TWI Industrial Member Report Summary 893/2008
By L Wei and I Hadley
Background
The European FITNET consortium was formed in 2002 with the aim of preparing the basis of a European fitness-for-service (FFS) procedure, covering assessment of flaws in welded and non-welded metallic structures. The procedure, which has now been submitted to CEN for publication as a CEN document, is intended for use by all industry sectors, and covers all relevant failure/damage modes (fracture, fatigue, creep and corrosion).
As part of its contribution to the validation of the procedure, TWI examined the results of around 300 large-scale and full-scale fracture tests. The test behaviour was analysed mainly in terms of existing FFS procedures such as BS 7910 (BSI, 2005), which has many elements in common with FITNET. The results of this validation exercise are reported in full in Hadley and Moore (2006a). Selected cases are also included in the FITNET validation document (Volume III/Section 13 of the FITNET procedure) and in a paper presented at the final FITNET conference (Hadley and Moore, 2006b).
The FITNET fracture assessment procedure includes a hierarchy of methods, designated Options 0 to 5. In general, the assessment becomes less conservative and more accurate as the option number increases. To date, TWI's validation work has considered only Options 0, 1, 3 and 5 of the FITNET fracture assessment procedures, which are closely analogous to various levels of fracture assessment by BS 7910. Two particular simplifying assumptions lie behind all fracture validation considered in the previous TWI work:
- Handbook solutions can be used for the determination of the applied stress intensity (KI) and reference stress (?ref) associated with a cracked body.
- The material has homogenous tensile properties, ie there is no effect of weld metal strength mismatch on the assessment.
The validation exercise revealed that there were several cases that would be amenable to a more detailed analysis, ie where the above assumptions would not be necessary. For example, several cases were identified where FITNET Option 4 could be applied. This requires a numerical analysis of the cracked body, from which a direct calculation of the crack driving force, J, is derived, and compared with the lower bound fracture toughness, Jmat, of the material. Other cases were found where FITNET Option 2 (which allows weld metal strength mismatch to be incorporated into the analysis) could be applied; these are the subject of a separate report within this project.
Objectives
- Re-analyse the results of selected full-scale fracture tests using Option 4 of the FITNET fracture analysis procedure.
- Compare the results with those produced by other options and by BS 7910.