Acoustic monitoring of cutting conditions to detect tool break events

A system for detecting cutting tool break events while machining a workpiece comprising:

• a broadband accelerometer which is most sensitive to frequencies around a resonant frequency and is positioned on a machine tool to sense vibrations at the tool-workpiece interface and convert these and other vibrations to an electrical signal;
• analog signal processing means including means for high-pass filtering said vibration signal to discriminate against lower frequency machinery noise, and means to rectify and low-pass filter said signal to detect the signal energy in an acoustic frequency band between 30 KHz and 100 KHz;
• means for sampling the unipolar output signal of said analog processing means and converting each sample to digital form;
• and
• digital pattern recognition circuitry to detect changes in cutting noise due to tool break events that can damage the workpiece or machine tool;
• said digital circuitry comprising means for calculating the running mean signal level of a selected number of samples, a transient detector to compare each new sample with the running mean and detect abrupt increases and decreases in signal level that may be the result of such a tool break event, and means for testing for a persistent shift in mean signal level that remains outside amplitude acceptance limits for a preset confirmation period after triggering by said transient detector, and generating a tool break alarm.

The method of detecting tool breakage capable of damaging the workpiece or machine tool comprising:

• sensing vibrations at the tool-workpiece interface and converting these and other machine tool vibrations to an electrical signal;
• preprocessing said vibration signal to discriminate against lower frequency machinery noise and rectify and low-pass filter said signal to detect the signal energy in an acoustic frequency band below 100 KHz;
• sampling the processed vibration signal and converting each sample to digital form;
• and
• analyzing said signal samples in only two basic phases to detect changes in cutting conditions due to tool break events capable of damage, the first phase comprising detecting an abrupt increase or decrease in signal level that may be a cutting noise change by comparing every sample with the running mean signal level of N previous samples, the second phase comprising testing for a shift in cutting noise that persists for a minimum confirmation period by continually comparing the mean signal level against amplitude acceptance limits, and generating a tool break alarm when outside said limits, otherwise dismissing and returning control to the first phase.