Vibration Diagnostics: Interpreting the Language of Machines

1. Definition: What is Vibration Diagnostics?

Vibration Diagnostics is an advanced form of condition monitoring where vibration data is not just collected, but is deeply analyzed and interpreted to determine the health of a machine and pinpoint the root cause of specific faults. It is the process of translating raw vibration signals into actionable maintenance information.

While vibration monitoring may track overall vibration levels, diagnostics focuses on the “why.” It seeks to answer questions like: Is this vibration caused by unbalance or misalignment? Is that bearing failing? Is there a problem with the gears?

2. The Diagnostic Process

A typical vibration diagnostics process follows a structured approach:

1. Data Acquisition: Collecting high-quality vibration data using sensors like accelerometers and a data analyzer. This involves selecting the right sensor, mounting it correctly (per ISO 5348), and choosing appropriate measurement settings (e.g., Fmax, resolution).
2. Signal Processing: Converting the raw time waveform signal into a more useful format, most commonly a frequency spectrum using the FFT (Fast Fourier Transform) algorithm. Other tools like phase analysis and enveloping are also used.
3. Spectral Analysis: This is the core of diagnostics. The analyst examines the frequency spectrum to identify specific patterns. Different machine faults generate energy at predictable frequencies. For example:
   • Unbalance: High amplitude at 1x the rotor’s running speed.
   • Misalignment: High amplitude at 1x and especially 2x running speed, often with high axial vibration.
   • Bearing Defects: Non-synchronous, high-frequency peaks at specific bearing fault frequencies (BPFO, BPFI, BSF, FTF).
   • Gear Defects: Peaks at the Gear Mesh Frequency (GMF) and its sidebands.
4. Fault Confirmation: Using multiple data types to confirm a diagnosis. For instance, analyzing the time waveform shape to look for impacting (indicative of bearing faults) or using phase analysis to differentiate between unbalance and a bent shaft.
5. Reporting and Recommendation: Clearly communicating the findings, including the identified fault, its severity, and a recommended course of action for maintenance personnel.

3. Key Tools and Techniques

Vibration diagnostics relies on a variety of specialized analytical tools:

• Spectrum Analysis (FFT): The primary tool for identifying the frequencies present in a signal.
• Time Waveform Analysis: Useful for observing signal shape, impacts, and modulating events that can be missed in the FFT.
• Phase Analysis: A crucial tool for confirming unbalance, misalignment, looseness, and performing balancing.
• Envelope Analysis (Demodulation): A technique for detecting the very low-energy, repetitive impacts associated with early-stage bearing and gear defects.
• Order Analysis: Used for variable-speed machines, it relates vibration to the multiples (orders) of the running speed rather than fixed frequencies.
• Operating Deflection Shape (ODS): An animation that shows how a machine or structure is moving at a specific frequency, useful for diagnosing resonance and structural weakness.

4. The Goal: From Reactive to Proactive

The ultimate goal of vibration diagnostics is to support a proactive maintenance strategy. By identifying the root causes of failure (like misalignment, resonance, or improper lubrication), organizations can move beyond simply fixing broken machines and begin to eliminate the conditions that cause them to fail in the first place, leading to significantly improved reliability and reduced costs.

← Back to Main Index