Development of a Reliable Model for Evaluating the Ductile Fracture Propagation Resistance for High Grade Steel Pipelines

The recent experience on ductile fracture propagation control on gas pipelines has shown that the applicability of the Battelle Two Curve Method (based on Charpy-V energy) to high grade steel pipes from API5L-X80 to X120 (ISO3183-L555M to L830M) operated at very high hoop stress values (≥500 MPa) is highly questionable.The reduced geometry of the specimen, the intrinsic low value of ductility of very high strength steels, as low work-hardening and low value of the strain at maximum load are pointed out as the main causes of the mismatch.Starting from these assumptions a new EPRG (European Pipeline Research Group) project has been launched with the aim to develop, with reference to the ductile fracture propagation resistance, a suitable fracture parameter(s) with an associated laboratory methodology based on a simple sample which would be able to take into account the role of the ductility of the material on this specific fracture event.The present paper shows the approach adopted in this EPRG Project: an innovative approach based on “plastic damage model” which allows to describe the stable ductile crack propagation by means of stress-state parameters (named triaxiality and deviatoric parameters).Moreover the proposed “damage model” has been implemented inside a commercial finite element code and used to predict the fracture crack propagation behaviour of Single Edge Notch Bend (SENB) tests in terms of load-displacement diagram and residual plastic deformation.One of the main topics of this project was the application of this method to six selected grade steels (with grades in the range of API X65 – X100) many of them coming from experimental full scale burst tests. The comparisons between experimental results and numerical simulations are substantially good; besides the results confirm that Charpy-V specimens, during the fracture propagation, work in different “constraint” conditions with respect to pipe and that DWTT specimen is in the middle between the two. Finally the “damage model approach” seems also able to discriminate between low and high grade steels in terms of failure deformation at rupture. So it resulted very promising to quantify the role of both ductile of the steel and geometrical constraint of the specimen in the ductile fracture propagation event.