Development of a chatter detection methodology for variable spindle speed milling

Regenerative chatter is one of the most detrimental phenomena affecting machining operations both in terms of tool integrity and quality of the machined surface. Most of the developed chatter detection algorithms, being based on the separation of the spindle-related frequencies from the asynchronous frequency components, fails to be effective under non-stationary spindle speed conditions. In order to overtake this limitation, in this research a chatter detection algorithm suitable for a generic spindle speed regime was developed. More specifically, the algorithm is based on the order tracking methodology and allows to perform a process stability assessment through the definition of a suitable chatter indicator, in the spindle angular domain. The developed procedure was then validated performing real milling operations in which the spindle speed variation technique was adopted.