Vibration measurement and analysis is one of the most powerful diagnostic tools in a predictive maintenance (PdM) programme, capable of detecting developing faults in rotating machinery weeks or months before they cause unplanned failure. Significantly reducing maintenance costs, production losses, and safety risks. In Singapore's manufacturing, process, and utilities sectors, where equipment reliability directly impacts production efficiency and regulatory compliance, vibration-based condition monitoring has become an essential element of modern maintenance strategy. This guide covers vibration measurement fundamentals, sensor types, analysis techniques, ISO severity standards, and how to apply these methods effectively in Singapore's industrial environment.
Why Measure Vibration?
All rotating machinery generates vibration as a by-product of operation. A machine in good condition has a low-level, consistent vibration signature; as faults develop (bearing defects, imbalance, misalignment, looseness, cavitation), the signature changes in characteristic ways revealing both the nature and severity of the fault. Trending these changes lets maintenance teams detect developing faults before failure, identify the specific fault type to target corrective action precisely, validate repairs (confirming balancing, alignment or bearing replacement returned vibration to acceptable levels), plan shutdowns based on actual condition rather than arbitrary calendar intervals, and avoid over-maintenance of machines running well. The economic case is well established: the ratio of planned-to-emergency maintenance cost is typically 1:3 to 1:5, so every unplanned breakdown avoided through early detection saves multiple times the programme's cost.
Vibration Measurement Parameters and Sensors
Vibration is characterised by three interrelated parameters: displacement (peak-to-peak movement in mm or µm, relevant for low-frequency vibration below 10 Hz such as shaft orbit measurement), velocity (rate of change of displacement in mm/s RMS, the most widely used parameter for overall severity assessment and ISO 10816 compliance, relatively flat across the 10–1000 Hz range most relevant to structural damage), and acceleration (rate of change of velocity in m/s² RMS or g, sensitive to high-frequency vibration above 1 kHz, preferred for bearing fault and gear mesh analysis). Piezoelectric accelerometers are the most widely used sensor, converting vibration into a charge or IEPE/ICP voltage signal, selected by frequency range (general purpose 0.5 Hz–10 kHz, high-frequency up to 100 kHz for bearing/gear analysis), sensitivity (10–100 mV/g typical), temperature rating (standard -50 to +120°C, with high-temperature versions available), and mounting method (stud mounting most accurate, adhesive, magnetic base for portable surveys, or probe tip). Velocity sensors (seismometers) produce a voltage directly proportional to velocity, the traditional standard now largely replaced by accelerometers with electronic integration, still used for large machines at frequencies below 10 Hz. Eddy current proximity probes measure the gap to a non-ferrous shaft surface continuously and non-contactly, the standard sensor for journal bearing shaft orbit measurement in large turbines, compressors and pumps, where radial position and orbit shape reveal bearing condition, imbalance and misalignment directly.
ISO Standards for Vibration Severity
ISO 10816 (now superseded by ISO 20816, which uses the same approach) provides internationally recognised vibration severity criteria for evaluating machine condition based on vibration velocity measurements on the machine structure (non-rotating parts). The standard categorises machines into groups by power and mounting, and provides four zones:
| Zone | Description | Action |
|---|---|---|
| A | New machine, good condition | No action required |
| B | Acceptable for long-term operation | Continue trending |
| C | Alarm. Investigate cause | Plan corrective maintenance |
| D | Danger. Immediate damage risk | Stop machine; repair urgently |
The specific velocity thresholds for each zone depend on the machine group. For example, for Group 1 (large machines on rigid foundations, >300 kW), Zone C starts at 11.2 mm/s RMS. These thresholds provide an objective basis for maintenance decision-making and are widely referenced in Singapore's industrial maintenance practice.
Vibration Analysis Techniques
The simplest approach, overall (broadband) level, measures the RMS velocity or acceleration and compares it to ISO severity criteria or historical baseline. It detects general deterioration but is insensitive to faults affecting only narrow frequency bands, so an alarm may not trigger until a fault is already advanced. Frequency spectrum analysis (FFT) decomposes the waveform into its constituent frequencies, each characteristic of a specific fault: 1× running speed indicates imbalance; 2× indicates misalignment or looseness; bearing defect frequencies (BPFI, BPFO, BSF, FTF), calculated from bearing geometry and running speed, appear as sidebands when bearing defects are present; gear mesh frequency (number of teeth × running speed) grows in amplitude as teeth wear; and blade pass frequency (number of blades × running speed) is relevant for cavitation and fouling. FFT requires data collectors with spectrum storage; portable analysers from established instrumentation brands allow field collection and software-based analysis, letting Singapore maintenance teams build a diagnostic picture without sending equipment to a specialist facility. Envelope analysis (demodulation) detects the amplitude modulation that bearing defects impose on high-frequency (2–10 kHz) stress waves, giving earlier warning of bearing defects than broadband or FFT analysis alone, and is particularly effective for early-stage rolling element fatigue before it progresses to spalling.
Setting Up a Monitoring Programme and Calibration
An effective vibration monitoring programme requires equipment prioritisation (focusing survey frequency on critical machinery whose failure causes production loss, safety hazard, or environmental consequence), defined measurement points (marked or photographed on each machine, bearing housings at both ends, radial and axial directions, for repeatability between surveys), a baseline established while the machine is newly installed or recently overhauled and in good condition, a survey frequency (monthly for critical machines, quarterly for others, shortened when trending shows acceleration), and dedicated software to store trending data, generate alerts, and produce maintenance reports.
Vibration instruments (accelerometers, vibration meters, data collectors) require periodic calibration against a reference vibration calibrator (shaker) providing a known vibration level at a specified frequency, per ISO/IEC 17511 and ISO/IEC 16063-21. For compliance measurements, such as ISO 20816 acceptance tests on new machine installations or MOM workplace vibration assessments under the WSH Act, calibration by an ISO/IEC 17025 accredited laboratory provides the required traceability. Unitest Instruments' SAC-SINGLAS accredited calibration laboratory (accreditation LA-2023-0845-C) provides calibration for vibration measurement equipment; contact Unitest Instruments at +65 6659 8878 for details. See also our related article on thermal imaging for predictive maintenance, which complements vibration monitoring as part of a comprehensive PdM programme.
Vibration and Singapore's Workplace Safety Regulations
Beyond machinery condition monitoring, vibration measurement has a direct workplace health application. MOM's Workplace Safety and Health Act and the WSH (General Provisions) Regulations address hand-arm vibration (HAV) and whole-body vibration (WBV) exposure to workers. Prolonged occupational exposure can cause Hand-Arm Vibration Syndrome (HAVS), a progressive and irreversible condition affecting fingers, hands, and arms. Employers with workers using vibrating tools (drills, grinders, jackhammers, chainsaws) should conduct vibration risk assessments per ISO 5349 (HAV) or ISO 2631 (WBV) to verify exposure is within safe limits.
