An investigation has been completed to determine the feasibility of using vibration measurement to develop an online tool wear monitoring system.Conventionaltransducers and FFT signal analysers have been used as the starting point for data collection and analysis. Additional software has been developed in order to obtain additional and enhanced analysis using statistical data such as percentile analysis and 4th statistical moments. The signal analysers have been interfaced with an IBM compatible PC to allow efficient data collection and analysis. • The measurement of surface finish via the use of centre line average has been selected as the primary indicator of tool wear. An experimental programme has been completed which examines the relationship between vibration measurement and surface finish during turning operations. This has provided sufficient general rules and guidelines to enable the method to be extended to other processes. It also provides information from which commonly occurring faults such as looseness of the tool holder, changes in material properties and swarf presence in the toolholder can be identified. Tests have been completed using cast iron, ENS and EN3 steels. An expert system has been developed. It has been demonstrated that sufficient data can be collected during the setting-up stage or commissioning part of a machining process from which, using the expert system, reliable tool wear monitoring can be achieved for all subsequent tests. In addition to predicting'the surface for all stages of tool life, the system can identify common faults such as looseness of the toolholder, the presence of swarf in the toolholder and changes in material properties. This system has been evaluated in relation to the available data, the present expectation of quality assesment required from the machine operation and relative to available commercial monitors. The system offers significant improvements.

In manufacturing, monitoring machine health is an important step when implementing Industry 4.0 and ensures effective machining operations and minimal downtime. Monitoring the health of cutting tools during a machining process helps contain the faults associated with gradual tool wear, because they can be tracked and responded to as wear worsens. Left unchecked, tool failures can lead to more severe problems, such as dimensional and surface issues with machined workpieces and lower overall productivity during the machining process. This research explores the use of a machine vision setup used internally by the McMaster Manufacturing Research Institute (MMRI) in their three lathe machines. This machine vision setup provides a direct indication of the tool's maximum flank wear (VBmax), which, according to ISO 3685:1993(E), is set to be 300 µm. Also investigated was the use of image processing and analysis methods to determine the flank wear without removing the tool from the machine. This new, in-machine vision setup is intended to replace the use of an external optical microscope, which requires extended downtime between cutting passes. As a result of this replacement, the experimentation downtime was decreased by around 98.6%, leading to the experiment time to decrease from 5 weeks or more to just a couple of days. In addition, the difference in measurement between a commonly used optical microscope and in-machine vision setup was found to be ±3µm. / Thesis / Master of Science (MSc)

A viabilização do conceito de manufatura enxuta depende da redução da necessidade de intervenções do operador no processo. Na usinagem dos metais, uma das principais situações onde está a intervenção se faz necessária é a determinação da condição de fim de vida das ferramentas de corte, a qual é muitas vezes subjetiva e não pode ser prescrita com segurança. Sendo a emissão acústica (EA) um sinal gerado pelos principais fenômenos envolvidos na usinagem, um sistema de monitoramento de desgaste de ferramentas baseado neste tipo de sinal foi avaliado em operação de torneamento de eixo de saída de transmissão mecânica caracterizado por geometria complexa. Verificou-se que há uma relação entre a trajetória da ferramenta e a \"assinatura\" do processo (tensão RMS plotada contra o tempo). Além disso, há uma forte correlação destes mais a adesão de material da peça à ferramenta determinaram pequena diferença de amplitude do sinal entre os estágios iniciais e finais de vida das ferramentas. Além disso, a influência do sistema de fixação na EA mostrou-se diversas vezes maior que a influência do desgaste. Portanto, um sistema de monitoramento de desgaste baseado apenas em emissão acústica não seria adequado para a situação estudada. / The Lean Manufacturing concept depends on the reduction of the necessity of interventions in the process by the operators. ln metal machining, one situation where these interventions are necessary is the determination of tool life, which is subjective. Besides that, the toollife prediction is not reliable. Acoustic emission signals are generated by many activities involved in machining. Therefore an acoustic emission based tool wear monitoring system was evaluated in a turning operation of a mechanical transmission main shaft characterised by complex geometry. lt was found that there is a relationship between the tool track and the AE \"signature\" of the process (RMS voltage plotted versus time). Furthermore, there is a strong correlation between crater and flank tool wear and the AE signal, but the combination of them plus adhesion phenomena determine low amplitudes between initial and final stages of the tool life. Besides that, the influence of the fixture system was found to be several times higher than the influence of the wear. Therefore, a tool wear monitoring system based on acoustic emission only would not be adequate for this application.

A viabilização do conceito de manufatura enxuta depende da redução da necessidade de intervenções do operador no processo. Na usinagem dos metais, uma das principais situações onde está a intervenção se faz necessária é a determinação da condição de fim de vida das ferramentas de corte, a qual é muitas vezes subjetiva e não pode ser prescrita com segurança. Sendo a emissão acústica (EA) um sinal gerado pelos principais fenômenos envolvidos na usinagem, um sistema de monitoramento de desgaste de ferramentas baseado neste tipo de sinal foi avaliado em operação de torneamento de eixo de saída de transmissão mecânica caracterizado por geometria complexa. Verificou-se que há uma relação entre a trajetória da ferramenta e a \"assinatura\" do processo (tensão RMS plotada contra o tempo). Além disso, há uma forte correlação destes mais a adesão de material da peça à ferramenta determinaram pequena diferença de amplitude do sinal entre os estágios iniciais e finais de vida das ferramentas. Além disso, a influência do sistema de fixação na EA mostrou-se diversas vezes maior que a influência do desgaste. Portanto, um sistema de monitoramento de desgaste baseado apenas em emissão acústica não seria adequado para a situação estudada. / The Lean Manufacturing concept depends on the reduction of the necessity of interventions in the process by the operators. ln metal machining, one situation where these interventions are necessary is the determination of tool life, which is subjective. Besides that, the toollife prediction is not reliable. Acoustic emission signals are generated by many activities involved in machining. Therefore an acoustic emission based tool wear monitoring system was evaluated in a turning operation of a mechanical transmission main shaft characterised by complex geometry. lt was found that there is a relationship between the tool track and the AE \"signature\" of the process (RMS voltage plotted versus time). Furthermore, there is a strong correlation between crater and flank tool wear and the AE signal, but the combination of them plus adhesion phenomena determine low amplitudes between initial and final stages of the tool life. Besides that, the influence of the fixture system was found to be several times higher than the influence of the wear. Therefore, a tool wear monitoring system based on acoustic emission only would not be adequate for this application.

Fundação de Amparo a Pesquisa do Estado de Minas Gerais / The main goal of this work is the development of techniques for application of acoustic emission signal in tool wear monitoring in face milling operation. In this work a Sensis (DM 42) equipment for acquisition of the acoustic emission signal was used during the milling of a high strength low alloy steel (Din 38MnS6). A milling cutter with 125 mm diameter for eight inserts with specification R245 125Q40-12M was used. The ISO specification of the inserts were SEMN 12 04 AZ TiN coated. The tests were divided into two parts. In the first part, wear and some other machining phenomena were monitored through the construction of a luminous intensity map. For these tests it was used one, two, four or eight inserts simultaneously. In this stage the acoustic emission signal was evaluated using the RMS values. In the second part of this work the acoustic emission raw signal was used. Amongst the statistical parameters that correlate to tool wear extracted from the raw signal, the best fit ones were selected to train and validate a Probabilistic Neural Network. The results of the PNN indicate that the acoustic emission signal can be used to recognition of tool wear levels in the milling process. / O principal objetivo deste trabalho é o desenvolvimento de técnicas para a aplicação da emissão acústica no monitoramento de desgaste de insertos de metal duro no fresamento frontal. Neste trabalho foi utilizado um equipamento da Sensis (DM 42) para aquisição do sinal de emissão acústica durante o fresamento frontal de um aço microligado perlítico de forja (Din 38MnS6). Foi utilizada uma fresa com especificação R245 125Q40-12M, de 125 mm de diâmetro e com possibilidade de montagem de até oito insertos. Os insertos utilizados são de metal duro R245-12 T3 M-PM 4030 revestidos com TiN. A fresa e os insertos foram fabricados pela Sandvik Coromant. Os testes foram divididos em duas partes. A primeira consistiu no monitoramento do desgaste e também de alguns fenômenos que ocorrem na usinagem através da construção de um mapa de intensidade luminosa. Nestes testes foram utilizados um, dois, quatro e oito insertos simultaneamente. Nesta etapa foi empregado o sinal RMS da emissão acústica. Na segunda etapa dos ensaios utilizou-se o sinal bruto de emissão acústica. Foram estimados parâmetros estatísticos destes sinais e destes foram escolhidos os que melhor se relacionavam com o fenômeno estudado. Este procedimento foi realizado através da ferramenta estatística boxplot. Finalmente, os parâmetros escolhidos foram usados para treinar e validar uma Rede Neural Probabilística, que teve a função de reconhecer e classificar os estágios de desgaste da ferramenta. Os índices de acerto nas classificações da rede indicam que a emissão acústica apresenta bons resultados no reconhecimento de níveis de desgaste das ferramentas na operação de fresamento. / Doutor em Engenharia Mecânica