| Abstract Scope |
Friction stir welding (FSW) is a solid-state welding process which utilizes a non-consumable, rotating tool to generate frictional heat within base metals while simultaneously contributing to the mechanical mixing of the materials. Tool selection has become known as the most influential aspect of the FSW process. Conventionally, the tool is responsible for the heat generated during welding, as well as the mixing of the base metals. A threaded, cylindrical pin with a concave shoulder is typically used; however, complex tool geometries have produced significant improvements.
The toolpath used conventionally in FSW is fixed parallel to the weld seam. This offers a simple and effective approach to FSW, as it does not require any advanced programming to generate this type of path. An offset may be applied in which a larger portion of the tool travels along one material. This was found to influence the strength of welds of dissimilar materials. While temperature was not recorded within these studies, it is hypothesized that this technique may be used to bias the heat generated within the workpiece to achieve a desired temperature. However, this is likely at the sacrifice of reduced intermixing of the base metals because of the limited reach of the tool into one of the materials.
The path tested herein formed a curling pattern in which the tool crossed the seam of the joint traveling approximately perpendicular to it. The aim of this toolpath is to promote the intermixing of the base metals, which is typically accomplished through tool selection. A distinct advantage of toolpath modification over complex tool use is its high level of flexibility. Complex tooling does not easily allow for alteration of parameters, which may be necessary for successful welding of different materials. The use of a complex toolpath, however, removes this limitation, as the toolpath may be changed expeditiously according to the material/desired outcomes. Another advantage this toolpath has over complex tooling is its relatively low investment cost. Additionally, it is hypothesized that this toolpath will offer the user greater control over temperature distribution due to the asymmetric shape of the toolpath.
While the greatest advantage of this toolpath may be realized when welding dissimilar metals, this initial investigation explored the welding of 6061-T651 aluminum. A critical modification of this tool is the convex shoulder, shown exaggerated in the figure. Conventional cylindrical FSW tools are axially angled towards the trailing end of the toolpath to mitigate plowing of the material from the leading edge of the tool and to force material into any voids which may form during welding. By creating an angled shoulder, plowing of the material can be largely eliminated; however, this is at the cost of creating large voids.
Throughout all tests, the tool's rotational speed was maintained at 1800 rpm clockwise. The tool was plunged to 1.99 mm (halfway up the angled shoulder) at a rate of 2.92 mm/min. Once this depth was reached, the tool was set to dwell for 30 seconds. The tool then traveled at 152.4 mm/min while following the tool path. Temperature was measured utilizing thermocouples placed offset from the seam of the weld. Tensile strength was measured at two different locations relative to the path.
The maximum temperature is not seen directly at the seam of the weld for tests exhibiting the curly toolpath. A maximum of +14.6% and minimum of -4.5% temperature difference between the seam and 6.35 mm from the seam was discovered. The difference between temperatures at the measured locations was found to be highly dependent on the amplitude of the path, while the amplitude of the temperature was found to be dependent on the wavelength. It was found that as amplitude increases, higher temperatures are found further from the seam of the weld and as wavelength increases, temperature magnitude decreases. This intuitive trend can be explained as an increase in energy input to the workpiece resulting from a longer toolpath.
It was found that a majority of tests displayed a positive relationship between welding force and welding temperature. It is hypothesized that the increase in temperature strengthened the material. Conventional toolpaths would not be affected by this increase in strength since strengthening would occur at the trailing edge of the tool and the thermal softening effects of the temperature increase would dominate over any strengthening of the material. However, because the curly toolpath passes over material which has been previously welded, it is possible that the strengthening of the material becomes crucial.
The curly toolpath resulted in an improvement in tensile strength over the conventional FSW toolpath for various parameters. The greatest improvement in average tensile strength was 22.2% relative to the conventional FSW toolpath. A decreasing trend was found in tensile strength as the overlap is increased. The location of the tensile test relative to the toolpath produced a significant difference.
The use of a complex, curly toolpath was investigated herein for its feasibility for use in FSW. A cylindrical pin was used throughout testing without an axial tilt. For the parameters tested, it was found that the curly toolpath is able to improve the strength of the weld by a maximum of 22.2%. Additionally, utilizing the curly toolpath was able to close visible voids apparent in baseline testing. Several trends were determined between the tested parameters, such as a negative relationship between wavelength and welding temperature and a positive relationship between wavelength and tensile strength. However, these relationships cannot be generalized for all tests, as they depend on several other factors.
A supplemental study was conducted in which the tool’s rotation was reversed. This caused significantly less void creation, leading to a 34.9% increase in strength over the conventional toolpath. This result provides evidence towards the significance of tool rotation when using this type of tool and toolpath. Therefore, future studies will incorporate both clockwise and counterclockwise rotations. |