| Abstract Scope |
Introduction: Researchers at the Naval Surface Warfare Center, Carderock Division and Carpenter Technology Corporation have collaborated to develop a high-strength steel solid wire welding electrode based on the Fe-10wt.% Ni metallurgical system (dubbed ‘10Ni steel’). As development of the 10Ni steel welding consumable progressed, a more granular understanding of its on-heating and on-cooling transformation behavior was needed to explain microstructural and mechanical property observations and to provide data for increasing the accuracy of computational models. This presentation reports the results of Gleeble-based dilatometry testing of 10Ni steel gas metal arc (GMA) weld metal and use of the resulting data to explain effects of chemistry and heat affected zone (HAZ) peak temperature and cooling rate on the alloy’s transformation behavior. In addition, effects of heating rate on recrystallization and the development of residual stresses during weld fabrication will also be discussed.
Experimental Procedure: Material for this study was produced by fabricating a 300 x 100 x 50 mm (12 x 4 x 2 in.) weld buildup using the GMA welding process and 1.2 mm (0.045 in.) diameter bare solid wire 10Ni steel welding electrode. Deposit chemistry was measured in multiple locations to identify the extent of dilution from the non-matching HY-130 base plate, and 6 mm (0.25 in.) round, 70 mm (2.75 in.) long dilatometry specimens were extracted from undiluted regions of the weld deposit.
A Gleeble 3500 and a linear variable different transformer (LVDT) extensometer were used to apply welding-relevant thermal cycles to the specimens and measure radial dilation as a function of temperature. First, austenite start and finish temperatures were measured for heating rates of 5 to 500 °C/s (9 to 900 °F/s). The lever law approach was used to calculate the temperature-dependent transformation rates. Next, specimens were heated at a rate of 5 °C/s (9 °F/s) and held at a peak temperature of 1200 °C (2192 °F) for 4 min. before cooling at rates between 1.3 and 66 °C/s (2.3 and 119 °F/s). On-cooling phase transformations were identified and compared to data reported by previous researchers for a base material of similar composition but higher carbon content. Finally, specimens were subjected to thermal cycles including heating at 200 °C/s (360 °F/s) to peak temperatures of 800 or 1350 °C (1472 or 2462 °F) followed immediately by cooling at 5 to 75 °C/s (9 to 135 °F/s). The resulting dilation data were used to develop continuous cooling transformation (CCT) diagrams relevant to multipass weld deposit HAZ thermal histories. Coefficients of thermal expansion were also calculated from the dilation data.
Optical microscopy, scanning electron microscopy, and microhardness testing were used to characterize the transformation products.
Results and Discussion: On-heating transformation temperatures between 650 and 800 °C (1202 and 1472 °F) were measured and shown to be stable with variations in heating rate. The difference between the minimum and maximum values were 13 °C (23 °F) and 14 °C (25 °F) for Ac1 and Ac3, respectively. In comparison to similar data for ferritic/pearlitic structural steels, the 10Ni steel weld metal Ac1 values were less rate-dependent. The austenite transformation temperatures measured in this work were not in good agreement with the data reported by Barrick et. al. for 10Ni steel base material. Differences were attributed to the higher base metal carbon and retained austenite contents relative to the weld metal studied in this work.
During cooling after soaking at 1200 °C (2192 °F), the measured 10Ni GMA weld metal martensite start and finish temperatures averaged 400 °C (752 °F) and 281 °C (538 °F), respectively. These values were significantly higher than reported by Fonda and Spanos for a 9 wt% Ni base material that had a higher nominal carbon content but was otherwise similar in composition to the material in this work (average values of 345 and 180 °C [653 and 356 °F], respectively). They were also higher than previously reported by Sinfield et. al. (340 and 210 °C [644 and 410 °F], respectively) for 10Ni steel gas tungsten arc (GTA) weld metal.
Construction of welding-relevant CCT diagrams demonstrated that the on-cooling transformations in this material were of the martensitic type across all investigated peak temperatures and cooling rates. Measured martensite hardness increased marginally with increasing cooling rate and decreasing peak temperature. The measured coefficient of thermal expansion for 10Ni steel weld metal martensite was lower than reported in previous work on martensitic HY-80 plate. The austenite thermal expansion rates were similar for the two materials.
Conclusion: This work has explored the effects of heating rate, cooling rate, and peak temperature on the solid state phase transformations of 10Ni steel gas metal arc weld metal. By comparison to prior work, it has demonstrated the magnitude of alloy chemistry and welding process influences. Because weld metal transformation behavior and thermophysical properties drive the development of strength, residual stress, and distortion in welded structures, the data generated in this work represent an important contribution to the development of this alloy system. |