Person in safety gear using equipment inside a large industrial tank, illuminated by beams of light from above.

Discover the advantages of the vibration analysis (VA) method, understand the basic principles behind VA, and explore the variety of techniques for applying vibration analysis in nondestructive testing and condition monitoring across industries.

Table of Contents

Advantages and Limitations
Basic Principles
Techniques
Industry Applications

What Is Vibration Analysis and How Is It Used in NDT?

Vibration analysis is an NDT method that monitors and evaluates the vibration signatures produced by rotating and reciprocating machinery to detect developing faults and assess equipment condition. Every machine — every motor, pump, gearbox, fan, and turbine — generates a characteristic pattern of vibration when it runs. When a component begins to wear, crack, loosen, or fall out of balance, that pattern changes in predictable ways.

What makes VA especially valuable is that it works on equipment while it is in operation. Unlike most NDT methods that inspect static components for existing defects, VA detects changes as they develop — often weeks or months before a failure occurs. This allows maintenance teams to schedule repairs during planned downtime rather than responding to unexpected breakdowns.

A skilled VA practitioner uses their knowledge of machinery dynamics, signal processing, and fault signatures to configure the measurement, collect data, and interpret the results to identify what is wrong, where, and how urgently it needs attention.

The primary objectives of VA in industrial applications include:

  • Detecting Developing Faults: VA identifies bearing defects, imbalance, misalignment, looseness, and gear wear before they progress to failure — giving maintenance teams time to act.

  • Establishing Machine Condition: By comparing current vibration data against baseline measurements, VA determines whether a machine is running normally, deteriorating, or approaching a critical threshold.

  • Guiding Maintenance Decisions: VA provides objective data that supports decisions about when to inspect, repair, or replace machinery — reducing both over-maintenance and unexpected failures.

Every rotating component in a machine generates vibration as it moves. Bearings create vibration at frequencies related to their geometry and rotational speed. Gears create vibration at frequencies tied to the number of teeth and shaft speed. Imbalance and misalignment create vibration at the fundamental rotational frequency and its harmonics. When a fault develops — a spall forming on a bearing race, a gear tooth wearing unevenly, a fastener loosening — it adds new vibration energy or shifts existing energy in ways that a trained analyst can recognize.

VA practitioners use accelerometers, velocity transducers, and proximity probes to capture this vibration data. The raw time-domain signal is then processed — typically using Fast Fourier Transform (FFT) analysis — to produce a frequency spectrum that shows how much vibration is occurring at each frequency. This spectrum is the primary diagnostic tool: peaks at specific frequencies point to specific fault types and component locations within the machine.

Because the relationship between rotational speed, machine geometry, and fault frequencies is mathematically defined, VA allows practitioners to identify not just that something is wrong, but what is wrong and which component is the likely source.

Advantages and Limitations of Vibration Analysis in NDT

VA is used across industries that rely on rotating machinery — including power generation, manufacturing, oil and gas, and transportation. Before applying VA to an asset management program, it helps to understand the main advantages and limitations of the method.

Advantages of Vibration Analysis

  • Non-Intrusive: VA is performed on machinery in operation. No shutdown, disassembly, or surface preparation is required.

  • Early Fault Detection: VA can identify developing faults weeks or months before they cause a failure, allowing planned maintenance rather than emergency repairs.

  • Fault-Specific Diagnosis: The frequency content of a vibration signal identifies not just that a fault exists, but what type of fault it is and which component is affected.

  • Continuous Monitoring Capability: Permanently installed sensors can monitor critical machinery around the clock, triggering alerts the moment vibration levels exceed defined thresholds.

  • Cost Effective at Scale: Route-based surveys cover large fleets of machines efficiently, and online systems protect the highest-value assets continuously.

  • Broad Applicability: VA applies to virtually any rotating or reciprocating machinery — motors, pumps, fans, compressors, turbines, gearboxes, and more.

Limitations of Vibration Analysis

  • Rotating Equipment Only: VA is designed for machinery with rotating or reciprocating components. It is not applicable to static structures, welds, castings, or non-moving components.

  • Requires Baseline Data: Meaningful interpretation depends on comparing current readings to known-good baseline measurements. Without baselines, detecting gradual change is more difficult.

  • Skill-Dependent Interpretation: Frequency spectra can be complex, particularly on multi-stage gearboxes or machines with multiple bearings. Accurate fault diagnosis requires training and experience.

  • Signal Masking: In machines with many rotating components, strong vibration from one source can mask weaker signals from a developing fault elsewhere.

  • Speed Sensitivity: Analysis is most reliable when machines run at consistent, rated speed. Variable-speed equipment requires more sophisticated processing to extract meaningful fault information.

  • Not a Direct Flaw Sizing Tool: VA identifies fault types and trends but does not precisely size or locate internal cracks or material loss the way volumetric NDT methods do.

How Vibration Analysis Works: Basic Principles

All rotating machinery produces vibration as a natural consequence of movement. Each rotating shaft, bearing, gear, and impeller contributes its own vibration energy at frequencies directly related to its rotational speed and geometry. In a healthy machine, these vibration levels are low and stable. When a component develops a fault — a bearing beginning to spall, a shaft going out of balance, a coupling becoming misaligned — the vibration pattern changes.

VA captures these changes by measuring vibration with sensors mounted on or near the machine, then analyzing the resulting data. The most important analytical tool is the frequency spectrum — a display of how much vibration energy is present at each frequency. Because every fault type produces vibration at predictable, calculable frequencies, the spectrum allows practitioners to read the condition of a machine the way a fingerprint reveals identity.

VA practitioners use their knowledge of machinery dynamics, bearing and gear geometry, and signal analysis to set measurement points, configure instruments, collect data, and correctly interpret what the frequency content reveals about machine condition.

The Three Properties of a Vibration Signal

Every vibration measurement is described by three properties. Understanding what each reveals is the foundation of vibration analysis.

Amplitude

How much the machine is vibrating — measured as displacement, velocity, or acceleration depending on the frequency range and the fault being sought. Amplitude indicates severity. Rising amplitude on a specific frequency component is one of the clearest signs that a fault is developing and progressing.

Frequency

The rate at which vibration repeats, expressed in cycles per second (Hz) or cycles per minute (CPM). Frequency identifies the source. Each type of fault — imbalance, misalignment, a specific bearing defect, a gear tooth problem — produces vibration at a characteristic frequency that can be calculated from machine geometry and running speed.

Phase

The timing of the vibration peak relative to a fixed reference point on the rotating shaft. Phase helps distinguish between fault types that occur at the same frequency. Imbalance and misalignment both create vibration at running speed, but their phase relationships differ in ways that allow an experienced analyst to tell them apart — and to determine whether correction is needed in one plane or two.

How Frequency Content Reveals Fault Type

One of the most powerful aspects of VA is that different fault types produce vibration at predictable frequencies. Once the running speed and machine geometry are known, the frequencies at which each type of fault would appear can be calculated in advance. Finding a peak at one of those frequencies — especially one that is growing over time — identifies both the fault and the component.

Imbalance produces a dominant peak at 1x running speed (one cycle per revolution). It is one of the most common causes of elevated vibration and is usually correctable by rebalancing the rotating element.

Misalignment typically produces elevated vibration at 1x and 2x running speed, often with a strong axial component. Correcting coupling or shaft alignment eliminates the source.

Bearing defects each produce vibration at specific calculated frequencies — one for the inner race, outer race, rolling element, and cage — derived from bearing geometry and running speed. These frequencies are non-integer multiples of running speed, making them easily distinguishable from imbalance or misalignment in the spectrum.

Gear mesh frequency is the product of the number of teeth on a gear and its rotational speed. A healthy gear produces a clean peak at gear mesh frequency. Defects on individual teeth add sidebands around that peak; worn or damaged gears raise the overall noise floor at those frequencies.

Mechanical looseness generates a spectrum rich in harmonics of running speed — 2x, 3x, 4x, and higher — because the looseness allows the component to move in a non-linear, repeating pattern with each revolution.

Three workers in safety gear operate machinery on a high platform, with a clear blue sky and landscape in the background.

Manual Scanning

  • Handheld MFL scanners are pushed or pulled across a surface by the technician. This is the standard approach for tank floor inspections, where the scanner moves in a systematic pattern to cover the entire surface. Data is recorded continuously and reviewed after the scan. Manual systems are well suited to confined spaces, irregular layouts, and inspections where setup time needs to be minimized.

Automated and Robotic Scanning

  • Motorized scanners move across large surfaces automatically at controlled speed, providing consistent coverage and reliable signal quality. These are commonly used for large tank floors and structural plate inspections.

  • Robotic crawlers extend this capability to vertical surfaces, underwater structures, and other locations where manual scanning would be impractical or unsafe. Automated systems typically integrate signal threshold alerts that notify the operator when a discontinuity exceeds a set level.

In-Line Pipeline Inspection (Pigging)

  • MFL pipeline pigs are inspection tools that travel inside a pipeline, propelled by the flow of product. As the pig moves, its magnets saturate the pipe wall and sensors inspect the full circumference for corrosion, pitting, and mechanical damage.

  • High-resolution pigs use denser sensor arrays and stronger magnets to detect smaller defects and provide more detailed data for fitness-for-service assessments.

How Vibration Analysis Is Conducted in NDT

VA uses different measurement configurations and analysis approaches depending on the machinery being monitored, the criticality of the asset, and how frequently data needs to be collected.

Add Vibration Analysis Certification to Your Qualifications

ASNT certifications enable you to become a qualified Level II or Level III in VA.

What Certification Is Right for Me?

Industry Applications of Vibration Analysis

Vibration analysis is used wherever rotating machinery is critical to operations. Because VA works on equipment in service, it integrates naturally into maintenance programs across sectors where machine reliability directly affects safety, production, or cost.

Energy

In power generation, VA is used to monitor turbines, generators, boiler feed pumps, cooling fans, and balance-of-plant equipment. Wind turbine drivetrains — gearboxes, main bearings, and generators operating in remote, difficult-to-access locations — are prime candidates for continuous online monitoring. VA is also applied to rotating equipment in nuclear, oil, and gas facilities where unplanned downtime has significant safety and financial consequences.

"A composite image showcasing various energy sources: solar panels in the foreground, oil pump jacks in the middle ground, and wind turbines and a power plant in the background. The scene illustrates the diversity of energy production methods at sunset.

Aerospace

In aerospace, VA is used to monitor jet engine health — tracking vibration signatures from compressor and turbine stages to detect blade damage, bearing wear, and rotor imbalance. Ground test stands and maintenance facilities use VA as part of engine acceptance testing. VA also monitors ground support equipment, hydraulic power units, and auxiliary power units.

A technician performing maintenance or inspection work on the landing gear of a large commercial airplane inside an aircraft hangar. The scene is illuminated with a blue tint, highlighting the aircraft's engines and the structural details of the hangar.

Transportation

VA is a standard tool in rail maintenance for monitoring traction motors, wheel bearings, axle gearboxes, and HVAC equipment in rolling stock. In marine applications, VA monitors propulsion machinery, thrusters, and auxiliary equipment aboard ships and offshore platforms. Automotive manufacturers use VA in production testing to verify that assembled drivetrain components meet vibration specifications before vehicles leave the plant.

A modern high-speed train moving swiftly through a train station at sunset. The motion blur effect emphasizes the train's speed, with vibrant colors in the sky and station lights creating a dynamic and futuristic atmosphere.

Manufacturing

Manufacturing relies heavily on VA to protect the rotating equipment that keeps production lines running — motors, pumps, fans, compressors, spindles, and gearboxes. Machine tool manufacturers use VA to verify spindle condition and detect bearing wear that would otherwise degrade part quality. VA programs in process industries reduce unplanned downtime by identifying equipment that needs attention before it fails during production.

A modern manufacturing facility with robotic arms working on an automated assembly line. The scene is well-lit with blue overhead lighting, showcasing advanced machinery and precision engineering in a clean, industrial environment.

Infrastructure

VA protects the rotating equipment that keeps critical infrastructure operating — water and wastewater treatment plant pumps, HVAC systems in large facilities, escalators and elevator machinery, and data center cooling equipment. Municipal utilities apply VA to high-value pumping stations where an unplanned failure would interrupt service to large populations.

A large infrastructure project featuring a highway under construction. Several cranes are positioned along the unfinished sections of the elevated roadway and bridge. The scene is set on a clear, sunny day with blue skies and some scattered clouds.

Example: VA in the Real World

In a large petrochemical plant, dozens of pumps, compressors, and motors run continuously to keep production moving. Many of these machines are critical — if one fails unexpectedly, it can shut down an entire process unit. Traditional maintenance approaches either replace components on a fixed schedule (wasting useful life) or wait for failure (which is costly and sometimes dangerous).

VA changes this by giving maintenance teams advance warning. During a routine route-based survey, a technician measures vibration on a cooling water pump and notices that a frequency matching the pump's outer race bearing defect frequency — calculated from the bearing dimensions and running speed — has risen significantly since the last measurement. The trend over three months shows steady growth.

The bearing is not yet making audible noise, and the machine is still running within temperature limits. But the VA data gives the maintenance team enough confidence and lead time to schedule a bearing replacement during the next planned outage — avoiding an unplanned failure that could damage the shaft, seal, and surrounding equipment.

A worker in protective gear operates a machine inside a dimly lit tank, casting a shadow on the wall near a bright light source.

Deeper Learning About Vibration Analysis

ASNT offers both members and nonmembers learning opportunities and resources for NDT specialists certified in VA.

Webpage

Electromagnetic Testing (ET) Courses

Advance your skills and knowledge with courses and webinars on electromagnetic testing.

Read More
Book

Programmed Instruction Series: Introduction to NDT

Master the fundamentals of all 16 NDT methods with this comprehensive guide for Level I and II candidates. Covers theory, principles, applications, equipment, and interpretation—plus quiz questions with detailed answer explanations.

Members SAVE 25%: $74.25

Non-Members: $99.00

Read More
Book

Nondestructive Testing Handbook, Vol. 5: Electromagnetic Testing (ET), 3rd ed.

Comprehensive electromagnetic testing reference with an emphasis on modern, digital approaches. Covers core theory, instrumentation, and practical applications across major industries, supporting inspectors, engineers, trainers, and others who evaluate or specify ET work.

Members SAVE 25%: $149.25

Non-Members: $199.00

Read More
Chat Window Trigger