United States Nuclear Regulatory Commission - Protecting People and the Environment

Nondestructive and Destructive Examination Studies on Removed-from-Service Control Rod Drive Mechanism Penetrations (NUREG/CR-6996)

On this page:

Download complete document

Publication Information

Manuscript Completed: April 2009
Date Published: July 2009

Prepared by:
S.E. Cumblidge, S.R. Doctor, G.J. Schuster, R.V. Harris,
S.L. Crawford, R.J. Seffens, M.B. Toloczko, S.M. Bruemmer

Pacific Northwest National Laboratory
P.O. Box 999
Richland, WA 99352

W.E. Norris, NRC Project Manager

NRC Job Code Y6867

Office of Nuclear Regulatory Research

Availability Notice


The U.S. Nuclear Regulatory Commission (NRC) and the Electric Power Research Institute (EPRI) conducted a collaborative research effort to address issues related to cracking of nickel-base alloys and degradation of reactor pressure vessel heads. Control rod drive mechanism (CRDM) nozzles and J-groove weldments were removed from the decommissioned North Anna Unit 2 reactor pressure vessel (RPV) head and shipped to Pacific Northwest National Laboratory (PNNL) in Richland, Washington, and Westinghouse Electric Company LLC in Pittsburgh, Pennsylvania, for study. The primary objectives of the research were to evaluate the effectiveness and reliability of nondestructive examination (NDE) methods as related to the in-service inspection of CRDM nozzles and J-groove weldments and to enhance the knowledge base of primary water stress corrosion cracking (PWSCC) through destructive characterization of the CRDM assemblies.

The North Anna 2 nozzles have been the subject of three series of nondestructive examinations. The first series consisted of the in-service examinations that determined that the RPV head required replacement. The second series of examinations was performed at PNNL by four inspection teams. The examinations were administered by EPRI as part of the cooperative agreement with the NRC. Finally, an extensive series of examinations was performed on several nozzles at PNNL and Westinghouse under laboratory conditions.

Ultimately, the laboratory studies focused on two CRDMs—Nozzle 31, which had cracks, as evidenced by through-wall leakage and the in-service inspection data, and Nozzle 54, which had circumferential defect indications in the penetration tube outer diameter, as evidenced by the in-service examinations.

The NDE procedures used to examine the CRDM assemblies followed standard industry techniques for conducting in-service inspections of CRDM nozzles and the crown of the J-groove welds and buttering. These techniques included eddy current testing (ET), time-of-flight diffraction ultrasound, and penetrant testing. In addition, laboratory-based NDE methods were employed at PNNL to conduct inspections of the CRDM assembly with particular emphasis on inspecting the J-groove weld and buttering. These techniques included volumetric ultrasonic inspection of the J-groove weld metal and visual testing via replication of the J-groove weld. The results from these NDE studies were used to guide the development of the destructive characterization plan.

The comparison of the examination results from the three series of nondestructive examinations performed (i.e., in-service, inspection team, and laboratory) indicate that the testing results are generally consistent. Many of the indications detected during the in-service examinations were detected also during the examinations performed by the inspection teams and in the laboratory. Many of these indications were confirmed through the destructive evaluation (DE) of the CRDMs. However, DE also showed that some significant indications were either not detected or mischaracterized. For example, several flaws were found by DE in the buttering region of Nozzle 54 that were not detected during the in-service examination. Volumetric inspection of the J-groove weld of nozzle 31 found many fabrication flaws but was not able to detect the through-weld crack, as the crack was axially oriented and presented almost no surface area to the ultrasonic beam. Eddy current testing was somewhat more reliable based on a comparison of results. However, eddy current testing also exhibited inconsistencies. In one area of a CDRM assembly, numerous surface-breaking indications were detected by ET were not confirmed in the laboratory.

Some differences between the results of the different inspection methods were expected. The reliability of NDE can vary with surface condition and the complexity of the geometry of a component. CRDM assemblies are very complex in this regard. However, there were two interesting findings from the comparisons of the results of NDE and DE. The first finding was that a significant number of the NDE indications were actually determined to be fabrication-related. The second finding was a realization that the meandering and branched nature of PWSCC can greatly affect detection and characterization. PWSCC cracks are generally tight at the surface. These findings provide a basis for explaining why it can be more difficult to detect cracks than leaks through in-service inspections.

The industry has been working to improve inspection methods and the quality of inspections. The NRC has amended Title 10 of the Code of Federal Regulations, Part 50, paragraph 50.55a(g)(6)(ii)(D), to require that all licensees of pressurized water reactors (PWRs) augment their inservice inspection programs by implementing the recently developed American Society of Mechanical Engineers (ASME) Code Case N-729-1, “Alternative Examination Requirements for PWR Reactor Vessel Upper Heads with Nozzles Having Pressure-Retaining Partial-Penetration Welds, Section XI, Division 1.”

The goal of in-service inspection of nuclear reactor piping and pressure vessels is to reliably detect service-related defects in a timely manner and, thereby, maintain the structural integrity of the inspected components. Thus, relative to this goal, the results of the research reported here indicate that NDE can be improved to be a more effective tool for detecting and characterizing indications in CRDM assemblies.

Page Last Reviewed/Updated Friday, October 18, 2013