The Time Waveform: The Foundation of Vibration Analysis

Definition: What is a Time Waveform?

The time waveform (also known as the time domain signal) is the raw, unprocessed signal from a vibration transducer, such as an accelerometer or proximity probe. It is a graph that plots the instantaneous amplitude of the vibration on the vertical (Y) axis against time on the horizontal (X) axis. It is a direct representation of the physical back-and-forth movement of the machine at the sensor location over a short period.

The Role of the Time Waveform in Diagnostics

While the frequency spectrum (FFT) is the primary tool for diagnosing most steady-state machinery faults, the time waveform is an indispensable and complementary tool. The FFT calculates the frequency content *averaged* over the time of the sample. In doing so, it can sometimes obscure short-duration, transient, or non-periodic events. The time waveform, however, shows exactly what happened from moment to moment, making it superior for analyzing:

• Impulsive Events: It clearly shows sharp impacts, which are often the first sign of bearing or gear defects.
• Modulation and Beats: The classic rise-and-fall pattern of beating is most clearly seen in the time waveform.
• Transient Events: It can capture random, one-time events that would be averaged out in an FFT.
• Signal Clipping: It immediately shows if the sensor signal has exceeded the input range of the analyzer, which would invalidate the FFT.
• Rubs: The sharp, distorted signals from a rotor rub are often most obvious in the waveform.

A skilled analyst always reviews both the spectrum and the time waveform to get a complete picture of the machine’s health.

How to Analyze a Time Waveform

Analyzing a time waveform involves examining its shape and key characteristics.

1. Peak Amplitude

The maximum amplitude (Peak) in the waveform is a direct measure of the maximum force or stress of an event. A high peak amplitude in an otherwise low-energy signal is a strong indicator of impacts.

2. Overall Shape

A healthy, well-balanced machine will often have a clean, sinusoidal waveform at the running speed frequency. Distortions in this shape indicate the presence of other frequencies or forces. A “flattened” or “clipped” appearance, for example, is a classic sign of mechanical looseness where the component’s motion is being constrained.

3. Repetitive Patterns and Periodicity

By placing cursors on the plot, an analyst can measure the time between repeating events.

• Measuring the time between major peaks gives the period of the fundamental vibration, which can be inverted to find its frequency (Frequency = 1 / Period).
• Looking for smaller, repetitive impacts “riding” on the main waveform can help identify the precise frequency of bearing or gear faults.

4. Statistical Parameters

Statistical values calculated from the time waveform are powerful diagnostic indicators:

• RMS (Root Mean Square): Measures the overall energy content of the signal.
• Crest Factor: The ratio of the Peak amplitude to the RMS value. A high crest factor (>>3) indicates the presence of strong impacting.
• Kurtosis: A measure of the “peakedness” of the signal. A high kurtosis value is highly sensitive to early-stage bearing faults.

Waveform vs. Spectrum: A Partnership

The time waveform and the frequency spectrum are two different views of the same data, and they work best together:

• The spectrum excels at separating multiple, closely-spaced, steady-state frequencies.
• The waveform excels at showing the true amplitude of impacts and the nature of non-steady-state events.

For example, the spectrum might show a raised noise floor, but the waveform will reveal that the cause is a series of low-amplitude, repetitive impacts from a developing bearing fault.

← Back to Main Index