Chapter 53. Stress/Life Behavior of C/SIC Composites in a Low Partial Pressure of Oxygen Environment - Part III: Life Prediction Using Probabilistic Residual Strength Model

  1. Hua-Tay Lin and
  2. Mrityunjay Singh
  1. David J. Thomas1,
  2. Anthony M. Caiomino2 and
  3. Michael J. Verrilli2

Published Online: 26 MAR 2008

DOI: 10.1002/9780470294741.ch53

26th Annual Conference on Composites, Advanced Ceramics, Materials, and Structures: A: Ceramic Engineering and Science Proceedings, Volume 23, Issue 3

26th Annual Conference on Composites, Advanced Ceramics, Materials, and Structures: A: Ceramic Engineering and Science Proceedings, Volume 23, Issue 3

How to Cite

Thomas, D. J., Caiomino, A. M. and Verrilli, M. J. (2002) Stress/Life Behavior of C/SIC Composites in a Low Partial Pressure of Oxygen Environment - Part III: Life Prediction Using Probabilistic Residual Strength Model, in 26th Annual Conference on Composites, Advanced Ceramics, Materials, and Structures: A: Ceramic Engineering and Science Proceedings, Volume 23, Issue 3 (eds H.-T. Lin and M. Singh), John Wiley & Sons, Inc., Hoboken, NJ, USA. doi: 10.1002/9780470294741.ch53

Author Information

  1. 1

    Ohio Aerospace Institute Brook Park, Ohio 44142

  2. 2

    NASA Glenn Research Center Brook Park, Ohio 44135

Publication History

  1. Published Online: 26 MAR 2008
  2. Published Print: 1 JAN 2002

ISBN Information

Print ISBN: 9780470375785

Online ISBN: 9780470294741

SEARCH

Keywords:

  • C/SiC compposites;
  • ceramic matrix composites;
  • methodology;
  • extrapolation;
  • reliability

Summary

Life prediction of C/SiC composites under stress rupture loading conditions was modeled using a Probabilistic Residual Strength (PRS) technique. The model considered the material's initial static strength, time-to-failure, and intermediate residual strength to be random variables. Statistically significant strength and stress-rupture data sets were generated at 1200 °C for this material system as part of an in-house testing program. These data sets were used for the calibration and validation of the PRS model. Predictive features of the model, such as the generation of constant reliability design curves, are presented; and other key benefits of this lifing approach are discussed.