Measurement of Bulk Residual Stress Distributions in Thick-section Components using the Contour Method

ABSTRACT

Residual stresses are known to play a significant role in many material failure processes (e.g., fatigue, fracture, and stress corrosion cracking). For example, pressure vessels typically contain welded joints. In many cases these welded joints are large (e.g., traveling around the entire circumference), contain significant amounts of residual stress, have reduced material properties, and contain defects. For these and other reasons, the welded joints tend to be critical locations in terms of design and performance of pressure vessels. In the aerospace industry, where large integral components are often machined from a single piece of material, bulk residual stresses can lead to significant distortion of the machined part. Thus, the ability to accurately quantify residual stresses through measurement is an important engineering tool.

The contour method is as an effective tool for generating 2D maps of residual stress normal to a plane in relatively thick components [1, 2]. The contour method consists of three basic steps. First, the specimen containing residual stress is cut in half (wire EDM typical). Cutting creates a new traction-free surface within the body, which results in deformation due to residual stress release. Second, the normal direction displacements resulting from residual stress release are measured over both cutting surfaces (opposite sides of the cut). Third, an analysis is performed to calculate the initial residual stress acting normal to the surface of the cut from the measured displacements [1, 2].


Recently, a cooperative research program was initiated between the US Nuclear Regulatory Commission (NRC) and industry to address weld residual stress issues. The NRC and the Electric Power Research Institute (EPRI) signed a memorandum of understanding that allows and encourages cooperation in nuclear safety research related to dissimilar metal weld (DMW) residual stress fields [3]. As part of the program, measurements and predictions were performed on welded components including safety and relief nozzles from the pressurizer of the canceled WNP-3 plant. One nozzle contained a section of stainless steel pipe welded to the stainless steel safe-end of the nozzle. A schematic of the nozzle configuration is shown in (Figure 1). This nozzle was 711 mm long with a 201 mm outer diameter and an inner diameter of 113 mm at the location of the DMW (44 mm wall thickness).

The contour method was used to measure a two-dimensional map of the hoop residual stress at two angular locations separated by 90° and a two-dimensional map of the axial residual stress at the center of the weld in section of the nozzle between them. All measurement planes are shown in Figure 1; the contour method provides a map of the residual stress component normal to the plane versus position within the plane. The contour measurements were performed in succession with the first hoop residual stress measurement (70°) completed before the second hoop residual stress measurement (160°), and finally the axial stress measurement.

The contour method was used to measure a two-dimensional map of the hoop residual stress at two angular locations separated by 90° and a two-dimensional map of the axial residual stress at the center of the weld in section of the nozzle between them. All measurement planes are shown in Figure 1; the contour method provides a map of the residual stress component normal to the plane versus position within the plane. The contour measurements were performed in succession with the first hoop residual stress measurement (70°) completed before the second hoop residual stress measurement (160°), and finally the axial stress measurement.

Illustration of nozzle used for residual stress measurement

Figure 1 – Illustration of nozzle used for residual stress measurement


Due to the relative proximity of one measurement plane to another, each measurement affects the residual stress remaining in the nozzle at the locations of subsequent measurements. To account for the partial residual stress release from previous measurements, a multi-cut contour method approach was employed [4]. For example, the initial hoop residual stress residual stress at the 160° measurement plane is the sum of the residual stress released at the 160° measurement plane during the measurement performed at 70° and the stress subsequently measured (remaining) at the 160° location. Because the stress at the 70° plane was measured, it could be used to estimate the stress release at 160°. A similar approach is used for the axial measurement planes (with two previous cuts to use for the correction). Strain gage data provided measured deformation due to prior cuts, and validated use of measured stresses to account for partial stress release from prior cuts.

Contour plots of the measured hoop residual stress in the nozzle are shown in Figure 2. In general, significant compressive hoop residual stress exists on the inner diameter near the DMW. The region of compressive hoop residual stress grows larger through the butter and into the carbon steel region. Tensile hoop residual stress exists near the outer diameter, in a region that is shifted towards the stainless steel safe-end of the nozzle. A contour plot of the measured axial stress is shown in Figure 3.

Two-dimensional map of the hoop residual stress for the nozzle at the 70-deg location and 160-deg locations

Figure 2 – Two-dimensional map of the hoop residual stress for the nozzle at the 70-deg location and 160-deg locations

Two-dimensional map of the measured axial residual stress for the nozzle (70-deg to 160-deg, counter-clockwise from x-axis)

Figure 3 – Two-dimensional map of the measured axial residual stress for the nozzle (70-deg to 160-deg, counter-clockwise from x-axis)

This presentation will provide an overview of the contour method and will discuss results from recent programs where the contour method was used to quantify residual stress in thick sections. Results from the NRC/EPRI weld residual stress program will be presented. In addition, results from recent measurements on aluminum aerospace components will be shown.

Next post:

Previous post: