A B Fig. 8 — Micrographs showing solidification cracking in Alloy 57. A — 42 mm/s travel speed; B — 85 mm/s travel speed. welding where the solidification rates would be greater than the conditions in this work, the minimum Creq/Nieq to ensure primary ferrite solidification may increase above 1.75. Crack Severity The binary representation of crack or no-crack on the weldability diagrams does not completely capture the differences in the extent of the cracking observed. The severity of cracking was evaluated using the measured total crack lengths given in Table 2. Crack lengths presented are the total for the entire area of all three cross sections of an alloy. Total crack length varied significantly between different alloys, and varied to a lesser extent with travel speeds for a given alloy. Total crack length as a function of travel speed is shown in Fig. 7 for three groups of alloys — 21-6-9 alloys with impurity content greater than 0.035 wt-%, 21-6- 9 alloys with impurity content less than 0.035 wt-%, and other alloys. The high impurity content 21-6-9 alloys with large (>3000 mm) total crack lengths showed relatively constant crack severity as travel speed increased. Alloys other than 21-6-9 showed the highest crack lengths at 42 mm/s travel speed, but this trend is not observed in the other materials. Overall, the crack severity is much more a function of chemical composition than travel speed. For alloys that showed cracking at all three travel speeds, there was no clear trend in change in crack length with travel speed. While total crack length is roughly constant for a given alloy as travel speed varied, the number of cracks increased at 85 mm/s travel speed for all alloys except Alloy 55. Excluding Alloy 55, it was observed that the length of individual cracks decreased and the number of cracks increased for 85 mm/s travel speed. Alloy 55 had complete sample separation at 85 mm/s; the “cracks” were characterized as three cracks, being one through-thickness centerline crack in each cross section. For the majority of alloys, at 85 mm/s travel speed, the cracking shifted away from large centerline cracks to multiple cracks on solidification grain boundaries away from the weld centerline. The change in weld pool shape and growth angle as travel speed increases may affect the location of the cracking. Figure 8 shows the change in cracking observed for Alloy 57 as travel speed increased from 42 to 85 mm/s, typical of the shift observed from large centerline cracking to multiple noncenterline cracks. The variation in crack severity in the experimental 21-6-9 alloys was presented previously. Comparing total crack length between commercial alloys with similar impurity levels, the Nitronic 50 and Nitronic 60 alloys (33–35) showed much greater cracking than the 21-6-9 Alloy 10. Type 21-6-9 alloys only showed large total crack lengths in the experimental alloys with high impurity contents. The other commercial alloys that showed cracking, 30 and 32, also showed very low total crack lengths. Dual-mode solidification conditions tended to show lower total crack lengths compared to primary austenite solidification in the same alloy, which is likely due to the cracking occurring only in regions of primary austenite solidification. With a lower volume of weld pool that undergoes primary austenite solidification that is susceptible to cracking, decreased crack lengths would be expected. Chemical Composition and Solidification Cracking Care must be used when applying the results of the limited study presented here on the effects of S and P on solidification cracking in 21-6-9 to other high-N, high-Mn austenitic stainless steel alloys. In general, for high-N, high-Mn stainless steels, the other minor constituents of an alloy could significantly change the crack susceptibility. One indication of this is the lower impurity levels that caused cracking in Nitronic 50 alloys compared to 21-6-9 observed in the weldability diagram at 21 mm/s travel speed. Ritter and Savage (Ref. 30) showed that solidification cracking in Nitronic 50 is related to the formation of a niobium carbonitride eutectic in the interdendritic region during solidification. Extensive cracking was also observed for the Nitronic 50 alloys in this work. The larger total crack lengths observed for Nitronic 50 alloys relative to 21-6-9 at similar impurity content could be a function of the niobium rich final solidification products expected for Nitronic 50 alloys. Even within 21-6-9 alloys there is variable cracking at impurity contents WELDING RESEARCH 416-s WELDING JOURNAL / NOVEMBER 2016, VOL. 95
Welding Journal | November 2016
To see the actual publication please follow the link above