Improvement of Modeling Predictions in Friction Stir Welding by More Accurate Measurement of Heat Transfer Between Tooling and Workpiece

Brigham Young University, Provo UT

Investigators

Abstract

Friction stir welding is a solid-state joining method that is finding increased application in joining aluminum alloys that are very difficult to join by conventional fusion welding processes. Industrial sectors that will directly benefit from better quality welded assemblies include core American industries such as aerospace, light rail, marine, and automotive. However, most friction stir welding development is done by experimental trial-and-error, limiting its impact, and slowing its introduction into potential weight-saving applications. Numerical simulation of friction stir welding began about 20 years ago, with advances made in predicting key process conditions (welding temperatures, material flow) and mechanical joint properties. Unfortunately, the predictive value of these models is limited because order of magnitude variations exist in reported friction and heat transfer coefficient model input values. This research aims to utilize thermal wave techniques to measure heat transfer coefficients more accurately than previously achievable. A better understanding and measurement of heat transfer, leading to improvements in modeling predictions, will speed development of friction stir welding, enabling the production of lighter vehicle structures, safer pressure vessels, and more durable nuclear waste canisters, among others. If successful, the technique can also be applied to conventional machining processes where prior thermal measurement efforts with thermocouples have been indirect and approximate at best. In addition to the technical aspects, this project will engage graduate and undergraduates in research and will provide them with opportunities to interact with industrial users, thus increasing their workforce preparedness. Outreach activities are aimed at the university’s Women in Engineering group, a local technical college’s welding program, and local high school agriculture and technology teachers. The research objective of this work is to determine the feasibility of utilizing thermal waves to measure heat transfer coefficients under dynamic processing conditions. Thermal waves are temperature variations in a material that are created by modulating the intensity of an incident laser heat source and are measured as a modulated change in the optical reflectance of the polished surface of the tool (or baseplate). In this system, the waves penetrate from inside the tool into the workpiece, and the magnitude of the thermal resistance between the two parts changes the measured amplitude and phase of the thermal wave. Heat transfer coefficient values will be obtained by fitting the resulting phase to a multi-layered thermal quadrupole model. To verify this novel metrology technique, a dual fiber optic probe will be designed and placed inside the tool and baseplate to create and detect thermal waves during the welding process. The ability of the probe to accurately measure heat transfer coefficients will be verified by static compression tests between discs of H13 steel and aluminum alloys. The measured values will also be compared to well-established analytical models that predict thermal contact resistances of materials with known surface roughness values and static contact pressures. The measured parameters will then be used as inputs in friction stir welding finite element models to demonstrate how improved input parameter values can advance the predictions of loads, temperatures, and material flow for a range of conditions and tool designs. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.