Influence of Reloading on Stress Relaxation for 19Cr-9Ni-1.4Mo-1.4W-Nb Bolting Steel

Abstract

Empirical equations to describe reloading stress relaxation data were proposed in the previous papers⁸⁾⁹⁾ on both ferritic and martensitic bolting steels. In this paper, application of the equations proposed was discussed to austenitic 19Cr-9Ni-1.4Mo-1.4W-Nb bolting steel which shows a peculiar stress relaxation behavior caused by structural changes during test¹⁰⁾¹¹⁾.
Reloading stress relaxation tests on the steel were carried out at 650°C and at the initial stress (σ₀) of 29.3kgf/mm² which corresponds to the value of stress for the total strain of 0.20 percent. Repeated loadings were undertaken up to forty-five times at specific reloading time intervals (t_i=24, 72, 240 and 720 hours), and at specific residual stresses (σ_rj=0.9σ₀, 0.8σ₀, 0.7σ₀, 0.6σ₀, 0.5σ₀, 0.46σ₀kgf/mm²).
It was confirmed that the reloading stress relaxation curves of the present steel could not be represented by the creep constitutive equations as mentioned in the previous papers. The reloading stress relaxation data obtained for the steel showed a decrease in relaxation strength with increasing loading cycles except for the earlier numbers of loadings. Moreover, the specimens with specific residual stresses lower than 0.7σ₀ resulted in failure. Their fracture mode was of intergranular type caused by linkage of grain boundary cavities.
From examinations of accumulated plastic strain data, it was confirmed that slightly modified equations to describe the reloading stress relaxation data were applicable to the present steel. The modified equation for specific reloading time intervals data is
log(σ₀-σ_ri)=a₀+a₁logt_i+a₂(logt_i)²+a₃logN
For specific residual stresses data, the expression is the same as the previous one:
logt_j=b₀+b₁log(σ₀-σ_rj)+b₂{log(σ₀-σ_rj)}²+b₃logN
where σ_ri is the residual stress, t_j is the testing time, N is the number of loadings, and a₀, a₁, a₂, a₃, b₀, b₁, b₂, b₃ are constants.