Abstract: A computer-implemented process determines, based on bearing fault frequencies, optimum values for the maximum frequency (Fmax) and the number of lines of resolution (Nlines) to be used in collecting machine vibration data so as to adequately distinguish between spectral peaks for identifying faults in machine bearings. The process can be extended to any other types of fault frequencies that a machine may exhibit, such as motor fault frequencies, pump/fan fault frequencies, and gear mesh fault frequencies. Embodiments of the process also ensure that the time needed to acquire the waveform is optimized. This is particularly useful when collecting data using portable vibration monitoring devices.

Abstract: A statistical method is used to separate periodic from non-periodic vibration peaks in machine vibration spectra. Generally, a machine vibration spectrum is not normally distributed because the amplitudes of periodic peaks are significantly large and random relative to the generally Gaussian noise. In a normally distributed signal, the statistical parameter Kurtosis has a value of 3. The method sequentially removes each largest amplitude peak from the peaks in a frequency region of interest in the spectrum until the Kurtosis has a value of three or less. The removed peaks, which are all considered to be periodic, are placed into a candidate peak list. As the process of building the candidate peak list proceeds, if the kurtosis of the remaining peaks in the frequency region of interest falls to three or less, the process stops and the candidate peak list is defined.

Abstract: A computer implemented method processes time waveform machine vibration data that are indicative of operational characteristics of a machine. The data, which were measured on the machine over a period of time having a begin time and an end time, are accessed from a memory or storage device. An integer number M of waveform samples are determined from the data to be averaged, and an asymptotically decaying DC bias component in the data is derived using a moving average of the M number of waveform samples. The DC bias component is extrapolated from the begin time of the waveform back to an earlier time and from the end time of the waveform forward to a later time. The DC bias component is then subtracted from the time waveform data, and a Fast Fourier Transform is performed on the data to generate a spectrum.

Abstract: A computer-implemented process determines, based on bearing fault frequencies, optimum values for the maximum frequency (Fmax) and the number of lines of resolution (Nlines) to be used in collecting machine vibration data so as to adequately distinguish between spectral peaks for identifying faults in machine bearings. The process can be extended to any other types of fault frequencies that a machine may exhibit, such as motor fault frequencies, pump/fan fault frequencies, and gear mesh fault frequencies. Embodiments of the process also ensure that the time needed to acquire the waveform is optimized. This is particularly useful when collecting data using portable vibration monitoring devices.

Abstract: A method for detecting defects in a rotational element of a machine based on changes in measured vibration energy includes: (a) collecting vibration data over an extended period of time using vibration sensors attached to the machine; (b) processing the vibration data to generate a time waveform comprising processed vibration values sampled during sequential sampling time intervals within the extended period of time; (c) detecting multiple time blocks within the extended period of time during which the processed vibration values exhibit sustained increases at progressively increasing rates; and (d) generating alerts based on detection of the multiple time blocks during which the processed vibration values exhibit sustained increases at progressively increasing rates.

Abstract: An apparatus is described that determines an estimated rotational speed of a rotating component of a machine in the absence of a reliable tachometer signal to indicate an actual rotational speed.

Abstract: An apparatus is described that determines an estimated rotational speed of a rotating component of a machine in the absence of a reliable tachometer signal to indicate an actual rotational speed. The apparatus includes a processor that produces a spectral plot of the vibrational data, locates peaks in the spectral plot, and scans the spectral plot in predetermined rotational speed increments to provide candidate rotational speeds. For each candidate rotational speed, associated harmonics are identified, closest peaks in the spectral plot to the candidate rotational speed and its harmonics are located, gaps between the closest peaks and the candidate rotational speed and its harmonics are measured, and a sum of the gaps is recorded. The estimated rotational speed is the candidate rotational speed associated with a minimum sum of the gaps.

Abstract: A vibration analyzer for use in determining a rotational speed. The vibration analyzer includes an input for sensing vibration data, a memory for storing the vibrational data, and a processor. The processor produces a spectral plot of the vibrational data, locates peaks in the spectral plot, receives an input rotational speed, and scans the spectral plot in predetermined rotational speed increments to provide a candidate rotational speeds. For each candidate rotational speed, a number of associated harmonics is identified, closest peaks in the spectral plot to the candidate rotational speed and its harmonics are located, gaps between the closest peaks and the candidate rotational speed and its harmonics are measured and summed, and a sum of the gaps is recorded. The candidate rotational speed that is associated with the maximum peak sum is selected as the nominal rotational speed.

Abstract: A vibration data collection device monitors vibration of a machine, generates original machine vibration waveform data based on the monitored vibration, and removes portions of the original machine vibration waveform data that do not indicate an occurrence of a vibration event related to a potential fault in the machine. The vibration data collection device then stores thinned waveform data that includes blocks of the original machine vibration waveform data and excludes the portions that have been removed. A data analysis computer generates a thinned waveform plot based on the thinned waveform data. In some embodiments, the thinned waveform plot includes the blocks of original machine vibration waveform data separated in time by gaps representing the portions that have been removed.

Abstract: A vibration measurement and analysis system identifies faulty bearings in a machine based on spectral vibration data. The system includes vibration sensors attached to the machine that generate vibration signals. A vibration data collector generates vibration spectral data based on the vibration signals. The vibration spectral data comprises vibration amplitude versus frequency data that includes peak amplitudes at corresponding peak frequencies. At least some of the peak amplitudes are associated with vibration generated by the faulty bearings.

Abstract: A computer-implemented method analyzes periodic information in digital vibration data associated with a machine. The method involves generating a spectral periodic information plot (PIP) based on the digital vibration data, and locating amplitude peaks in the PIP at frequencies associated with fundamental frequencies of interest. Peaks occurring at fundamental fault frequencies and at related harmonic frequencies are removed from the PIP, while retaining energy values associated with the removed peaks. Remaining peaks in the PIP are classified as synchronous periodic peaks and non-synchronous periodic peaks. The remaining peaks in the PIP are graphically plotted along with the fault frequencies and related harmonic frequencies in different colors or different line styles on a display device to identify different groups of frequencies of interest.

Abstract: A computer-implemented process determines, based on bearing fault frequencies, optimum values for the maximum frequency (Fmax) and the number of lines of resolution (Nlines) to be used in collecting machine vibration data so as to adequately distinguish between spectral peaks for identifying faults in machine bearings. The process can be extended to any other types of fault frequencies that a machine may exhibit, such as motor fault frequencies, pump/fan fault frequencies, and gear mesh fault frequencies. Embodiments of the process also ensure that the time needed to acquire the waveform is optimized. This is particularly useful when collecting data using portable vibration monitoring devices.

Abstract: A vibration data collection device monitors vibration of a machine, generates original machine vibration waveform data based on the monitored vibration, and removes portions of the original machine vibration waveform data that do not indicate an occurrence of a vibration event related to a potential fault in the machine. The vibration data collection device then stores thinned waveform data that includes blocks of the original machine vibration waveform data and excludes the portions that have been removed. A data analysis computer generates a thinned waveform plot based on the thinned waveform data. In some embodiments, the thinned waveform plot includes the blocks of original machine vibration waveform data separated in time by gaps representing the portions that have been removed.

Abstract: A data compression process reduces the amount of machine spectral data transmitted over a network while maintaining the details of spectral peaks used for machine health analysis. The data compression process also provides for the calculation of various types of spectral parameters, such as spectral band parameters, with negligible loss of accuracy.

Abstract: Spectral machine condition energy peaks are graphically represented in a spectral plot using color coding, different line types, and/or filtering. This allows visual differentiation of spectral peaks associated with various fault frequency families from one another, whereby a machine condition analyst using computer-based analysis software can easily see each family of spectral peaks individually, without all the other spectral peaks, or in combinations of families that are relevant to a machine fault under investigation. In addition to current spectral data, the analyst can also view a historical trend of related scalar parameters plotted in conjunction with current spectral data, wherein the spectral data plot is synchronized with a time-based cursor on the trend plot.

Abstract: A method for filling a gap of missing vibration data in a set of vibration data. At least one reference waveform on a first side of the gap is selected and at least one adjacent waveform on an opposing second side of the gap is selected. It is determined whether the gap is in a section of the vibration data where a frequency of the vibration data is one of increasing, decreasing, and steady state. Where the gap is in a section of the vibration data where the frequency of the vibration data is changing substantially linearly, an analytical method is applied to at least one of the at least one reference waveform and the at least one adjacent waveform to approximate the vibration data that is missing in the gap.

Abstract: A computerized apparatus for presenting vibration data that is read under different conditions as a single trend line on a chart. A sensor measures vibration and produces vibration data. A memory stores the vibration data, an indicator of the condition under which the vibration data was produced, and a type associated with the vibration data. A processor reads the vibration data, the condition, and the type from the memory, and selectively plots the vibration data on the trend line when the type of the vibration data matches a given value, even though the condition might be different from data point to data point on the trend line. The processor selectively creates a flag indicating a condition change, and an interface presents a plot of the trend line, and the flag when a condition change occurs between data points.

Abstract: A method for producing a thinned representation of vibration waveform data. The waveform vibration data is received and divided into sequential blocks. For each sequential block, each serially designated in turn as a current block, the following steps are performed. When the current block is also a first block, the current block is passed as a reference block. A representative value for the current block is computed and compared to the representative value for the reference block to determine a difference. The representative value for the current block is compared to a minimum representative value. The current block is transformed into a spectrum and compared to the spectrum for the reference block to determine a correlation value.

Abstract: A vibration sensor mounting structure improves the stability and repeatability of a measured signal generated by a machine vibration sensor. The structure has an outer annular surface that contacts the machine under test and a shallow recess inside the outer annular surface. The recess causes resonant vibrations of the mounting structure to occur at frequencies that are above the intended measurement range of the sensor. The recess also allows the mounting force to be positioned away from the center mounting screw and onto the more stable outer annular surface. With the mounting force away from the center, lateral forces have less effect on the measured signal.

Abstract: A data compression process reduces the amount of machine spectral data transmitted over a network while maintaining the details of spectral peaks used for machine health analysis. The data compression process also provides for the calculation of various types of spectral parameters, such as spectral band parameters, with negligible loss of accuracy.