Fig. 1 — Sigmajig test fixture in the laser welding enclosure. regarding influence on crack susceptibility. Several researchers have shown that P has a larger effect on increasing solidification cracking susceptibility than S in both austenitic stainless steels (Refs. 15– 18) and high-manganese steels (Ref. 19). The differing effects of P and S on solidification cracking were also observed for 21-6-9 (Ref. 1) and were taken into consideration for the weldability diagrams developed. The goal of Part 2 of this work was to characterize the solidification cracking behavior of 21-6-9 during laser welding. Weldability diagrams for laser welded 21-6-9 were developed, relating solidification mode and solidification cracking to chemical composition and impurity content. The weldability diagrams were created using a large number of heats of 21-6-9 for a range of solidification rates to observe the shift in critical Creq/Nieq for primary ferrite solidification as solidification rate increases. Experimental A brief recapitulation of the experimental work is given here. For further details, see Part 1 of this investigation (Ref. 1). A wide range of chemical compositions of high-nitrogen, highmanganese stainless steel materials, found in Table 1, were tested to examine their solidification mode and solidification cracking behavior. All chemical compositions were determined using A B optical emission spectroscopy (OES) for the majority of the elements, and Leco inert gas fusion techniques for nitrogen, carbon, and sulfur. The chemical composition presented is the average of three measurements for each alloy. Sigmajig weldability testing (Ref. 20) was used to compare the solidification crack susceptibility of the various alloys. Figure 1 shows the Sigmajig fixture with sample after welding. Samples 32 ¥ 24 ¥ 2 mm were used with the stress applied along the 32-mm length and welding along the 23-mm length. A stress of 310 MPa (45 ksi) was used for all samples. The welding power source was a 1-kW multimode IPG fiber laser, and three travel speeds of 21, 42, and 85 mm/s (50, 100, and 200 in./min) were used. The laser power was adjusted at each travel speed to maintain a complete-jointpenetration weld on the 2-mm sample thickness. The laser power used was 555, 755, and 1166 W at 21, 42, and 85 mm/s travel speed, respectively. The Sigmajig samples were inspected for cracking by examining the surface and three transverse cross sections from the end of the weld as shown in Fig. 2. The transverse cross-sections were all taken at the end of the weld because that is the location where transverse tensile stresses are known to be highest (Ref. 21), and where cracks were observed in the surface inspection when present. All three cross sections were examined to characterize solidification mode and solidification cracking through microstructural observation. Standard metallographic preparation procedures were used and electrolytic etching was done with 10% oxalic acid. The solidification mode was characterized with light optical microscopy at a variety of magnifications. Examination for solidification cracks was also conducted with light optical microscopy, with a maximum of 1000¥ magnification. Correlation of the solidification mode, solidification cracking response, and the chemical composition was used to develop the weldability diagrams for the three travel speeds used. Detailed description of the primary solidification modes and representative microstructures can be found in Part 1. Quantifying the length of cracks (when present) was used to assess the severity of solidification cracking for each cross section. Results and Discussion Weldability Diagram The materials used span a wide range of Creq/Nieq, from roughly 1.1 to 1.9. Experimental laboratory heats were used to expand the composition range of 21-6-9 alloys beyond the somewhat limited range of the commercial alloys available. The weldability diagrams were developed using the established convention of plotting total impurity content vs. Creq/ Nieq. Previous researchers have defined total impurity content as P + S. In this work, a coefficient of 0.2 for S was used on the vertical axis based on the regression analysis results presented in Part 1. Hereafter, total impurity content refers to P + 0.2 S. To plot the Cr and Ni equivalencies of the alloys studied here, the equivalents developed by Espy (Ref. 22) were used, which are given by the following equations WELDING RESEARCH 410-s WELDING JOURNAL / NOVEMBER 2016, VOL. 95 Fig. 2 — Schematic Sigmajig sample with weld overfill and bead morphology shown. A — After welding; B — after sectioning showing the three transverse cross sections taken for microstructural analysis. All dimensions in mm.
Welding Journal | November 2016
To see the actual publication please follow the link above