Session: MF-12-01 Leak Before Break

Paper Number: 107396

107396 - Estimation Scheme for Weld Residual Stress Effect on Crack Opening Displacements 

Leak before Break (LBB) concepts have been used in many industries (nuclear, space, oil, and gas, etc.) over the years to help ensure that pressurized pipes, vessels, and other components operate safely.  With LBB a pressurized component will leak, and the leak must be detected so that the component can be depressurized, and the flaw repaired, or pipe or vessel replaced before a catastrophic break occurs.  The nuclear industry has been one of the leaders in LBB technology development over the years.  LBB is used in nuclear plants to simplify the plant design to ensure jet impingement shields, pipe whip constraints, and other safety features are not required.

The Standard Review Plan in NUREG-0800 was developed by the U.S. Nuclear Regulatory Commission and provides LBB evaluation procedures to exclude the dynamic effects of postulated pipe ruptures.  In particular, the Standard Review Plan (SRP) Section 3.6.3 allows analyses to eliminate postulated pipe ruptures from the design basis of plant SSCs.  Part of this analysis plan is to determine the leak rate through postulated cracks that are smaller than the critical size with safety factors applied on both loads and crack size.  The leak rate codes require inputs of the crack length and the crack opening displacement (COD), the separation or opening between crack faces, to predict the leak rate.

Analytical methods for estimating COD from operating loads have been developed to eliminate the necessity of performing finite element (FE) solutions for each individual geometry and loading condition on a case by case basis.  However, cracks in piping often occur in welds or in the heat affected zone of welds, and weld residual stresses can also affect the COD.  The effect of the WRS and COD is usually ignored in LBB calculations mainly because no adequate estimation method has been available to predict this effect to date.  This paper examines various analytical methodologies for estimating WRS-induced COD and provides an estimation scheme similar to the classic GE/EPRI method that can be used to predict the effect of WRS on COD. 

Validation examples are provided where COD estimates are compared directly to full WRS solutions with a crack added, for a large number of cases.  This is followed by a summary of the results and suggestions for possible additional work.

Presenting Author: Frederick (bud) Brust emc-sq

Presenting Author Biography: F. W. Brust is experienced in computational modeling.

Authors:

Frederick (bud) Brust emc-sq
Edward Punch Engineering Mechanics Corporation of Columbus
Elizabeth Twombly Engineering Mechanics Corporation of Columbus
Jay Wallace US Nuclear Regulatory Commission