Regenerative chatter, which is a well-known unstable phenomenon in macro-scale machining, can also occur in micro milling and leads to insufficient quality of products, premature tool wear, and tool damage. To predict the chatter stability, the dynamics of the machine tool system should be identified either by experiments or by (semi-)analytical models. Due to the process-machine-interactions, it is usually insufficient to treat the micro end mills as an oscillator or as a cantilever beam for chatter prediction. In this dissertation the systematic modeling of a micro milling machine tool system is presented considering the micro end mills, the spindle–tool holder–tool assembly, and other fundamental components. The frequency response functions at the tool tip point and the mode shapes of the machine tool structure are discussed to gain a deep insight into the system characteristics. The possibilities of model reduction are also given to improve the computational efficiency with acceptable accuracy.
For analytical stability prediction, simplified cutting force models based on the conventional milling are adopted for high feed rates. The stability behavior and chatter frequencies are discussed by analytical and experimental results regarding diverse influencing factors, such as the rotational effect, the tool clamping condition, the process damping effect, and the tool geometry. Furthermore, a new measuring method for online chatter detection is presented by application of piezoelectric elements. During the cutting processes, chatter frequencies can be identified with an axial depth of cut lower than the actual stability boundary.