5-axis double-flank CNC machining of spiral bevel gears via custom-shaped milling tools -- Part I: modeling and simulation

A new category of 5-axis flank computer numerically controlled (CNC) machining, called \emph{double-flank}, is presented. Instead of using a predefined set of milling tools, we use the shape of the milling tool as a free parameter in our optimization-based approach and, for a given input free-form (...

Descripción completa

Detalles Bibliográficos
Autores: Bo, P., Gonzalez, H., Calleja, A., Lopez de Lacalle, N., Barton, M.
Tipo de recurso: artículo
Estado:Versión aceptada para publicación
Fecha de publicación:2019
País:España
Institución:Basque Center for Applied Mathematics (BCAM)
Repositorio:BIRD. BCAM's Institutional Repository Data
OAI Identifier:oai:bird.bcamath.org:20.500.11824/1045
Acceso en línea:http://hdl.handle.net/20.500.11824/1045
Access Level:acceso abierto
Palabra clave:5-axis CNC machining
flank milling
custom-shaped tools
tangential movability
spiral bevel gears
free-form shape manufacturing
Descripción
Sumario:A new category of 5-axis flank computer numerically controlled (CNC) machining, called \emph{double-flank}, is presented. Instead of using a predefined set of milling tools, we use the shape of the milling tool as a free parameter in our optimization-based approach and, for a given input free-form (NURBS) surface, compute a custom-shaped tool that admits highly-accurate machining. Aimed at curved narrow regions where the tool may have double tangential contact with the reference surface, like spiral bevel gears, the initial trajectory of the milling tool is estimated by fitting a ruled surface to the self-bisector of the reference surface. The shape of the tool and its motion then both undergo global optimization that seeks high approximation quality between the input free-form surface and its envelope approximation, fairness of the motion and the tool, and prevents overcutting. That is, our double-flank machining is meant for the semi-finishing stage and therefore the envelope of the motion is, by construction, penetration-free with the references surface. Our algorithm is validated by a commercial path-finding software and the prototype of the tool for a specific gear model is 3D printed.