Thesis (Ph. D.)--Ohio State University, 2007. / Title from first page of PDF file. Includes bibliographical references (p. 131-136).

Thesis (M. S.)--Mechanical Engineering, Georgia Institute of Technology, 2009. / Committee Co-Chair: Bement, Matt; Committee Co-Chair: Liang, Steven; Committee Member: Griffin, Paul; Committee Member: Mayor, Rhett; Committee Member: Melkote, Shreyes; Committee Member: Zhou, Chen.

Analytical force and dynamic models for material removal processes such as end and face milling do not account for material and process related uncertainties such as tool wear, tool breakage and material inhomogeneity. Optimization of material removal processes thus requires not only optimal process planning using analytical models but also on-line monitoring of the process so that adjustments, if needed, can be initiated to maximize the productivity or to avoid damaging expensive parts. In this thesis, a Polyvinylidene Fluoride (PVDF) sensor based process monitoring method that is independent of the cutting conditions and workpiece material is developed for measuring the cutting forces and/or torque in milling. The research includes the development of methods and hardware for wireless acquisition of time-varying strain signals from PVDF sensor-instrumented milling tools rotating at high speeds and transformation of the strains into the measurand of interest using quantitative physics-based models of the measurement system. Very good agreement between the measurements from the low cost PVDF sensors and the current industry standard, piezoelectric dynamometer, has been achieved. Three PVDF sensor rosettes are proposed for measuring various strain components of interest and are shown to outperform their metal foil strain gauge counterparts with significantly higher sensitivity and signal to noise ratio. In addition, a computationally efficient algorithm for milling chatter recognition that can adapt to different cutting conditions and workpiece geometry variations based on the measured cutting forces/torque signals is proposed and evaluated. A novel complex exponential model based chatter frequency estimation algorithm is also developed and validated. The chatter detection algorithm can detect chatter before chatter marks appear on the workpiece and the chatter frequency estimation algorithm is shown to capture the chatter frequency with the same accuracy as the Fast Fourier Transform (FFT). The computational cost of the chatter detection algorithm increases linearly with data size and the chatter frequency estimation algorithm, with properly chosen parameters, is shown to perform 10 times faster than the FFT. Both the cutting forces/torque measurement methodology and the chatter detection algorithm have great potential for shop floor application. The cutting forces/torque measurement system can be integrated with adaptive feedback controllers for process optimization and can also be extended to the measurement of other physical phenomena.

Turning is one of the most widely used processes in machining and is characterized by a cutting tool moving along the axis of a rotating workpiece as it removes material. A detrimental phenomenon to productivity in turning operations is unstable cutting or chatter. This can reduce the life of tooling, dimensional accuracy, and the quality of a part's surface finish because of severe levels of vibration. Ideally, cutting conditions are chosen such that material removal is performed in a stable manner. However, it is sometimes unavoidable because of the geometry of the cutting tool or workpiece. This work seeks to develop a dynamic cutting force model that can be used to predict both the point of chatter instability as well as its amplitude growth over time. Previous chatter models fail to capture the physics of the process from a first-principles point of view because they are oversimplified and rely on various "cutting force coefficients" that must be tuned in order to get a desired correlation with experimental results. The proposed approach models the process in a geometrically rigorous fashion, also giving treatment to the strain, strain rate, and temperature effects encountered in machining. It derives the forces encountered during a turning operation from two sources: forces due to chip formation and forces due to plowing and flank interference. This study consists of a detailed derivation of two new cutting force models. One relies on careful approximations in order to obtain a closed-form solution; the other is more explicit and obtains a solution through numerical methods. The models are validated experimentally by comparing their prediction of the point of instability, the magnitude of vibration in the time and frequency domains, as well as the machined surface topography during chatter.