Date of Award


Document Type

Campus Access Dissertation


Mechanical Engineering

First Advisor

Anthony P Reynolds


At present Friction Stir Welding process design and implementation is mostly dependent on empirical information gathered through experience. Basic science of friction stir welding and processing can only be complete when fundamental interrelationships between process control parameters and process response variables and resulting weld microstructure and properties are established to a reasonable extent. It is known that primary process control parameters like tool rotation and translation rate and forge axis force have complicated and interactive relationship to the process response variables such as peak temperature, time at temperature, torque and power and in-plane forces.

Of primary influence to the other process response parameters is the temperature and its gradient at the deformation and heat affected zone. Through exhaustive series of experiments and some closely related finite element simulations this work seeks to understand the nature and effects of temperature transients by varying thermal conditions at tool, work piece and backing plate boundaries.

Various thermal management methods are applied at work-piece, tool and backing plate boundaries. This is achieved by performing series of welds using various parameter sets in regular lab air, submerged under water and under auxiliary coolant. Series of welds with various parameter sets are also made over various backing plates made out of materials ranging from low diffusivity ceramic tile to high diffusivity aluminum. Some welds were also made using Densimet and Nimonic as tool materials to assess the effects of thermal gradient at the tool. The use of enhanced work piece surface quenching in 6.35 mm thick AA7050 by welding underwater and under auxiliary coolant resulted in reduced probe temperature, increased torque and power consumption and better mechanical properties. No reduction in time at temperature in the HAZ was observed when the welding speed was greater than 5.1mm/sec. Joint Properties, peak temperature and other process variables were significantly affected by backing plate diffusivity and applied forge axis force in welds made with 25.4 mm thick AA6061 and 4.24mm thick AA6056. Through-thickness homogeneity was successfully achieved in up to 25.4 mm thick welds in terms of grain size and micro-hardness by the use of low conductivity backing plate. A novel composite backing plate that can retain through thickness homogeneity at the same time enhance cooling rate at the HAZ is introduced. Hardness testing has indicated that composite backing plate results in homogenous and superior joint property.