Abstract
The purpose of the research study was to analyze residual stresses induced and develop predictive mathematical models for mechanical properties like ultimate tensile strength, microhardness and percentage elongation. The Box Behnken design was used to generate the set of weld matrices from three process parameters that were allowed to vary. The welding speed (N), friction force (FR) and forging force (FF) were the parameters that varied while the rest were kept constant. The minimum and maximum welding speeds, friction forces and forging forces of 1500RPM and 3500RPM, 3 kN and 7 kN, and 5 kN and 14 kN respectively, were utilized in the generation of the weld matrix.
The analysis of the joints revealed superior mechanical properties when compared to the base material. A maximum ultimate tensile strength (UTS) of 1043 MPa was obtained at the welding speed of 2500 RPM and 7 kN friction force. The peak hardness of all weld joints was obtained at the weld centre zone (WCZ) with the maximum peak of all being 406 HV0.3. The maximum achievable elongation was 26 %. Based on the experimental and predicted values of UTS, micro-hardness (MH) and percentage elongation (% E), the welding speed and forging force had more effect on these output responses. The study concluded that based on the fluctuations obtained when these input parameters were varied.
The scanning electron microscope (SEM) micrograph of the weld joints revealed a gradual change in the microstructure of the base material to the centre of the weld interface. The metallurgical analysis of the base material revealed equi-axed α grains within the matrix of transformed β. The SEM image of the weld interface suggested
Page | VI
Analysis and characterisation of residual stresses on Rotary Friction Welded Ti6Al4V alloy rods by MC Zulu
that there were three distinct weld zones obtained, which were the WCZ, Thermo-mechanically affected zone (TMAZ) and heat-affected zone (HAZ). The WCZ showed evidence of recrystallisation because of weld temperatures exceeding transus temperature. Very fine α grains were obtained at this zone. A basket-weave microstructure was observed at a low welding speed due to a steeper temperature gradient. Elongated grains arranged in a flow pattern suggested material flow in a radial direction at the TMAZ. Closer to the boundary of WCZ, TMAZ had very fine grains with traces of martensitic and acicular α grains. The HAZ had fully transformed β phase and α phase was clearly visible and coarser. The energy dispersive x-ray spectroscopy (EDS) analysis revealed that there was no compositional change due to RFW in Ti6Al4V alloy.
The X-ray diffraction (XRD) patterns for the phase identification of friction-welded samples were measured in two locations. The patterns from all welded samples were similar with slight differences in the intensity peaks between the two measuring positions. The difference in intensity was linked to the variation in microstructure between these measurement positions. Ti6Al4V, Ti3AlN TiFe2Al and Ti are the phases identified after comparing the results to the available database. The d-spacing results obtained from all welded samples were within the range of 1.3190 Å to 1.3278 Å with the minimum and maximum deviations of 1.6E-4 Å and 1.3E-3 Å respectively across all measurements. The d-spacing values were measured across three principal directions and two gridlines.
The residual stress results suggested that friction-welded Ti6Al4V alloy samples were mainly dominated by stresses that are compressive in nature. The maximum negative
Page | VII
Analysis and characterisation of residual stresses on Rotary Friction Welded Ti6Al4V alloy rods by MC Zulu
value of 380 MPa was obtained along the axial direction at the gridline x=0 while the maximum positive of 100 MPa was obtained at gridline x=4 at all principal directions. The maximum compressive residual stress value of 420 MPa is achievable at gridline X=0 for the weld joints of low friction force and welding speed. In addition, the results suggested that residual stresses have an inverse relationship with welding speed and friction force.