where there is more than normal scatter, the curve from the specimen giving the lowest data is also shown in the figures. The Jid corresponding to the crack-initiation point (by the procedure in [2]) and J0.2 (corresponding to a crack-extension of 0.2 mm) estimated from the mean key-curve power-law are given in Table 1 for comparison with the estimates by the Schindler procedure. Figure 1 shows the key-curve results for the CW condition. The results from the PCVN specimens are close together and is well represented by the common fit curve shown (PCVN-JR curve). The pseudo-JR curve is much higher than the PCVN-JR curve and to bring the pseudo-JR curve into coincidence with the PCVN-JR curve, the pseudo-JR curve was shifted down using Q = − 4 (Eqn. 3). The Q-factor used is shown in Figure 1 by the side of the arrow indicating the shift. The JR curve obtained by this procedure is referred to as the shift-JR curve. 4.2. JR Curves by the Schindler Procedure The constants of the power-law fit obtained by the Schindler procedure as also the estimated J0.2 are given in Table 1. Application of the Schindler procedure to CVN specimens gives JR curves that are higher than those obtained by applying the Schindler procedure to PCVN specimens (Sch,PCVN in Figure 1). In the following, Schindler curve refers to the JR curve obtained by applying the Schindler procedure to (unprecracked) CVN specimen; this is almost in coincidence with or slightly lower than the key-curve PCVN-JR curve. Schindler curve is poor in reproducing the slope of the PCVN-JR curve. The shift-JR curve better reproduces the slope of the PCVN-JR curve. 4.3. General Discussion Results for ST, H5 and GT material conditions are similar to those in Figure 1, but with different Q values as shown in Table 1. Figure 2 for the H8 material shows anomalous behaviour. This material has been aged for 1000 h at 1073 K and is expected to have extensive precipitation along grain boundaries. This results in intergranular ductile fracture with a very low impact energy (~ 28 J: see Table 1) [5]. In such cases, bluntnotched CVN specimen reportedly shows lower toughness than the PCVN specimen [6]. Hence the anomalous behaviour in Figure 2: CVN JR curve is lower than the PCVN-JR curve. Also, the Schindler curves from CVN specimens are lower than those from the Schindler curves from PCVN specimens. Moreover, the key-curve JR curves from PCVN specimens show excessive scatter and odd behaviour. In this case, the pseudo-JR curve, without any shift (Q = 0), is ultra-conservative (lower) with respect to the PCVN-JR curves. Schindler has also stated that his procedure is applicable for CVN energies greater than 30 J [4]. From the above, it is evident that, when homogeneous deformation prevails and abnormalities like grain boundary fracture or other preferential fracture paths are not active, using a Q factor − 2 to − 4 for generating the shift-JR curve from the pseudo-JR curve gives conservative results. In the absence of the above mentioned abnormal deformation and fracture behaviour, the shift-JR curve is in coincidence with or slightly
RkJQdWJsaXNoZXIy MjM0NDE=