ICF10B

thermal expansion mismatch in the vicinity of a weld can significantly affect the fatigue behavior. Thus, knowledge of the fatigue crack growth behavior of such welds is of practical importance. The objectives of this research are to investigate the fatigue crack growth behavior in the presence of the martensite layer found adjacent to the fusion line for as-welded (AW) 309L stainless steel/A285 carbon steel dissimilar metal welds. Included in this objective is an understanding of the residual stress pattern that can develop in such dissimilar metal weld overlay coatings. The results of this research should provide insight as to whether the martensite layer in such dissimilar welds should be of concern in the presence of cyclic loading conditions during service. EXPERIMENTAL PROCEDURE Carbon steel designation ASTM A285 Grade C, of dimensions 61 cm x 15 cm x 4 cm, was utilized as the substrate material. The austenitic overlay used was iron-based 309L stainless steel. The weld strip electrode size was 30 mm × 0.5 mm. Strip electrode and substrate compositions are provided in Table 1. Electroslag welding was used to deposit the strip electrode continuously along the top edge of the substrate at 565 A, 24 V, and 3.0 mm/s travel speed. The electroslag process was chosen because of its abilities to produce a relatively flat fusion line and to provide a uniform cladding with low dilution levels [8]. The resulting dilution level was 17 %. Compact-tension (C(T)) specimens were fabricated from specimens comprised of bulk electroslag welds and electron-beam (EB) extensions of the overlay material. EB welding was chosen to produce a minimal heat affected zone, allowing the original dissimilar metal electroslag fusion line microstructure to remain unchanged during the process. The resulting C(T) specimens were of dimensions 63.5 mm x 60.96 mm x 12.7 mm. Starter notches having a 1.52 mm radius of curvature were inserted by wire electro-discharge machining (EDM) as close to the fusion line as possible as to enable crack growth along the fusion line (parallel to the welding direction). However, since the welding process resulted in a fusion line that was not perfectly flat, it was impossible to insert the starter notch perfectly along the fusion line. Figure 3 shows a photomacrograph of a typical C(T) specimen with starter notch. Neutron diffraction measurements were utilized to determine the residual stress pattern in three orthogonal directions near the dissimilar alloy C(T) specimen weld fusion line. Measurements were performed on the 309L stainless steel AW specimens. Due to differences in crystal structure between the substrate and the overlay, the measurements could not be conducted directly on the fusion line. Thus, for each sample, strains were measured at distances of 1, 3, and 5 mm from the dissimilar alloy weld fusion line (into both the substrate and overlay) at the notch tip, 10 mm, and 20 mm from the notch tip, as illustrated in Figure 4. Fatigue crack propagation tests, conforming to ASTM Standard E647 [13], were performed on AW specimens and on wrought 309L stainless steel and A285 specimens to provide a baseline. Instron servohydraulic fatigue testing equipment was utilized, along with constant amplitude loading and a sine waveform. All testing was performed at a frequency of 25 Hz. Software from Fracture Technology Associates was used to allow for computer-controlled testing. Constant R (equal to 0.10) tests were utilized to generate steady state and higher crack growth rate data. Constant Kmax tests, during which the R ratio increased as the load decreased, were utilized to obtain low crack growth rate and threshold data. Crack lengths were determined by the compliance function and visually verified. Crack growth rates were calculated by the modified secant method. Stereomicroscopy and light optical microscopy were used to characterize the fracture surface and microstructure of the welds following testing. RESULTS AND DISCUSSION Residual Stress Figure 4 shows the residual stress measurements for the 309L stainless steel/A285 carbon steel dissimilar metal weld in the AW condition for the normal (N), transverse (T), and longitudinal (L) directions of the stress. In this figure, the x-axis represents the magnitude of the residual stress, with positive values corresponding to tensile stresses and negative values corresponding to compressive stresses. The y-axis exhibits the distance from the fusion line, with positive values corresponding to locations within the austenitic overlay and negative values corresponding to locations within the ferritic substrate. The legend key illustrates the measurements taken at the notch tip and at 10 mm and 20 mm from the notch tip. From this figure, the residual stresses are tensile in the

RkJQdWJsaXNoZXIy MjM0NDE=