Evaluation of Existing Mass Distribution Model for
Multi-Orifice Flows
by
Jason Sydney Barnwell
Submitted to the Department of Mechanical Engineering
in Partial Fulfillment of the Requirements for the Degree of
BACHELOR OF SCIENCE IN MECHANICAL ENGINEERING
at the
MASSACHUSETTS INSTITUTE OF TECHNOLOGY
January 2000
ABSTRACT
Droplet-Based Manufacturing (DBM) is an application of the Uniform Droplet Spray (UDS) process developed at the MIT Laboratory for Manufacturing and Productivity. DBM creates deposits using metal droplets that are generated at a consistent rate and size. These droplets can be collected in a specific pattern to produce a near-net-form end product or part. The quality of the end product is governed by the uniformity of the droplets and the ability to predict droplet behavior after droplet generation. Most of the research performed at MIT to date has focused on single jet sprays. Further development of this technology requires the use of multiple sprays to meet increasing mass flux requirements for various industrial applications.
In this research, an existing simulation model's ability to predict deposit geometry was evaluated for multiple jet sprays. This was accomplished by capturing deposits and comparing them to the output of the simulation. The existing model, developed by Godard Abel (Abel, 1993), has the capacity to predict droplet flight trajectory, thermal state, and deposition geometry. The model calculates droplet flight trajectory by summing the theoretical forces on the droplets the simulation generates, and the simulation tracks the location where the droplets land. This model is accurate for cases where the liquid metal jets are perfectly normal to the plane of the crucible bottom plate, but when an array of jets that are not normal to the crucible bottom plate is employed, the original simulation lacked the capacity to fully characterize the initial spray conditions.
Abel's simulation model was modified to allow for a more thorough
characterization of the angular divergence of the initial jet trajectory
with respect to the crucible bottom plate. The original simulation allowed
axial variation from the normal vector of the crucible bottom plate as
an angle of elevation, but only in one preset azimuthal direction. The
modifications added a degree of freedom that permits initial trajectories
with any angle of elevation and any azimuthal angle. The axial divergence
for an orifice configuration was found empirically, and the simulation
inputs were adjusted accordingly. The output of the modified simulation
more accurately predicted the experimental deposit geometry.