Session: MF-09-01 Mechanistic Modelling of Deformation and Fracture

Paper Number: 155870

155870 - A Procedure to Estimate the Effect of Constraint on Temperature Shifts to the Reference Toughness Curve for Low Toughness Line Pipe Seam Welds 

Abstract: 

This work was conducted for the assessment of fracture resistance of lower toughness vintage line-pipe steel welds during a PRCI project to address DOT/PHMSA MegaRule toughness requirements. In a review of vintage linepipe databases of Charpy data, high-constraint fracture toughness data, and full-scale data, a procedure was developed to determine the lowest toughness for ductile fracture initiation of vintage line-pipe base metal and welds. In a few of the weld cases, the procedure predicted that the ductile fracture initiation transition temperature was warmer than the minimum operating temperature. Hence a procedure like the ASME Reference Toughness curve (or Master Curve) was needed to determine the toughness. However, the Charpy upper-shelf toughness can be so low that the nil-ductility reference temperature (T_RT-NDT) or the Master Curve T_o value could not be determined. The full-size Charpy upper shelf values ranged from 4 to 25 ft-lb. Therefore, a new Material-Specific Reference Toughness curve needed to be established, with a different way of defining a Reference Temperature (T_ref). The shape of this new Material-Specific Reference Toughness curve was consistent with the past ASME and prior toughness data, and bounded them as well as C(T) data on these lower CVP steels. The key concept used was that the equation for the Reference Toughness curve was the boundary between cleavage fracture initiation and ductile tearing, which is similar to Kirk’s T_us concept for the Master Curve for nuclear vessel applications.

The prediction of the lowest temperature for ductile initiation was previously developed by use of hundreds of pipe fracture tests and thousands of specimen tests, and validated against many additional full-scale pipe fracture tests afterwards. This Master-Curves of Fracture Transition Temperatures procedure allowed the use of Charpy transition temperature data to be correlated to the transition temperature of through-wall-cracked or surface-cracked pipes of different thicknesses. The through-wall-cracked transition temperature also correlated well to high-constraint fracture mechanics test specimens [i.e., SEN(B)]. As will be shown, the Charpy transition temperature correlation predicted the high-constraint C(T) specimen fracture toughness transition temperature as well, allowing for the determination of a new T_ref, which is a link from the traditional Master Curve to predicting through-wall-cracked and surface-cracked pipe transition temperature differences. Together the modified Master Curve and the Master-Curves of Fracture Transition Temperatures procedure created a powerful pragmatic tool.

With the fact that the high-constraint C(T) specimen initiation toughness corresponds to the surface crack having an a/t of 0.7 and knowing that on the upper-shelf the surface-cracked pipe toughness changes with a/t allows to constraint adjustments. Shallower surface cracks have higher upper-shelf toughness values, so that adds a shift to the Material-Specific Reference Toughness curve. Additionally surface cracks have a lower transition temperature than high-constraint specimens, so that is another transition temperature shift. If the surface crack is shallower than a/t=0.375, then there is an additional lowering of the brittle-to-ductile transition temperature, or further shift to the Material-Specific Reference Toughness curve.

The correlations, validations, and example cases for different types of welds will be shown.

Presenting Author: Gery Wilkowski Engineering Mechanics Corporation of Columbus

Presenting Author Biography: To be given later.

Authors:

Gery M. Wilkowski Engineering Mechanics Corporation of Columbus