Head of fracture, The Welding Institute, Abington, Cambridge
Pages: 29 Published: Jan 1988
Although crack-tip opening displacement (CTOD) analysis procedures are well established for the prediction of tolerable crack sizes in ferritic steels, no specific procedures exist which incorporate a resistance (R) curve to facilitate estimation of the ductile instability conditions for a fully ductile material.
This paper discusses a modification to the CTOD procedure, based on the strip yield model, to enable the prediction of maximum nominal stress and the extent of stable ductile crack growth, using a CTOD R-curve as input.
The effectiveness of this procedure is assessed by predicting the behavior of wide plate tension tests and burst tests in API 5LX56 and A533B Class I ferritic parent steels and of wide plates in stainless steel plate and weldments. The predictions are then compared with physical data obtained from the tests.
Reasonable predictions of maximum applied stress are achieved. However, the accuracy is very dependent on the plastic collapse estimations. This is particularly true for the high work-hardening stainless steel plate where, to avoid an overestimate of collapse, the effective flow stress must be taken to be the stress at 3% strain in a uniaxial tension test.
Predictions of the test results are also made using maximum load toughness values, measured using CTOD test specimens complying with BS 5762, with either the CTOD design curve or the modified strip yield model approach. Predictions give safe estimates of nominal stress.
It is concluded that maximum load values provide the most practical general approach when detailed resistance curve information is not available.
elastoplastic analysis, fracture toughness, fracture mechanics, critical values, ductile fracture, low-alloy steels, austenitic stainless steels, welded joints, pressure tests, static fracture tests
Paper ID: STP23285S