An Analysis of Noise Characteristics of Drill Bits

There have been papers that analyze the relationship between bit design and a bit’s vibrational characteristics. These papers typically are based on the analysis of three-axis near-bit down-hole vibration sensors. In this paper, the authors take a simpler approach. Using a standard microphone literally pointed at the bit, they record the noise of the bit/rock interaction while drilling and analyze the resulting noise for these but vibrational characteristics. The data were gathered at the Colorado School of Mines in Golden, Colorado. The noise of a PDC core, roller cone, and diamond core bits were recorded under various weight and rotary speeds using a microphone and a vertically mounted uniaxial geophone (used for confirming the data recorded on the microphones). Using a Fast Fourier Transform, the frequency spectra were extracted from the recorded data and analyzed. The data was normalized for rotational speed. The results of the frequency analysis of the roller cone, the PDC, and the natural diamond bits are presented in this paper. The major differences in the three bit frequency characteristics could be detected and furthermore, for drag bits, the frequency characteristics could be related to the bit’s design. The frequency spectra of the roller cone bit can best be described with a general high amplitude level that is relatively evenly distributed over the whole frequency spectrum. The drag bit data showed a strong relationship between the number and arrangement of cutting elements and frequency peaks on a plot of amplitude vs. cycles per revolution. Frequency peaks were observed at multiples of the number of cutting elements. In general this relationship was strongly visible on the PDC bit data but not as strongly visible on the diamond bit data. The conclusion is that bit characteristics can be determined using only the noise of a bit. Potential applications of this research include detecting and diagnosing bit problems (e.g. broken teeth, bit balling) in real time using simple microphone based acoustic data.