Date of Award
Program or Major
Master of Science
It is well known that cutting forces increase as a cutting tool wears out. Current commercially available Tool Condition Monitoring (TCM) systems use this fact to set a threshold for the allowable percent power increase corresponding to a worn out tool. Typically, a training process is required whereby the first few parts cut with a sharp tool provide a baseline for comparison. This method works well when making large numbers of the same part with a single tool but is not as useful when making a single part or when the tool life is shorter than the time required to make a single part.
The goal of this research is to develop a TCM system that can estimate tool wear without the necessity of a training process and in the presence of different modes of tool wear. Our hypothesis is that tool wear can be accurately correlated with the coefficients of a tangential cutting force model. The model coefficients are estimated by online measurement of spindle motor power. One of the model coefficients (Ktc) relates cutting forces to chip thickness. The other coefficient (Kte) relates to rubbing or edge forces. Research results indicate that Kte correlates well with flank wear while Ktc is a good indicator of edge chipping.
A number of experiments are performed with helical flat end mills made of High Speed Steel (HSS) and carbide tools cutting 1018 steel. Spindle motor power is used to calibrate the model coefficients at periodic intervals. Generally, the percent power increase at tool failure is much higher when flank wear is the dominant mode. Tool wear is estimated based on a weighted combination of the two model coefficients thereby establishing a threshold that is independent of the combination of chipping and flank wear.
Preliminary research was also performed to explore the potential of using a contact microphone mounted to the spindle for TCM. A contact microphone can provide frequency information based on the vibrations as well as relate the RMS value to cutting power.
Desfosses, Bennett, "An improved power threshold method for estimating tool wear during milling" (2007). Master's Theses and Capstones. 265.