WELDING RESEARCH Fig. 5 — Metallographic cross sections through the high-power laser welds for the following: A — Ni-Ar-HP (3.4-mm penetration); B — Nivac HP (9.8-mm penetration); C — Ti-Ar-HP (2.7-mm penetration); D — Ti-vac-HP (8.7-mm penetration). There is about 4× increase in penetration for both materials when welding under reduced atmosphere. Magnification and scale are the same in all macrograph. NOVEMBER 2016 / WELDING JOURNAL 423-s responds to 721 projections for the samples used in this study. The component was positioned at a distance from the x-ray source that ensured that the projected image fell within the confines of the detector sides at all angles of turntable rotation. X-ray settings (kV and mA) were selected that sufficiently penetrated the component at all angles without saturating the detector. A 1-mm-thick copper filter was placed in front of the x-ray source to reduce saturation by low-energy x-rays from the Bremsstrahlung emission of the source. When a scan was completed, the captured digital radiographs were processed through a reconstruction algorithm into a 3D representation of the component, sometimes referred to as a CT volume. During reconstruction, each pixel from the detector at every angular increment was processed into a three-dimensional volume element with grey level and position (these are known as voxels). For the settings used here, the reconstructed voxel size varied from 15.3 to 26.1 m, depending on the CT reconstruction setup for different samples, as summarized in Table 3. Once generated, the CT volumes were loaded into a visualization software package for analysis, using Volume Graphics VGStudio Max Version 2.2. This package displays a 3D rendering of the sample along with 2D slice views in the each of the three axes, plus a rotational slice view about the vertical axis. Typically, features are observed in the 2D slices followed with creation of slice planes within the 3D volume rendering. A second CT visualization software package, Avizo (Ref. 12), was also used for quantification of the pore volumes in the same manner performed in a previous investigation (Ref. 1). To calibrate the threshold intensity for quantification, different threshold values were used until the analysis returned a correct volume for the flatbottomed hole. The output of the analysis is a spreadsheet that lists all segmented porosity and their respective volumes and the amount of porosity as a function of the distance from the front of the weld as performed in a previous study (Ref. 1). After performing the CT scans, the welds were cross sectioned, cold mounted in epoxy resin, and metallographically prepared using standard lapping and polishing procedures. After polishing, the nickel welds were chemically etched and photographed using a digital microscope to observe the weld microstructure and to measure the weld pool dimensions. The weld fusion zone cross-sectional area was further measured using the digital microscope software to calculate the area contained within a closed path perimeter line drawn around the weld fusion boundary, including the weld crown. All of the weld measurements are summarized in Table 4. Results and Discussion Comparison of Laser Beam Welds Made in Vacuum and Ar Shielding To demonstrate the qualitative effect of vacuum on reduction in porosity, preliminary Ni and laser welds were made and radiographed using conventional 2D x-ray methods. The results are shown in Fig. 4 where dark areas represent porosity in the welds. Figure 4A, B compare the Ni welds in vacuum and Ar, respectively. Small amounts of porosity are present in the vacuum weld, but cannot be distinguished in this figure, whereas large amounts of porosity are clearly evident in the Ni weld made in Ar. Based on the results of the 2D radiographs on the preliminary welds, a more thorough study was performed where each metal was welded with two different power levels, referred to as low power (LP) and high power (HP) hereafter. These welds were made in both vacuum and atmospheric Ar shielding gas conditions, and the results were characterized using optical metallography techniques and 3D xray CT. Additional welds were made using the electron beam process, which is performed in vacuum only A C B D
Welding Journal | November 2016
To see the actual publication please follow the link above