Assuming a power-law relation, JR is given by the following relation: J = C (∆a)p (1) where C and p are fit constants (C and p from CVN specimens are indicated as pseudo). For constant slope, dJ/da must be constant. Therefore, dJ/da = C . p. (∆a)p-1 = Constant (2) This can be true only if the new JR curve is represented by: J = C (∆a)p + K (= Q.p) (3) where numerical constants K and Q are determined empirically by comparing the pseudo-JR curves with the PCVN-JR curves. Because of plasticity and notch-root effects, pseudo-JR curves are much higher than the PCVN-JR curves. Hence, for obtaining the PCVN-JR curve from the pseudo-JR curve, a negative shift must be applied to the pseudo-JR curve and Q is negative (see, however, Section 4.3). 3.2. Schindler's Procedure for Obtaining JR Curves In recent analyses, Schindler uses only the power-law (Eqn. 1) for estimating the JR curve as is done in the present paper. Schindler and coworkers [4] obtain constants C and p of the power-law (Eqn.1) from the following relations: C = (2/p)p . [ η(a 0)/{B.(b0) 1+p}]. Et p. Emp 1-p (4) p = (3/4). [1 + Emp/Et] -1 (5) Emp is the plastic energy upto Pmax (maximum load), Et is the total energy for the impact test and η(a0) is the well known eta-factor. We have used η(a0) given in [2]. 4. RESULTS AND DISCUSSION 4.1. Key-curve JR Curves and Application of the Shift Procedure The power-law constants for the key-curve JR curves are given in Table 1. Only the constants for the mean curves from the multiple specimens are given (separate fits for CVN and PCVN results in each heat-treatment condition). In most cases, the specimen to specimen scatter is small enough to justify this procedure. In making the fit, for each specimen, the maximum ∆a was chosen to be equal to be 10% of b0, the initial remaining ligament depth (= W − a0, where a0 is the initial a). The ASTM size criteria have not been evaluated since only the results from same size specimens are being compared. In cases
RkJQdWJsaXNoZXIy MjM0NDE=