The roadmap for the the multi-year project is outlined in Figure.1
 
 

Summary of First Year' s Results

 The task for the first year was to collect extensive experimental data from specimens containing different technologies, different architectures, and different manufacturing methods. A total of five specimen categories were identified, based on member interests (Updated specimen status). Four groups consisted of circuit card assemblies (CCAs), since this is the most effective package level for failure stimulation with exposure to accelerated stresses. The fifth group was a functional LRU box, to examine the capability to stimulate failures at higher packaging levels. Some of the specimens were daisy-chained engineering samples for testing substrates and interconnects, while the rest were fully functional assemblies. The schematic of the approach is shown in Figure 2
 
 

The tasks included:

design test matrix using vibration and temperature stresses,
select and acquire test specimens,
design and fabricate test setup/fixtures,
acquire, implement and calibrate control and data-processing hardware/software,
conduct accelerated stress tests,
collect and post-process data,
document results.

Methodology for Second Year

The damage analysis will be different for each observed failure mechanism. The failures identified during the first year of this project, in each of the five specimen categories, will be analyzed in detail. The CALCE failure mechanism libraries will be used, whenever appropriate. The model parameters will be updated using the accelerated test data, and quantitative acceleration transforms will be illustrated using the pof approach. New models will be generated,for new mechanisms for which models are not available in the CALCE library.

 The three units during this year will be allocated for stress analysis, damage analysis, and some new testing, as follows:
Unit 1: Analyze Hamilton Standard and Hamilton Standard CCAs.

• Conduct random vibration analysis for Hamilton Standard CCA on RS and ED machines. Conduct thermal stress analysis and superpose stresses.
• Conduct pof damage analysis using stresses computed above, to determine the answers to the three questions posed in this project.

Work Accomplished

Objective 1:Generic Guidelines for Accelerated Stress Testing
One of the deliverables of this multi-year project is to establish generic guidelines for conducting accelerated stress tests using a physics-of-failure (PoF) based approach. For further details please refer to the Internal report on Generic Guidelines for Accelerated Life testing using a PoF Approach.

Objective 2:Temperature-Vibration Synergy

First Internal Report

Second Internal Report