A mathematical and experimental analysis of friction stir welding of steel

AL-MOUSSAWI, Montadhar (2018). A mathematical and experimental analysis of friction stir welding of steel. Doctoral, Sheffield Hallam University.

Preview

PDF Al-Moussawi_2018_PhD_MathematicalExperimentalAnalysis.pdf - Accepted Version Creative Commons Attribution Non-commercial No Derivatives. Download (36MB) | Preview

Abstract

In the last decade there has been significant research into joining steel alloys using the Friction Stir Welding technique due to its ability to carry out welding below the melting point of the parent material and without using fillers such as in fusion welding techniques. This coincided with the increased use of DH36 and EH46 steel grades for ship building. The main reason for joining these steel grades by the friction stir welding technique is to reduce the weight of the vessel, as well as, the high welded joint quality especially mechanical properties such as fatigue and impact resistance. Other improved physical characteristics include increased tensile strength, microhardness and surface finish. This research project has attempted to model friction stir welding using Computational Fluid Dynamics (CFD). Three different approaches have been used when considering the interface between the tool and the workpiece; these are torque, sticking/slipping and fully sticking. The project also investigates the mechanical properties of the welded joints including tensile, fatigue and microhardness. The microstructural evolution of welded joints carried out using different welding parameters is also investigated. The phenomenon of elemental precipitation/segregation during the friction stir welding process has been investigated and the limit of tool rotational speed at which the segregation occurs has been determined by modelling and also by heat treatment to simulate FSW. The purpose of the heat treatment trials was to attempt to replicate the temperature and time that the parent materials experiences during the FSW process. Defects in the weld joints associated with unsuitable friction stir welding parameters were also investigated and two new types of defect have been identified for the first time. Finally, tool wear has been investigated in the different weld joints in order to understand the suitable welding parameters that can prolong tool life. The results from the mathematical modelling of FSW using CFD showed that the fully sticking assumption is the most effective approach for modelling friction stir welding of steel. The model also revealed that local melting at the advancing-trailing side of the tool is likely to occur at high tool rotational speeds. The experimental findings were in agreement with the results from the CFD model as the high tool rotational speed welded joints showed elemental segregation of Mn, Si, Al and O which only occurs when the peak temperature during welding approaches the melting point of steel. Experimental work has also shown significant improvement in the mechanical properties of the welded joints in terms of fatigue and tensile strength after friction stir welding compared to the parent metal. However, the joints welded at high tool rotational/traverse speeds have shown lower mechanical properties as a result of defects such as weld root defect and microcrackes which have been introduced. Tool wear was found to increase with the increasing tool rotational speed as a result of the tools W-Re binder softening. Tool wear was also found to increase with increasing plunge depth as a result of the high shear stress originating from the high thermo-mechanical action at the FSW tool surface. The current project has contributed to knowledge in the friction stir welding of steel by revealing the limits of tool speed that causes elemental segregation. The new technique for estimating the peak temperature and cooling rate using TiN precipitates can also be an alternative to thermocouple measurements which can significantly underestimate the tool-workpiece interface temperature.

Item Type:	Thesis (Doctoral)
Contributors:	Thesis advisor - Smith, Alan
Additional Information:	Supervisor: Professor Alan J. Smith.
Research Institute, Centre or Group - Does NOT include content added after October 2018:	Sheffield Hallam Doctoral Theses
Identification Number:	https://doi.org/10.7190/shu-thesis-00064
Depositing User:	Hilary Ridgway
Date Deposited:	27 Apr 2018 12:40
Last Modified:	26 Apr 2021 13:33
URI:	https://shura.shu.ac.uk/id/eprint/20985

Actions (login required)

View Item

Downloads

Publication Metrics