Modelling of tool wear and metal flow behaviour in friction stir welding (FSW)

Hasan, Ahmed Falh (2016) Modelling of tool wear and metal flow behaviour in friction stir welding (FSW). PhD thesis, University of Nottingham.

[thumbnail of Thesis after corrections]
Preview
PDF (Thesis after corrections) (Thesis - as examined) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (11MB) | Preview

Abstract

Friction Stir Welding (FSW) is a solid-state joining process that was invented in 1991; it is particularly useful for joints difficult to make using fusion techniques. Significant advances in FSW have been achieved in terms of process modelling since its inception. However, until now experimental work has remained the primary method of investigating tool wear in FSW.

In this project, two main objectives were set; the first one was to produce a numerical approach that can be used as a useful tool to understand the effect that worn tool geometry has on the material flow and resultant weld quality. The second objective was to provide a modelling methodology for calculating tool wear in FSW based on a CFD model.

Initially, in this study, a validated model of the FSW process was generated using the CFD software FLUENT, with this model then being used to assess in detail the differences in flow behaviour, mechanically affected zone (MAZ) size and strain rate distribution around the tool for both unworn and worn tool geometries. Later, a novel methodology for calculating tool wear in FSW is developed. Here a CFD model is used to predict the deformation of the highly viscous flow around the tool, with additional analysis linking this deformation to tool wear. A validation process was carried out in this study in order to obtain robust results when using this methodology. Once satisfied with the tool wear methodology results, a parametric study considering different tool designs, rotation speeds and traverse speeds was undertaken to predict the wear depth. In this study, three workpiece materials were used which were aluminium 6061, 7020 and AISI 304 stainless steel, while the materials used for the tools used were of H13 steel and tungsten-rhenium carbide (WRe-HfC) with different tool designs.

The study shows that there are significant differences in the flow behaviour around and under the tool when the tool is worn and it shows that the proposed approach is able to predict tool wear associated with high viscous flow around the FSW tool.

With a simple dome shaped tool, the results shows that the tool was worn radially and vertically and insignificant wear was predicted during welding near the pin tip. However, in other regions the wear increased as the weld distance increased. Additionally, from the parametric study that was undertaken for the two tool designs - a dome and a conical shape- the study has found that for both tool designs, wear depth increases with increasing tool rotation speed and traverse speed. It was also shown that, generally, the wear depth was higher for the conical tool design than the dome tool in the pin tip zone. The research concludes that a proposed methodology is able to calculate tool wear associated with high viscous flow around the FSW tool, which could be used as a method for calculating tool wear without the need for experimental trials. The CFD model has provided a good tool for prediction and assessment of the flow differences between un-worn and worn tools, which may be used to give an indication of the weld quality and of tool lifetime. Furthermore, from the results, it can be concluded that this approach is capable of predicting tool wear for different process parameters and tool designs and it is possible to obtain a low wear case by controlling the process parameters.

Item Type: Thesis (University of Nottingham only) (PhD)
Supervisors: Bennett, Chris J.
Shipway, Philip H.
Keywords: Friction stir welding, Mechanical wear, Testing
Subjects: T Technology > TS Manufactures
Faculties/Schools: UK Campuses > Faculty of Engineering
Item ID: 36154
Depositing User: Hasan, Ahmed
Date Deposited: 28 Oct 2016 10:24
Last Modified: 12 Oct 2017 21:21
URI: https://eprints.nottingham.ac.uk/id/eprint/36154

Actions (Archive Staff Only)

Edit View Edit View