Seminar: Understanding Laser Welding through Calorimetry and Dimensionless Parameters

Basic to our knowledge of the science of welding is an understanding of the process efficiencies and the dependencies of these dimensionless parameters on the process variables. Calorimetric studies by the author at Sandia National Laboratory of the GTAW, PAW, and LBW welding processes have measured the net heat input to the workpiece thereby quantifying the energy transfer efficiency and in turn permitting an accurate determination of melting efficiency, which indicates how much of the heat deposited by a fusion welding process is used to produce melting. It is indicated that the weld process variables can dramatically affect the melting efficiency and that all fusion welding processes can be optimized to produce the same maximum melting efficiency. This limiting value is shown to depend on the weld heat flow geometry as predicted by analytical solutions to the conduction heat flow equation and as demonstrated by empirical data.

Laser welds were typically made directly into an open Seebeck envelope calorimeter mounted on a CNC table in the workstation of a materials processing laser. Net heat input was measured for several materials, laser focus conditions, and automated welding parameters. Strong correlations were seen with the experimental data and dimensionless parameters that are fundamental to the understanding of fusion welding. It will be shown that heat flow geometry is critical in determining melting efficiency and is strongly affected by weld joint design and the travel speed of the fusion welding process. These conduction heat flow correlations have also been extended to an experimental matrix of 2D welds on important engineering alloys to estimate thermophysical property values for other fusion welding applications. Also to be described is characterization of laser beam quality and measurements of the focused laser beam, made using several contrasting experimental methods developed by the author including diamond apertures and Kapton film. Quality assurance and problem solving for the production of non-nuclear components at Sandia production vendors, and major efforts by the author in laser power measurement and calibration of welding lasers will also be discussed.

Lastly, a package of 14 extensible open source welding software applications called SmartWeld, which provide optimal weld schedules, will be described. The optimization applications enable the user to input desired weld shape dimensions, select a material, process parameters, and to constrain the search problem to meet the application requirements. The models are based on much of the experimental work described above and are thereby well validated. SmartWeld outputs weld procedures that will make improved welds when compared with traditional approaches. Procedures that have been developed only to meet a penetration requirement often are not ideal. Process efficiencies and other important figures of merit can often be improved. SmartWeld enables the user to ask "what if?" and instantly learn the effect of a change in process parameters on several important measures of quality. Applications of the software will be described.