Creep deformation and stress relaxation of a martensitic P92 steel at 650 °C

Saber Khayatzadeh, David W J Tanner, Christopher E. Truman, Peter E J Flewitt, David J. Smith

Research output: Contribution to journalArticle (Academic Journal)peer-review

17 Citations (Scopus)
325 Downloads (Pure)


This paper develops methods to predict creep stress relaxation in the presence of combined boundary conditions and explores the influence of primary-secondary stress dependent creep properties on predictions for a martensitic P92 steel at temperature of 650 °C. A series of forward creep and elastic follow-up experiments have been conducted. A summary is provided of empirical creep equations for forward creep and creep stress relaxation (elastic follow-up) tests, including the link to the experimental procedure. The creep stress relaxation tests have been performed with two rigs to give elastic follow-up factors of 1.17 and 1.7. Both time hardening and strain hardening approaches have been considered where the strain hardening model provided more accurate predictions compared to time hardening; except at relatively low stress levels. The difference between stress relaxations predicted using strain hardening and time hardening approaches are considerable. The model predicts the creep stress relaxation accurately in the early stage of relaxation, indicating that the majority of stress relaxation occurs where primary creep needs to be taken into account. This study highlights the importance of stress dependent creep model to predict stress relaxation, especially with high level of initial residual stresses.
Original languageEnglish
Pages (from-to)57-71
Number of pages15
JournalEngineering Fracture Mechanics
Early online date27 Feb 2017
Publication statusPublished - 15 Apr 2017


  • Primary and secondary creep
  • Stress relaxation
  • Elastic follow-up
  • P92 steel
  • Creep modelling


Dive into the research topics of 'Creep deformation and stress relaxation of a martensitic P92 steel at 650 °C'. Together they form a unique fingerprint.

Cite this