Integrated approach to reliability-based design of future electronics systems
暂无分享,去创建一个
Over the past 5 years, VEXTEC has developed a modeling tool set for analyzing variability at every product level. This is from the lowest level where physics of failure describes the process for fatigue crack initiation and development within the material microstructure to the complex electronic interconnect, the electronic control unit (ECU), and finally to the overall vehicle system of ECUs. The oversimplification that every element within the system of ECUs has a constant failure rate is not assumed. In fact, no assumptions are made about failure rates. Physics of failure models are used to predict failure rates based on measured statistical variations. With the advent of the computer, the understanding of the physics of failure and how to apply it directly to electronics design has greatly improved. The vast array of computer aided design and manufacturing tools now in existence has enabled the simulation of physics for thermodynamics, fluid flow and structural analysis for example. Statistical measurements of manufacturing process control determine the statistical variations in the geometric dimensions and product material properties. Warranty records, customer surveys and built-in diagnostics are used to determine the statistical variation is how the product is used. These statistical distributions are combined with the physics of failure within VEXTEC's virtual representation of the electronics product to simulate the product's real world performance. Thus, reliability can be estimated while the product is still on the drawing board. This paper will discuss the various aspects that make up this new electronics design framework. The validation work that conveys the realities of this framework will be discussed as well as the future plans for implementation of this concept within everyday automotive design
[1] John H. L. Pang,et al. Low cycle fatigue models for lead-free solders , 2004 .
[2] T. Cruse,et al. A reliability-based model to predict scatter in fatigue crack nucleation life , 1998 .