When using vibration analyses to determine the current state of a piece of equipment, multiple spectra are used to accurately diagnose any potential problems.  However, with the technology we have available to us today, a filtered signal is most frequently the one that is observed.  While able to generate a quick and accurate plot in the velocity or acceleration spectra, these spectra do not always tell the whole story.

The time waveform is the cumulative collection of all frequencies that are present in a given machine over time.  Whether related to a bearing failure, structural looseness, or even gear wear, impacts in the time waveform can be broken down into their individual elements.  Previously, this was the only method available to collect data, but it has been overtaken by filtering and FFT spectra which can create a more user friendly image in a shorter amount of time.  Despite this, time waveform can still be the most accurate and telling spectra for diagnosing problems, especially in more complex situations.

The typical time waveform sets up a spectrum that shows vibration amplitudes over a specific amount of time.  Usually this is measured in inches per second, or IPS, versus a small amount of time (approximately ten cycles of the unit being measured).  It is in these ten cycles that impacts or changes in signature can be identified based on their frequency and amplitude.  Some good opportunities to use time waveform include but are not limited to:

  • Gearboxes – The inherent nature of a gearbox is to slow down or speed up an input speed.  Over the course of one revolution of the input shaft, impacts can be observed and identified to diagnose problems like cracked, broken or deformed gear teeth.
  • Rolling element bearings (slow speeds) – Often, bearings operating at less than 300 RPM can be difficult to analyze without a very long collection time needed to filter out ambient noise.  The time waveform allows for analysis of these bearings in real time by breaking it down into impacts per revolution(s).  This can help to identify lubrication deficiencies or bearing defect frequencies that may not be visible in a filtered spectrum.
  • Motor start up/ramp down – Time waveform helps to identify internal motor problems or resonances because it takes data over time versus a snapshot or a series of averages like a regular velocity measurement.  This can be helpful during motor start up or shut down because this is the time when the frequencies that are being generated are changing the most.  Problems such as rotor or winding problems can be observed during these critical periods.  This data allows for system setting adjustments to be made to minimize the affect these frequencies have on the life of the unit.

Time waveform can also be used to help confirm previously identified anomalies from the spectral analyses as well.  As helpful as time waveform can be, it should not be used over spectral analysis, or vice versa.  However, using both in accordance with each other can add another layer of confidence and validity to the evaluation of any piece of equipment exhibiting synchronous motion.