No abstract

A fully coupled thermomechanical finite element analysis of the friction-stir welding (FSW) process developed in the authors' previous work is combined with the basic physical metallurgy of Ti-6Al-4V to predict/assess the structural response of FSW joints. A close examination of the experimental results reported in the open literature reveals that in most cases the heat-affected zone (HAZ) of the weld possesses the most inferior properties and tends to control the overall structural performance of the weld. Taking this observation into account, a microstructure evolution model is developed and parameterized for the Ti-6Al-4V material residing in the HAZ. Specifically, this model addresses the problem of temporal evolution of the globular a-phase particles located within prior b-phase grains (the dominant microstructural parameter in the HAZ) during the FSW process. Next this model is combined with the wellestablished property versus microstructure correlations in Ti-6Al-4V in order to predict the overall structural performance of the weld. The results obtained are found to be in reasonably good agreement with their experimental counterparts, suggesting that the present computational approach may be used to guide the selection of FSW process parameters in order to optimize the structural performance of FSW joints (at least while they are controlled by the HAZ-material microstructure/properties).

To respond to the advent of more lethal threats, recently designed aluminum-armor-based military-vehicle systems have resorted to an increasing use of higher strength aluminum alloys (with superior ballistic resistance against armor piercing (AP) threats and with high vehicle-light weighing potential). Unfortunately, these alloys are not very amenable to conventional fusion-based welding technologies and in-order to obtain high-quality welds, solid-state joining technologies such as Friction stir welding (FSW) have to be employed. However, since FSW is a relatively new and fairly complex joining technology, its introduction into advanced military vehicle structures is not straight forward and entails a comprehensive multi-step approach. One such (three-step) approach is developed in the present work. Within the first step, experimental and computational techniques are utilized to determine the optimal tool design and the optimal FSW process parameters which result in maximal productivity of the joining process and the highest quality of the weld. Within the second step, techniques are developed for the identification and qualification of the optimal weld joint designs in different sections of a prototypical military vehicle structure. In the third step, problems associated with the fabrication of a sub-scale military vehicle test structure and the blast survivability of the structure are assessed. The results obtained and the lessons learned are used to judge the potential of the current approach in shortening the development time and in enhancing reliability and blast survivability of military vehicle structures.

A review of the literature revealed that high-cycle fatigue data associated with friction stir-welded (FSW) joints of AA5083-H321 (a solid-solution-strengthened and strain-hardened/stabilized Al-Mg-Mn alloy) are characterized by a relatively large statistical scatter. This scatter is closely related to the intrinsic variability of the FSW process and to the stochastic nature of the workpiece material microstructure/properties as well as to the surface condition of the weld. Consequently, the use of statistical methods and tools in the analysis of FSW joints is highly critical. A three-step FSW-joint fatigue-strength/life statistical-analysis procedure is proposed in this study. Within the first step, the type of the most appropriate probability distribution function is identified. The parameters of the selected probability distribution function, along with their confidence limits, are computed in the second step. In the third step, a procedure is developed for assessment of the statistical significance of the effect of the FSW process parameters and fatigue specimen surface conditions. The procedure is then applied to a set of stress-amplitude versus number of cycles to failure experimental data in which the tool translational speed was varied over four levels, while the fatigue specimen surface condition was varied over two levels. The results obtained showed that a two-parameter weibull distribution function with its scale factor being dependent on the stress amplitude is the most appropriate choice for the probability distribution function. In addition, it is found that, while the tool translational speed has a first-order effect on the AA5083-H321 FSW-joint fatigue strength/life, the effect of the fatigue specimen surface condition is less pronounced.

No abstract