Understanding Spectral Analysis

Devices

Portable balancer & Vibration analyzer Balanset-1A

€1,975.00 + VAT (if applicable)

Parts

Vibration sensor

€90.00 + VAT (if applicable)

Parts

Optical Sensor (Laser Tachometer)

€124.00 + VAT (if applicable)

Devices

Balanset-4

€6,803.00 + VAT (if applicable)

Parts

Magnetic Stand Insize-60-kgf

€46.00 + VAT (if applicable)

Parts

Reflective tape

€10.00 + VAT (if applicable)

Devices

Dynamic balancer “Balanset-1A” OEM

€1,735.00 + VAT (if applicable)

Spectral analysis is the process of taking a complex signal — such as a time waveform from a vibration sensor — and decomposing it into its individual frequency components. Its goal is to transform the signal from the time domain (amplitude versus time) into the frequency domain (amplitude versus frequency). The output is a spectrum (plural: spectra): a graph showing the amplitude of vibration at each specific frequency. Spectral analysis is the most fundamental and powerful technique in vibration diagnostics, because it lets an analyst recognise the unique frequency signatures associated with different machine faults.

1. Definition: What is Spectral Analysis?

A machine’s vibration is the sum of many simultaneous periodic forces, and in the time domain those forces blur into a single complicated trace that resists interpretation. Spectral analysis untangles that trace. The mathematical foundation is the Fourier principle: any periodic signal, however complex, can be represented as a sum of simple sine waves, each with its own frequency, amplitude, and phase. Spectral analysis recovers that list of sine waves and displays it as a spectrum — converting “a confusing wiggle over time” into “a tidy set of peaks at known frequencies.” That conversion is the whole point, and everything else follows from it.

2. The Role of the Fast Fourier Transform (FFT)

Modern spectral analysis is made practical by a highly efficient algorithm called the Fast Fourier Transform (FFT). The FFT is a mathematical procedure programmed into every digital vibration analyzer. It takes the digitised time-waveform data as its input and produces the frequency spectrum as its output, performing in milliseconds a computation that would once have taken hours by hand.

The FFT allows a complex, seemingly chaotic signal — very difficult to read in the time domain — to be viewed as a clear series of distinct frequency peaks in the frequency domain. For the FFT to be valid, the signal must first be sampled correctly and protected from aliasing, which is why an anti-aliasing filter always precedes the transform.

3. The Diagnostic Power of the Spectrum

Spectral analysis is so effective because different mechanical and electrical problems in rotating machinery generate vibration at predictable, characteristic frequencies. By examining the spectrum, an analyst can diagnose the root cause of a problem by matching the frequencies of the peaks to a known fault pattern. Common examples include:

• A large peak at 1× the running speed often indicates unbalance — the cue for a balancing job.
• A large peak at 2× the running speed is a classic sign of misalignment.
• A row of running-speed harmonics commonly indicates mechanical looseness.
• A series of peaks at non-integer multiples of running speed can point to bearing defects, often best confirmed with envelope analysis.
• A high-amplitude peak at the gear mesh frequency with sidebands indicates gear defects.
• A peak at 2× the electrical line frequency can indicate a motor stator problem.

4. Key Parameters in Spectral Analysis

To acquire a useful spectrum, an analyst must define several key parameters. Get these wrong and even a perfect FFT will mislead you:

• Fmax (maximum frequency): the highest frequency included in the spectrum. It must be set high enough to capture the suspected faults — high for gear and bearing problems, lower for slow-speed unbalance.
• Resolution (lines of resolution): this determines the level of detail. More lines give finer frequency resolution, letting the analyser separate two peaks that lie close together — essential for resolving sidebands in gearbox analysis. The trade-off is a longer acquisition time, a relationship you can explore with an FFT Resolution Calculator.
• Averaging: to obtain a clean, stable spectrum, the analyser takes several “snapshots” and averages them, suppressing random noise and revealing the true steady-state vibration.
• Windowing: a mathematical function such as a Hanning window is applied to each data block to prevent spectral leakage, which would otherwise smear and distort the amplitudes and shapes of the peaks.

5. Spectral Analysis in the Field

Spectral analysis is not confined to the laboratory — it is exactly what a maintenance engineer performs on site with a portable instrument. A two-channel analyser such as the Balanset-1A samples the signal from an accelerometer on the bearing housing, runs the FFT, and presents the spectrum on the spot, so the technician can identify a fault and act before the next shift. Because the same instrument also reads shaft phase from a tachometer, a diagnosis of 1× unbalance can flow straight into a balancing correction without changing tools — diagnosis and repair in one workflow.

6. Spectral Analysis vs the Time Waveform

The spectrum is powerful, but it is not the whole picture. Averaging and the FFT can hide short, sharp transient events — the impacts of a cracked gear tooth or a spalled bearing — that stand out clearly in the raw time waveform. Skilled analysts therefore treat the two views as partners: the spectrum to locate which frequencies carry energy, the time waveform to study how that energy is delivered. In essence, spectral analysis is the core of modern vibration diagnostics, giving a clear “X-ray” view of the forces and movements happening inside a machine — and read alongside the waveform and a sound baseline reference, it turns raw vibration into confident decisions.

← Back to Main Index