Material failures in heat exchangers are often closely tied to events associated with the conditions of service and operating parameters. These events can generally be attributed to adverse load application and higher than optimum operating temperatures that could lead to changes in the microstructure of the materials and fatigue failure of the component. However, fatigue failure in heat exchangers is usually associated with the presence of a biaxial stress condition. Two nonparallel forces create a two-dimensional stress field at the free surface of the structural element where the process and mechanism of fatigue failure normally initiate. An experimental investigation was conduct6d to evaluate the biaxial fatigue behavior of commercially pure titanium Ti-50A (Grade 2) and low-alloy titanium Ti-Code 12 (Grade 12) heat exchanger materials. The biaxial state of stress was composed of an axial stress and a superimposed torsional stress, applied in a thin-wall tubular specimen machined from titanium tubing. Torsional stress was applied independently using a torsion machine and a torque fixer assembly devised as part of this study. After applying the desired torsion, the torsionally stressed specimen was mounted on a closed-loop electrohydraulic machine for the application of axial cyclic loading. A minimum of four tests were conducted for each of three alternating stress levels at both high and low torsional stresses. The biaxial fatigue test under load control condition was done under fully reversed cycles equivalent to a biaxiality ratio of -1. These test parameters were determined from an analytical formulation based on Mohr's circle. The results are presented in terms of the various measured or calculated quantities versus number of cycles to fracture. Biaxial fatigue curves were drawn through the experimental points corresponding to Weibull's mean life criterion. The four data points exhibit scatter that appears to be related to the applied stress amplitude. It was also found that a correlation exists between the magnitude of applied cyclic biaxial stress and fatigue life to failure. In addition, the results have been discussed taking existing failure criteria into account. / Graduation date: 1985

This dissertation involves the experimental investigation of commercially pure titanium (CP Ti) which was subjected to laser forming and mechanical forming processes. Commercially pure titanium grade 2 was formed to a radius of curvature of approximately 120 mm using three forming procedures, i.e. i) laser forming; ii) mechanical forming (stretched forming) and iii) a combined forming process (laser-mechanical forming). Fatigue testing revealed, for all the forming processes, that samples produced by laser forming performed the best at high load settings. However, mechanically formed specimens performed the best at low load settings, whereas the laser-mechanical process resulted in midway performance between laser and mechanical processing. Considering microstructure vs fatigue; impact vs fatigue; and residual stress vs fatigue; at high load settings it is evident that the microstructure is the dominant contributor to crack initiation and growth. Crack morphology of fatigue samples revealed that secondary cracks (parallel to main crack front) followed the grain boundaries of the Widmanstätten microstructure, whereas irregular secondary cracks grew parallel and through the twinning planes and along the grain boundaries of the equiaxed microstructure. Laser forming resulted in microstructural changes from equiaxed grains to a Widmanstätten structure due to fast cooling rates. Excessive twinning is developed within the equiaxed microstructure after the mechanical forming procedure. This is due to cold working / strain hardening. The combined process shows a combination of equiaxed grains and Widmanstätten microstructure. Residual stress relieved for all forming processes revealed an increase in the magnitude of the residual stress compared to the parent plate and that the maximum values were obtained at the inner radius of curvature (i.e. 118.4 mm). Laser forming revealed the highest values in residual stress whereas the other two processes i.e. mechanical and laser-mechanical forming exhibited an increase midway between the parent plate and laser forming. The second most influential factor with regards to fatigue was the magnitude of the residual stress, especially at medium to low load settings. When considering theoretical models to predict fatigue life it was found that the Goodman model showed the closest relation to the actual fatigue data when considering the entire theoretical curve. Vickers microhardness profiling was applied to the thickness of the samples for the parent plate and all forming processes. No significant hardening occurred due to the forming processes and differences in hardness were considered negligible. Charpy impact testing revealed that the laser formed specimens exhibited the most brittle behaviour when compared to the parent plate results. Mechanical formed specimens showed a slight increase in brittleness compared to parent plate whereas the combined process yielded results midway between the laser and mechanically formed specimens. Mathematical equations are formulated and presented for predicting the fatigue life of CP Ti grade 2 for the parent plate and the three forming processes. This study proved that the laser forming process can be successfully used as a production stage in the forming of CP Ti grade 2.

M. Tech. Metallurgical Engineering / Titanium and its alloys exhibit poor tribological characteristics. The poor resistance to sliding wear of Ti6Al4V alloy makes it susceptible to severe wear at the surface during sliding contact. This could cause galling and seizing during sliding contact. Ti6Al4V alloy also have poor corrosion resistance under critical conditions. Some problems with Ti6Al4V MMCs produced by laser cladding technique in most cases is poor bonding as a result of wetting properties between the ceramic and metal powders for reinforcement. Occurrence of porosity is another factor which can reduce the mechanical properties of MMCs. Occurrence of agglomerates is also a concern due to poor mixing of reinforcement powders. This project is aimed at investigating the effect of laser cladding of titanium alloy substrate with zirconium (Zr), titanium carbide (TiC), titanium (Ti) reinforcement additions. The effect of combination of these powders using various fractions and variable cladding parameters on the substrate will be investigated.