Creep crack growth behaviour of Type 316H steels and proposed modifications to standard testing and analysis methods

The influence of test duration and geometry on the creep crack initiation and growth behaviour of 316H steel

“…Generally this means that for ductile materials such as austenitic Type 316H stainless steel, the load applied to the specimen generates a reference stress in excess of the yield stress of the material, which implies that significant plasticity is introduced into the test specimens on loading. Recently, long term tests (of duration exceeding 16,000 hrs) have been performed on two compact tension, C(T), specimens [1] and on a DEN(T) specimen of Type 316H stainless steel at 550 °C. The C(T) specimen, which is subjected to combined bending and tension loading is generally considered to be a high constraint geometry in comparison to specimens under predominantly tensile loading such as the M(T) and DEN(T) whereas the SEN(B) and CS(T) have intermediate levels of constraint.…”

Effects of plastic pre-straining level on the creep deformation, crack initiation and growth behaviour of 316H stainless steel

“…It can be observed that the cracking mode in this test was, as termed in [21], 'brittle inter-granular', i.e. where the CCI and growth occurs with no clear evidence of grain deformation.…”

“…Additional validity criteria are also proposed in the R5 procedure [20] and in [21] which suggest that the non-dimensional crack velocity, λ, must be less than 0.5 and the engineering redistribution time, eng red t , must be…”

“…It has been discussed in [21] that the engineering redistribution time, compared to the transition time estimated using Eqn (10), is a more accurate criterion to check widespread creep conditions in CCG tests on 316H stainless steel. The λ and eng red t for these additional validity criteria can be calculated using…”

“…Also included in Figure 10 are the mean (solid line) and upper bound (UB) / lower bound (LB) (dashed lines) trend lines fitted to the short term (shown in red) and long term (shown in pink) AR data [21] . The mean and UB/LB fits to the 8% PC data set are also included in this figure.…”