Utilizando a vibração mecânica para monitorar o desgaste das ferramentas de corte e o acabamento superficial no processo de fresamento

• Ano de defesa: 1998
• Autor(a) principal: Sousa, Marcos Morais de
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Uberlândia; Brasil; Programa de Pós-graduação em Engenharia Mecânica
• Palavras-chave em Português: Desgaste de ferramenta; Acabamento superficial; Monitoramento; Vibraçao mecânica; Wear in tools; Surface finishing; Mechanical vibration; Monitoring; CNPQ::ENGENHARIAS::ENGENHARIA MECANICA
• Link de acesso: https://repositorio.ufu.br/handle/123456789/26694; http://dx.doi.org/10.14393/ufu.te.1998.1

This work shows a study on the viability o f the application of the mechanical vibration signals generated during the milling process to quantify the evolution of a cutting tool flank wear and to estimate the surface fmish of a milling tool. It is presented: the techniques used for real time monitoring, some definitions on milling, surface finishing, and the signal analysis techniques in the domain of time and frequency. Due to the fact that the milling machine presents variations in its vibration properties with the modifícation o f geometry during the work, a study on the dynamic sensitivity has been perfoimed in order to identify regions with less sensitivity to this variation in geometry, and such study was accomplished with the application o f Frequency Response Function. With the signals obtained during the monitoring of the milling passes on a AISI 1045 Steel bar, it has been realized that the parameters RMS, peak and kurtosis show good results for the quantification of the evolution of the wear of the cutting tool with the levei of vibration. The relation between work piece surface flnish and mechanical vibration has been obtained by using a third degree polynomial with coeffícients adjusted through common least squares. In the frequency domain, the evolution o f the wear of the cutting tolls can be detected through: spectral analysis, ceptral analysis, spectral analysis using Hiibert transforms and one third octave filters.