AN INVESTIGATION OF VARIABLE SPINDLE SPEED FACE MILLING FOR TOOL-WORKSTRUCTURES WITH COMPLEX DYNAMICS .1. SIMULATION RESULTS

Citation
R. Radulescu et al., AN INVESTIGATION OF VARIABLE SPINDLE SPEED FACE MILLING FOR TOOL-WORKSTRUCTURES WITH COMPLEX DYNAMICS .1. SIMULATION RESULTS, Journal of manufacturing science and engineering, 119(3), 1997, pp. 266-272
Citations number
20
Categorie Soggetti
Engineering, Mechanical","Engineering, Manufacturing
ISSN journal
10871357
Volume
119
Issue
3
Year of publication
1997
Pages
266 - 272
Database
ISI
SICI code
1087-1357(1997)119:3<266:AIOVSS>2.0.ZU;2-N
Abstract
A mechanistic dynamic model is used to simulate a face milling process during constant and variable speed machining. The model can be used t o predict the optimum speed trajectory that can provide a low level of vibration and consequently a large productivity rate and a small surf ace error. The model is used to investigate the vibration efface milli ng processes that have one, or multiple coupled modes of vibration act ing throughout the cut. For cutting processes having one dominant mode of vibration, the model predicts that variable speed machining is esp ecially effective over constant speed machining when the tool-work sys tem changes its dominant mode of vibration throughout the cut, or when the tool-work system has several modes of vibration coming from compo nent parts that are cut in the same time. For cutting processes having multiple dominant modes of vibration, the model predicts that variabl e speed machining is especially effective over constant speed machinin g when the tool-work modes of vibration are unequal and moderately cou pled to each other. Also, the model suggests that for tool-work system s having complex geometries with dynamics hard to predict, variable sp eed machining is safer to use than constant speed machining when tryin g to achieve high productivity rates. This is due to the fact that var iable speed machining is robust with respect to the dynamics of the to ol-work system. Finally, the model predictions are in good agreement w ith the experiment.