Ferrous friction stir weld physical simulation

• Main Author: Norton, Seth Jason
• Format: Dissertation
• Language: English
• Subjects: engineering, materials science; engineering, metallurgy
• ISBN: 9780542778414; 0542778416

Traditional fusion welding processes have several drawbacks associated with the melting and solidification of metal. Weld defects associated with the solidification of molten metal may act as initiation sites for cracks. Segregation of alloying elements during solidification may cause local changes in resistance to corrosion. The high amount of heat required to produce the molten metal in the weld can produce distortion from the intended position on cooling. The heat from the electric arc commonly used to melt metal in fusion welds may also produce metal fumes which are a potential health hazard. Friction stir welding is one application which has the potential to make full thickness welds in a single pass, while eliminating fume, reducing distortion, and eliminating solidification defects.
Currently the friction stir welding process is used in the aerospace industry on aluminum alloys. Interest in the process by industries which rely on iron and its alloys for structural material is increasing. While friction stir welding has been shown to be feasible with iron alloys, the understanding of friction stir welding process effects on these materials is in its infancy. This project was aimed to better that understanding by developing a procedure for physical simulation of friction stir welding.
Friction stir weld material tracer experiments utilizing stainless steel markers were conducted with plates of ingot iron and HSLA-65. Markers of 0.0625" diameter 308 stainless steel worked well for tracing the end position of material moved by the friction stir welding tool. The markers did not produce measurable increases in the loading of the tool in the direction of travel. Markers composed of 0.25" diameter 304 stainless steel did not perform as well as the smaller markers and produced increased loads on the friction stir welding tool. The smaller markers showed that material is moved in a curved path around the tool and deposited behind the tool. Material near the surface is moved a greater distance as it is acted upon by the tool shoulder.
A friction stir weld was made on a plate of HSLA-65 which had 0.0625" Inconel sheathed thermocouples embedded in the tool path at seven positions. Thermocouples on the top of the plate acquired data at the desired position until encountering the shoulder, at which point they were sheared by the shoulder and stirred behind the tool. Thermocouples on the bottom of the plate were deformed a relatively small amount and acquired data throughout the welding process. Heating rates calculated from the slope of the acquired temperature data show that the peak heating rate (∼1100°C on top and ∼500°C on the bottom) occurs on both the top and bottom of the weld at temperatures between 350°C and 500°C. An increase in the heating rate occurring at elevated temperature was associated with the transformation from ferrite to austenite. Comparison of phase transformation data acquired in rapid heating in the GleebleRTM suggests that austenite transforms back to ferrite at higher temperatures in the presence of strain than in its absence. Peak temperatures on the top of the plate exceeded 1200°C and peak temperatures acquired on the bottom exceeded 1000°C. The heating rate method of data analysis was sensitive enough to pick up variations in the heating rate which occurred at the same frequency as the rotation rate of the tool. (Abstract shortened by UMI.)