Back

A Practical Guide to ISO 20816: Mastering Non-Rotating Vibration Analysis for Structures & Reciprocating Machines

Aug 8, 2025

non-rotating vibration iso
Hero image for A Practical Guide to ISO 20816: Mastering Non-Rotating Vibration Analysis for Structures & Reciprocating Machines

You’ve done everything by the book. The new reciprocating compressor was precision-aligned, the rotating components were dynamically balanced, and the initial vibration readings on the crankshaft are textbook perfect. Yet, weeks later, the discharge piping develops a fatigue crack, the anchor bolts keep working themselves loose, and a nearby control panel is suffering from phantom electronic failures. What’s going on?

The culprit is often the invisible force you weren't looking for: non-rotating vibration. While traditional vibration analysis focuses on the health of shafts, bearings, and rotors, it often misses the bigger picture—the health of the entire machine system, including its casing, foundation, piping, and supporting structures. This structural vibration is the unsung villain of asset failure, causing a cascade of problems that can baffle even seasoned reliability teams.

This is where international standards like ISO 20816 come in. But this isn't just another article defining a standard. This is your definitive, practical implementation guide for 2025. We'll move beyond theory and into a step-by-step workflow designed for reliability engineers, maintenance managers, and facility operators who need to solve complex, real-world problems. We will dissect the standard, provide a complete implementation workflow, explore advanced troubleshooting, and show you how to integrate this critical data into a world-class reliability program.

Why Non-Rotating Vibration is the Unsung Villain of Asset Failure

For decades, the spotlight in vibration analysis has been on rotating components. While this is crucial, it's only half the story. The energy generated by a machine doesn't just stay within the rotating assembly; it's transferred directly into the machine's static components and support system. When this energy is excessive or excites a natural frequency, it creates a state of destructive structural vibration.

Beyond the Shaft: The Ripple Effect of Structural Vibration

Think of a machine's casing and foundation as the speakers for the music being played by the rotating parts. Even a perfectly tuned "song" can sound terrible if the speakers are cracked or rattling. Structural vibration propagates through an entire system with damaging consequences:

  • Foundation Degradation: Cracking, crumbling, and shifting of concrete pads and steel skids, leading to severe misalignment and catastrophic failure.
  • Piping and Weld Failure: The most common and often most dangerous outcome. Seemingly minor vibrations can excite a resonant frequency in a long pipe run, leading to rapid fatigue failure at weld points, elbows, and flanges—a major safety hazard in chemical, oil, and gas facilities.
  • Ancillary Equipment Damage: Gauges, sensors, transmitters, and electrical conduits attached to a vibrating structure will fail prematurely. This leads to unreliable process control and a constant, frustrating cycle of replacing small components.
  • Looseness and Bolting Issues: Chronic structural vibration is a primary cause of loosened hold-down bolts, mounting feet, and structural fasteners, creating a dangerous feedback loop of increasing vibration and further loosening.

The Unique Challenge of Reciprocating Machines

While structural vibration affects all machinery, it is a particularly acute problem for reciprocating machines like large-bore compressors and industrial engines. Unlike the smooth, continuous forces in a centrifugal pump or turbine, a reciprocating machine generates powerful, pulsating forces from two main sources:

  1. Inertial Forces: The constant starting, stopping, and changing of direction of heavy pistons and crossheads creates significant shaking forces, even in a perfectly balanced machine.
  2. Gas Forces: The compression and release of gas create powerful, cyclical pressure pulsations that act on the cylinders, frame, and associated piping.

These forces combine to create a complex vibration signature that is rich in harmonics (multiples of the running speed). This makes them especially effective at finding and exciting natural frequencies in the machine frame, skid, foundation, and connected piping, making non-rotating part vibration analysis an absolute necessity.

Demystifying the Standard: An Engineer's Guide to ISO 20816

To manage non-rotating vibration effectively, we need a standardized framework. That framework is the ISO 20816 series of standards, "Mechanical vibration — Measurement and evaluation of machine vibration."

From 10816 to 20816: What Changed and Why It Matters

Many experienced professionals will be familiar with the predecessor standard, ISO 10816. The transition to ISO 20816 wasn't just a name change; it represented a modernization and clarification of vibration evaluation. Key changes included:

  • Broader Scope: The 20816 series provides more specific guidance for a wider range of machine types in its various parts.
  • Clarified Measurement Conditions: More detailed instructions on operational conditions during measurement to ensure data is comparable and repeatable.
  • Emphasis on Both Absolute and Relative Vibration: While non-rotating part vibration (absolute) is our focus here, the new standard better integrates it with shaft (relative) vibration for a complete picture.

The most important thing to understand is that the core principles of using vibration velocity measured on non-rotating parts to judge machine health remain. If your procedures still reference 10816, it's time to update them to the current ISO 20816 to ensure you're aligned with global best practices.

Core Concepts of ISO 20816-1 (General Guidelines)

Part 1 of the standard lays the foundation for all the other parts. For our purposes, the key takeaways are:

  • Measurement Quantity: The primary metric for evaluating vibration on non-rotating parts is vibration velocity, typically measured in millimeters per second (mm/s) RMS (Root Mean Square). Why velocity? Because for the frequency range typical of most industrial machinery (around 10 to 1000 Hz), velocity gives the most balanced indication of vibration severity. Displacement over-emphasizes low frequencies, while acceleration over-emphasizes high frequencies. Velocity represents the destructive energy most consistently across this range.
  • Measurement Locations: The standard specifies that measurements should be taken on the "non-rotating parts" of a machine, typically at the bearing housings or other structurally significant points. The goal is to capture the dynamic forces being transmitted from the rotor to the stationary structure.
  • Measurement Axes: Measurements should be taken in three orthogonal directions: horizontal, vertical, and axial. Often, horizontal and vertical are the most revealing for structural issues.

Deep Dive into ISO 20816-6: Reciprocating Machines

This is the critical document for anyone managing large reciprocating compressors or engines. ISO 20816-6 provides specific guidance for machines with power ratings above 100 kW.

  • Scope: It applies to reciprocating machines, both separable and integral, operating between 120 and 1800 RPM. This covers the vast majority of large process gas compressors and industrial engines.
  • Specific Measurement Points: It provides more detailed locations than Part 1. For a typical compressor, this includes measurements on the frame near each main bearing, on each cylinder, and potentially on the crosshead guide.
  • Vibration Severity Zones: This is the most practical and widely used aspect of the standard. It divides vibration magnitude into four zones, providing a clear, actionable framework for assessment.
ZoneDescriptionOperational Guidance
Zone ANewly commissioned machinesThe vibration is within the expected range for new or refurbished equipment.
Zone BUnrestricted long-term operationThe machine is considered acceptable for continuous operation without any restrictions.
Zone CUnsatisfactory for long-term operationThe machine is considered unsuitable for continuous long-term operation. Corrective action should be planned and implemented. The machine can typically be operated for a limited time in this condition to avoid unscheduled shutdowns.
Zone DCritical, potential for damageVibration values in this zone are considered critical and likely to cause damage. The machine should be shut down immediately to prevent failure.

The specific velocity values (in mm/s RMS) for these zones depend on the machine type, its power rating, and its foundation type (rigid vs. flexible). It is essential to consult the tables within the ISO 20816-6 standard to determine the correct limits for your specific asset.

The Complete Workflow: Implementing a Non-Rotating Vibration Program

Having a standard is one thing; using it to prevent failures is another. Here is a step-by-step workflow for building a robust non-rotating vibration monitoring program.

Step 1: Asset Selection and Baselining

You can't monitor everything. Start by identifying the most critical assets where structural vibration poses the highest risk.

  • Who to Monitor:
    • All reciprocating compressors and engines >100 kW.
    • Large centrifugal pumps or fans mounted on flexible steel skids.
    • Machines with a history of foundation issues, piping cracks, or "mystery" failures.
    • Equipment with long, complex, or historically problematic piping systems.
  • How to Track: Use a modern CMMS with a comprehensive asset management module. Each critical asset should have a profile that includes its ISO 20816 classification (machine type, foundation type), defined measurement points, and alarm limits.
  • The Golden Rule of Baselining: The most valuable data you will ever collect is the baseline data from a new or newly overhauled machine. This "as-good-as-it-gets" signature is your benchmark for all future analysis. Collect a full set of readings (overall values and spectra) during commissioning and link them to the asset's record in your CMMS.

Step 2: Sensor Selection and Placement Strategy

The quality of your data is dictated by your sensor and how you mount it.

  • Sensor Choice: For most structural monitoring, a general-purpose industrial accelerometer with a sensitivity of 100 mV/g is the industry standard. It offers a wide frequency range that covers the typical vibration signatures of most machinery.
  • Mounting is Everything: How you attach the sensor is critical.
    • Best: Stud mounting. Drilling and tapping a hole provides the most rigid connection and the most accurate data, especially for high-frequency analysis.
    • Good: A powerful, two-pole rare-earth magnet on a clean, flat, unpainted machine surface. This is the standard for portable data collection. Ensure the surface is properly prepared to avoid "rocking" the sensor.
    • Avoid: Hand-held probes. They are not repeatable and should not be used for trendable data collection.
  • Placement Strategy: Follow the guidance in ISO 20816-6, but also use common sense. Place sensors at the primary load-bearing points. For a large compressor, a typical routine would include:
    • Horizontal, Vertical, and Axial readings at each main bearing housing.
    • Vertical and Horizontal readings on the top of each cylinder.
    • Readings on the foundation or skid at both the drive and non-drive ends.
    • Key locations on high-risk suction and discharge piping (e.g., after elbows, near flanges).

Step 3: Data Acquisition and Setting Alarms

Now it's time to collect data and make it work for you.

  • Configuration: Whether using a portable data collector or a continuous online system, configure your measurement points to collect broad-band overall velocity (mm/s RMS) from 10-1000 Hz, as recommended by the standard. Also, collect a high-resolution FFT spectrum (e.g., 3200 lines of resolution) for detailed diagnostics.
  • Setting Alarms: This is where the ISO zones become your action plan.
    • ALERT/ALARM 1: Set this at the boundary between Zone B and Zone C. This is your "investigate" threshold. It should trigger a notification to the reliability team to analyze the data more closely.
    • TRIP/ALARM 2: Set this at the boundary between Zone C and Zone D. This is your "danger" threshold. It should trigger an urgent warning, potentially leading to an immediate shutdown recommendation.
  • Automation is Key: Manually checking thousands of data points is impossible. A modern AI predictive maintenance platform can automate this entire process. Factory AI's Predict platform transforms raw data into decisive action. The process begins by seamlessly ingesting real-time data from your equipment sensors. Our machine learning models then analyze this information to learn the unique operational fingerprint of each asset, establishing a precise baseline for normal behavior.

When a deviation occurs, Predict instantly flags it as an anomaly. But it doesn't stop there. To provide a complete picture, the platform intelligently contextualizes the alert by correlating it with relevant operational data—such as production schedules, work orders, or recent maintenance activity. This transforms a simple anomaly into a rich, actionable insight, allowing your team to bypass tedious data sifting and focus directly on determining the next best action to prevent downtime.

Step 4: Data Analysis and Root Cause Identification

An alarm is just a question. The spectrum holds the answer. Going beyond the single overall velocity value is what separates a basic program from an advanced one.

  • Interpreting Overall Velocity Trends: The trend plot is your first diagnostic tool. Is the vibration increasing slowly over time (indicating wear) or did it jump suddenly (indicating an event like a broken bolt or process upset)?
  • The Power of the FFT Spectrum: The Fast Fourier Transform (FFT) spectrum breaks down the overall vibration into its individual frequency components. This is where true root cause analysis happens.
    • 1x RPM: Vibration at the fundamental running speed often points to imbalance or misalignment.
    • 2x RPM: In reciprocating machines, 2x vibration is very common and often relates to reciprocating unbalance or crosshead-related issues. On any machine, high 2x vibration can also indicate misalignment.
    • Harmonics (3x, 4x, 5x...): A "picket fence" of harmonics is a classic sign of mechanical looseness—either in the machine's components or, more relevant to our topic, in its mounting to the foundation.
    • Non-Synchronous Peaks: A large peak that is not a direct multiple of the running speed often indicates resonance. This is a red flag for a structural issue where a machine's operating speed is exciting a natural frequency of the foundation, skid, or attached piping.

Step 5: Action and Verification

Data is useless without action. Your analysis must lead to a concrete maintenance task.

  • From Data to Work Order: The final step in the analysis should be a clear recommendation. For example: "High 3x, 4x, and 5x harmonics observed on compressor frame horizontal measurement point. Recommend creating a work order to torque-check all foundation anchor bolts and frame-to-skid bolts."
  • Leverage Your CMMS: This recommendation should be seamlessly converted into a task within your work order software. This creates a closed-loop system, ensuring that findings from the reliability team are executed by the maintenance team. The work order should include the vibration data as an attachment for context.
  • Verify the Fix: After the maintenance work is complete, always take another set of vibration readings. This is the only way to confirm that the root cause was correctly identified and the problem was solved. The "after" data should be stored with the work order, providing a complete history of the event.

Advanced Troubleshooting Scenarios & Real-World Examples

Let's apply this workflow to some challenging, real-world scenarios that highlight the power of non-rotating vibration analysis.

Case Study 1: The "Good" Compressor on a "Bad" Foundation

  • Scenario: A large, newly installed process gas compressor passes all mechanical and rotational vibration checks. However, operators report excessive "shaking" of the entire concrete block foundation. The overall frame vibration is in ISO 20816 Zone C.
  • Analysis: The FFT spectrum from an accelerometer on the frame shows a massive peak at exactly 2x the running speed. The maintenance team insists the compressor internals are perfect. The reliability engineer places a second accelerometer on the foundation and performs a phase analysis. The data shows that the compressor frame and the foundation are moving in unison (in-phase), but with very high amplitude. This suggests the entire system is moving together. A "bump test" (impact testing) is performed on the foundation with the machine off. The test reveals a structural natural frequency of the foundation that is almost identical to the 2x running speed of the compressor.
  • Root Cause: A case of structural resonance. The compressor's normal 2x inertial forces were exciting a natural frequency of the improperly designed foundation.
  • Solution: The problem was not with the machine, but its support. The solution involved engineering a fix to stiffen the foundation and shift its natural frequency away from the compressor's forcing frequency. This was achieved by adding gussets and epoxy grouting under the skid. Post-modification testing showed vibration levels dropped well into Zone B.

Case Study 2: Chasing Ghosts in a Piping System

  • Scenario: A multi-stage centrifugal pump system experiences repeated fatigue failures of a small-bore drain line welded to the main discharge pipe, about 30 feet from the pump. Vibration on the pump bearing housings is low and well within Zone B.
  • Analysis: The team focuses on the structure. An analyst takes vibration readings along the length of the discharge pipe. The levels are low near the pump but increase dramatically as they approach the location of the failed drain line, peaking at a point midway between two pipe supports. The FFT spectrum is dominated by a single, sharp peak at the pump's blade-pass frequency.
  • Root Cause: Flow-induced vibration from the pump was exciting a resonant frequency of the long, unsupported section of pipe. The pipe itself was acting like a guitar string, with the point of maximum vibration (the anti-node) occurring right where the drain line was attached, causing rapid fatigue.
  • Solution: The fix had nothing to do with the pump. A simple, properly designed pipe support was installed at the point of maximum vibration. This changed the stiffness and mass of the pipe section, shifting its natural frequency far away from the blade-pass frequency. Subsequent readings showed negligible vibration along the entire pipe run, and the failures ceased. For more information on such complex systems, engineering organizations like ASME provide extensive resources.

Integrating Non-Rotating Vibration into Your Broader Reliability Strategy

A successful program doesn't exist in a silo. It must be woven into the fabric of your entire maintenance and reliability effort.

Beyond ISO: Complementary Standards and Techniques

While ISO 20816 is your foundational document, be aware of others. The International Organization for Standardization offers numerous standards for different machine classes. Additionally, for highly critical or problematic assets, advanced techniques may be required:

  • Operational Deflection Shape (ODS): Uses multiple accelerometers and phase data to create a 3D animation of how a structure is moving during operation. It's an incredibly powerful tool for visualizing resonance, looseness, and structural weakness.
  • Modal Analysis: Similar to ODS, but performed with the machine off using a calibrated impact hammer. It identifies the inherent natural frequencies, damping, and mode shapes of a structure.

The Role of a Modern CMMS in Vibration Management

A centralized hub is essential for managing the vast amount of data generated. A modern CMMS software is the brain of your reliability program.

  • Centralized Data: It stores all vibration data, alarm limits, and analysis reports against the specific asset record.
  • Automated Workflows: It automatically generates work orders from vibration alarms, ensuring no finding gets lost.
  • Complete Asset History: It provides a single source of truth for an asset's entire life, including all condition monitoring readings, maintenance actions, and associated costs. This is invaluable for long-term reliability analysis and identifying "bad actor" assets.

Building a Business Case for a Structural Vibration Program

Investing in equipment and training requires justification. Frame the business case around risk mitigation and cost avoidance.

  • Calculate ROI: Quantify the cost of past failures that could have been prevented. Include the cost of lost production, secondary damage (e.g., a $500 pipe support failure that leads to a $50,000 pump repair), and emergency labor. Compare this to the cost of the monitoring program.
  • Focus on Safety: In many industries, a piping failure is not just a maintenance event; it's a major safety and environmental incident. A robust structural vibration program is a key part of process safety management.
  • Embrace Proactive Methodologies: Explain how this program moves the organization up the maturity scale from reactive to predictive, and ultimately, to a state of prescriptive maintenance. Instead of just predicting a failure, you can prescribe the exact action (e.g., "Stiffen pipe support at location X-Y") needed to prevent it entirely. As noted by experts on platforms like Reliabilityweb, this proactive stance is the hallmark of a world-class operation.

Conclusion: Take Control of Your Structural Health

Non-rotating vibration is not a niche or secondary concern; it is a fundamental aspect of machine reliability that is directly addressed by the ISO 20816 standard. By neglecting the health of your machine's structure, foundation, and piping, you are ignoring a primary cause of downtime, safety incidents, and costly secondary failures.

By adopting the practical, workflow-oriented approach outlined in this guide, you can move beyond simply collecting data and start transforming it into actionable intelligence. This means:

  1. Standardizing your evaluation criteria using the clear zones defined in ISO 20816.
  2. Systematizing your program with a defined workflow from asset selection to corrective action.
  3. Analyzing beyond the overall number to diagnose root causes with spectral and phase data.
  4. Integrating your findings into a modern CMMS to create a closed-loop reliability system.

The tools and knowledge are available. It's time to shift your focus, look beyond the shaft, and take control of the structural integrity of your most critical assets.

Ready to implement a world-class condition monitoring program that sees the whole picture? Discover how Predict AI can automate your data analysis and turn ISO 20816 from a document into a dynamic failure-prevention strategy. Learn more today.

Tim Cheung

Tim Cheung

Tim Cheung is the CTO and Co-Founder of Factory AI, a startup dedicated to helping manufacturers leverage the power of predictive maintenance. With a passion for customer success and a deep understanding of the industrial sector, Tim is focused on delivering transparent and high-integrity solutions that drive real business outcomes. He is a strong advocate for continuous improvement and believes in the power of data-driven decision-making to optimize operations and prevent costly downtime.