Lifetime Prediction

Lifetime & Reliability Considerations during Product Development

One essential goal of the design process of an electronic product is to create a robust and reliable design that meets the demands from field operation. Typical measurement values, the number of failures per operation time and their distribution are evaluated. By the use of accelerated aging tests together with numerical simulations, it is possible to get access to these quantities and to predict the lifetime of a product for different load cases.

 

A typical lifetime prediction analysis involves the following steps:

We support our customers during each step in order to avoid over-engineering, increase power density and enhance the competitiveness of your products.

Application and Research

  • Development of lifetime estimation strategies (experimental and theoretical).
  • Statistical methods according to industrial standards and guidelines.
  • Numerical and analytical modeling of physical degradation mechanisms.
  • Objectives: Electronic systems, subsystems and components (e.g. converter, power modules, semiconductors, capacitors, inductors, etc.).

Example: Lifetime Analysis of Electronic Power Module

1 - Mission Profile  

 

  • Use cases, e.g. drive cycles
  • Operating data, e.g. V and I vs. time

 

2 - Power Loss Profile

  • Electrical simulations
  • Empirical mathematical relations

 

3 - Temperature Profile

  • 3D-transient thermal Finite-Element-Analysis (FEA)
  • 1D-modeling by electro-thermal analogue circuits

 

4 - Load Analysis

  • Constant, variable, block and random loads
  • Extraction of closed hysteresis loops from load signal by use of counting algorithms as input for lifetime model

 

5 - Lifetime Model

  • Determination of system response and damage variables
  • Accelerated ageing tests (e.g. power cycling, thermal cycling, shock and storage tests, etc.)
  • Non-linear transient finite-element-analysis
  • Modeling of damage evolution
  • Empirical, by observable variables, e.g. temperature swing, heating voltage, thermal  resistance
  • Physical, by internal variables, e.g. inelastic strain, plastic strain energy density, damage variables (CDM)

 

6 - Prediction of Lifetime

  • Accumulation of damage per load increment
  • Calculation of predicted lifetime
 

Lifetime Prediction for Power Electronic Assemblies

[ PDF 0.33 MB]

Related Research Areas at IISB

Explore the Entire Power Electronic Systems Value Chain