Robert F. Kwasnick

Determination of CPU use conditions

Use condition inputs to physics-of-failure models are required to use knowledge-based qualification of ICs. Modern CPUs have multiple voltage-frequency states which vary widely in reliability stress, but it is not obvious what time in the various states to use in product qualification. We present a methodology for developing a time in state model for CPUs which combines large scale user monitoring and lab-based studies. Results for a specific CPU family, including field validation and implications for knowledge-based qualification, are discussed.

Impact of VLSI technology scaling on HTOL

High Temperature Operating Life (HTOL) is a standard stress used in IC product qualification. With VLSI technology scaling, gate dielectric TDDB models have higher acceleration factors leading to an increase in predicted HTOL failure, particularly with the transition to high-k gate dielectrics. However, cumulative end-of-life field failures remain substantially unchanged from previous technologies. A calculator tool which comprehends both field and HTOL failure modeling illustrates the trend and guides product qualification expectations.

Ambient Use-Condition Models for Reliability Assessment

We describe methods of computing reliability acceleration for realistic temperature/humidity use-condition models. We extract outdoor temperature/humidity models based on NOAA data, and indoor and automotive interior models by using the NOAA data combined with additional data which characterizes thermal and human behavior (thermostat settings). We compare predictions of these models with traditional reliability assessments, and provide useful models to represent the results

Define electrical packing temperature cycling requirement with field measured user behavior data

In this paper, a user behavior based solder joint reliability modeling approach has been proposed to estimate the design and test requirements for the second level interconnect (SLI) reliability prediction. This approach uses a numerical tool to integrate solder joint creep damage during the actual use condition that was collected from a large user sample size. The resultant damage per time period was then input to the solder joint fatigue model to estimate equivalent damage to testing duration. The is a physics based approach and is expected to provide more accurate product life prediction and reliability performance demand for BGA package designs.

Setting use conditions for reliability modeling

Use conditions (UCs) are necessary inputs for knowledge-based qualification reliability modeling calculations, and are implied by standard-based qualification stress conditions. We describe a method for setting CPU UCs. A sampled distribution of field data is analyzed, accounting for the reliability model Weibull shape factor. That is combined with other information to determine a reference value for modeling, with appropriate conservatism. We present application to notebook PCs for both silicon and thermo-mechanical CPU wearout mechanisms.

Time-ordered events CPU reliability assessment

IC product use conditions (UCs) are needed to enable accurate reliability modeling in the context of knowledge-based qualification. We describe a method of use condition development which is based on the sequence of user foreground events from field surveys. Events are converted to temperature and voltage use condition traces accounting for lab data on representative workloads and thermal modeling. The temperature and voltage data are paired for client CPU UC data and used to estimate the reliability risks for both silicon and thermo-mechanical failure mechanisms. The new approach are validated using field consumer data. We also describe the future work needed on how to account for concurrent events and the implications on TOE methodology.

In Vehicle Infotainment and Advanced Driver Assistance Systems: Advantages of Knowledge-Based Qualification over Standard-Based Qualification for Solder Joint Reliability

In this paper, we proposed a novel approach of applying Knowledge-based Qualification (KBQ) to calculate temperature cycle (TC) requirements for Ball Grid Array (BGA)s in a given automotive electronics module. Realistic use (UC) condition temperature profiles were constructed by combining weather data, driver cabin environment, user behavior statistical data, and component thermal behavior. Individual package geometry, solder ball pin map and material properties were incorporated through Finite Element Analysis (FEA) modeling. Direct translation from use condition to KBQ test condition requirement was achieved by using a physics based damage metric - inelastic strain energy density (ISED). We compared KBQ results to Standard-based Qualification (SBQ) results for In Vehicle Infotainment (IVI) and Advanced Driver Assistance Systems (ADAS) automotive modules to demonstrate and explain why KBQ and SBQ differ and why SBQ requirements can be misleading. We also demonstrate how package geometry and material properties impact KBQ, while they are generally ignored by SBQ.

Platform qualification methodology: Face recognition

Platforms may be developed with a range of new features. We describe a methodology to qualify a platform for operational stability and functional reliability. This includes determining feature-specific goals and use conditions to achieve the desired user experience quality and reliability. We present detailed results for a client PC face recognition feature.

Telemetry for reliability

Knowledge of customer usage of IC products is an important part of establishing accurate use conditions inputs for product reliability modeling. We present a multi-stage framework for achieving this: derive telemetry metrics; develop telemetry software; acquire users; manage and store telemetry data; perform data analytics and visualization; decide use conditions. We discuss each stages objectives and requirements, and provide information relevant to successful implementation. Server telemetry examples are presented to elucidate the framework and its role in evaluating product reliability.