Vibration Analysis in Predictive Maintenance

Most machines vibrate when they are in operation. When you record and study the vibration pattern under specific conditions, you can identify changes in the machine's operational behavior, otherwise known as vibration analysis. Vibration analysis can be used to understand the normal operating condition of an asset and find a deviation from that condition that may signal a degradation in performance or an impending breakdown. This article walks through vibration analysis—how it works, how to do it, and how it can help your maintenance team eliminate downtime.

Key takeaways

• Vibration analysis is a condition-based and predictive maintenance technique that measures vibration to detect developing equipment issues like imbalance, misalignment, and bearing wear before they cause failure.
• A successful vibration analysis program depends on establishing a baseline vibration signature for healthy equipment, then regularly collecting data to spot deviations that indicate a problem.
• Modern analysis uses sensors and software to convert raw vibration data (time waveform) into a frequency spectrum (Fast Fourier Transform), which helps your team diagnose the specific root cause of a fault.
• Integrating vibration data with a computerized maintenance management system allows you to automate work order creation when the system detects anomalies, streamlining the entire process from detection to resolution.

What is vibration analysis?

Vibration analysis (VA) is a way to monitor machine health by measuring equipment vibrations. VA helps pinpoint abnormal vibration patterns that signal developing problems.

Maintenance professionals use VA to identify small failures before they become complete failures or to find signs of potential breakdowns to prevent failures and avoid unplanned downtime.

Traditionally, manufacturers relied on corrective maintenance, where manufacturers waited until after a breakdown to repair the equipment.

Modern manufacturers use vibration analysis as a form of condition-based maintenance (CBM) or predictive maintenance (PdM). Predictive maintenance and other preventive maintenance plans are excellent levers for efficiency and minimize downtime, repair costs, and overtime.

The results speak for themselves—while industry averages show most facilities still experience 5% to 15% unplanned downtime annually, companies using comprehensive condition-based and predictive maintenance strategies achieve significant improvements. Titan America, for example, achieved a 30% reduction in unplanned maintenance through their reliability-centered approach.

Once you've identified a potential issue with vibration analysis, you can create a work order in a computerized maintenance management system (CMMS) like MaintainX. If a part requires replacement or service, you can assign the task to a technician via the platform.

Everything you need to build a CMMS business case that gets a yes

Download Now

Understanding machinery vibrations

All rotating equipment, from production line conveyors to industrial pumps, produce vibrations as a normal part of their operation. These vibrations create a unique signature that acts as a fingerprint for the machine's health.

Vibration analysis is the process of measuring this signature to identify changes that signal developing problems in manufacturing and industrial equipment. This predictive maintenance technique helps maintenance managers detect issues before they cause costly downtime.

Key components of vibration measurement

The two primary components of a vibration analysis are frequency and amplitude.

• Frequency: Measured in Hertz (Hz) or cycles per minute (CPM), frequency indicates how often the vibration occurs. Different mechanical issues generate vibrations at specific, predictable frequencies. For example, a simple imbalance often appears at a frequency equal to the machine's running speed (known as 1X).
• Amplitude: This measures the intensity or severity of the vibration. A higher amplitude typically signals a more serious problem. By tracking amplitude changes over time, you can determine how quickly a fault is progressing.

For example, a maintenance manager monitoring a production line motor running at 1,800 RPM would expect to see normal vibration at 30 Hz (1X frequency). If the analysis reveals elevated readings at 60 Hz (2X frequency), this typically indicates shaft misalignment that requires attention during the next scheduled maintenance window.

By analyzing changes in both frequency and amplitude from an established baseline, your maintenance team can accurately diagnose developing problems. Common issues include bearing defects, gear wear, misalignment, or looseness—all detectable long before they lead to catastrophic failure.

The process of vibration analysis

VA is a four-step process:

• Collect vibration data: Use vibration sensors or a portable device to collect vibration signals and data from equipment.
• Start with a baseline: Collect vibration data from a machine operating in its ideal condition. Record the vibration data to establish a baseline.
• Continually track and store vibration data: Track and store vibration data so you can compare future data sets to historical data sets. You can automate this process by investing in an online monitoring system to collect vibration measurements.
• Analyze vibration data: As you collect data, compare it with the baseline to identify vibration levels, anomalies, and defects. If you find a defect, perform the necessary repairs or maintenance work to prevent machine failure and unplanned downtime.

How to collect data for vibration analysis

Here's what you'll need to get started with vibration analysis.

Mountable or portable equipment

You can install fixed vibration sensors in areas where it's unsafe for technicians to reach regularly.

Maintenance teams most commonly use accelerometer sensors like proximity probes, piezoelectric accelerometers, laser Doppler vibrometers, and microelectromechanical sensors (MEMS).

You'll also find vibration analysts using other types of sensors. However, the right type depends on the application. To choose the appropriate sensor, consider factors like noise, amplitude, sensitivity, mounting options, and sampling vibration frequency, among other indicators.

You also can use portable devices to collect vibration data. Portable devices are especially convenient if you run condition-based maintenance (CBM).

CBM is a form of preventive maintenance but differs from predictive maintenance.

You can use a handheld device alongside a computerized maintenance management system (CMMS) to automatically collect vibration data for equipment.

How vibration analysis software works

Vibration analysis software transforms raw sensor data into actionable maintenance insights. Today's software does the heavy lifting for you:

• Real-time monitoring: Cloud-based systems receive vibration data continuously for continuous vibration monitoring and alert teams when readings exceed preset thresholds
• Data visualization: Software converts complex measurements into easy-to-read charts and graphs for maintenance managers
• Automatic alerts: Systems trigger notifications when they detect potential anomalies, allowing proactive maintenance scheduling
• Historical trending: Platforms store data over time to identify patterns and predict future maintenance needs

Data analysis formats

Vibration analysis tools typically use two primary formats for recording and analyzing vibration data:

Time waveform

Time waveform is the raw data your vibration equipment produces. It compares the vibration's amplitude against time to detect the presence of a defect in equipment, but not the cause. This is where the frequency spectrum becomes helpful.

Frequency spectrum

The Fast Fourier Transform (FFT)—an algorithm that converts time-domain data into frequency analysis—creates a frequency spectrum from the raw time waveform. The frequency spectrum plots the frequency against time, detecting a broader range of potential issues with the equipment.

FFT also helps you get closer to detecting the exact cause of the defect because it provides specific waveforms per equipment component. After collecting the data, the VA tool either detects anomalies or provides the crucial data to reliability professionals for interpretation.

Training and certification

Several institutes offer certification to ISO 18436-2 (condition monitoring and diagnostics of machines using vibration analysis). Some institutions provide courses and training relevant to reliability engineers and vibration analysts.

Here's where to get properly trained:

• The Vibration Institute: The ANSI-recognized Vibration Institute offers a vibration analyst certification with four levels. Each level requires a specific number of training hours.
• Technical Associates of Charlotte: Technical Associates of Charlotte is a world-leading vibration training and certification provider.
• Mobius Institute: Mobius Institute offers vibration analysis certifications and training in other areas like condition monitoring.
• American Society for Nondestructive Testing (ASNT): ASNT is a leading accrediting institute for reliability engineers interested in nondestructive testing. Unlike other institutes on the list, ASNT doesn't focus entirely on vibration analysis but includes it in the coursework.

The final word: Vibration analysis can accelerate predictive maintenance with the right platform

Vibration analysis changes much about how you handle maintenance, but only when you can act on the data. Too many maintenance teams get stuck with systems that can't turn sensor alerts into actual work orders.

MaintainX bridges this gap by connecting vibration monitoring with mobile-first work order management. When your sensors detect an anomaly, your technicians get instant notifications with all the context they need to address the issue before it causes downtime.

Ready to see how predictive maintenance works when everything connects seamlessly? Sign Up for Free and start building a more reliable operation today.