ISO 13373-1: Condition monitoring and diagnostics of machines – Vibration condition monitoring – Part 1: General procedures

Summary

ISO 13373-1 establishes a systematic and repeatable procedure for performing vibration measurements and analysis as part of a condition monitoring program. It serves as a foundational “how-to” guide for setting up a monitoring program, detailing everything from selecting measurement points and parameters to data collection and basic analysis. The goal is to ensure that the collected vibration data is consistent, reliable, and suitable for detecting changes in a machine’s condition over time. This standard essentially formalizes the best practices for route-based data collection.

Table of Contents (Conceptual Structure)

The standard provides a step-by-step guide for establishing a robust vibration monitoring routine:

1. 1. Scope and Objectives:

   This foundational chapter explicitly defines the standard’s purpose, which is to establish a generic, systematic, and repeatable set of procedures for the entire process of vibration condition monitoring. The primary objective is to ensure that vibration data is acquired in a consistent and reliable manner, making it suitable for its intended purpose: detecting changes in a machine’s dynamic behavior over time. The standard is designed to be the procedural backbone for setting up a new vibration monitoring program or for auditing an existing one. It emphasizes that by following these procedures, an organization can create a high-quality database of machine vibration history, which is the essential prerequisite for effective fault detection, trend analysis, and diagnostics. It clarifies that this part of the standard covers the general methodology, while subsequent parts (e.g., ISO 13373-2) provide more detailed diagnostic techniques.

2. 2. Measurement and Sensor Selection:

   This chapter delves into the critical decisions that form the foundation of any vibration measurement. It mandates a structured approach to selecting measurement points, emphasizing that they should be as close as possible to the machine’s bearings to accurately capture the forces transmitted from the rotor. It provides detailed guidance on the orientation of measurements (horizontal, vertical, axial) to ensure a complete three-dimensional picture of the machine’s movement. A significant portion of this section is dedicated to sensor selection, explaining the trade-offs between different transducer types. It highlights that the accelerometer is the most common choice due to its wide frequency range and robustness, but also discusses the use of velocity probes and non-contact proximity probes for specific applications. Crucially, it stresses that the quality of data is directly dependent on the sensor’s mounting method, providing a strong recommendation to use permanent stud mounts for the highest-quality, most repeatable data, and referencing the detailed guidelines in ISO 5348.

3. 3. Measurement Parameters:

   This section is arguably the most technical, as it dictates the settings within the data collector that determine the quality and usefulness of the spectral and waveform data. It provides a detailed methodology for selecting these parameters based on the specific machine and the potential faults being monitored. Key parameters covered include:

   • Frequency Range (Fmax): The standard explains how to select the maximum frequency for the measurement. This must be high enough to capture the signatures of interest, such as the high-frequency tones from bearing defects or gear mesh, without being so high that it introduces unnecessary noise.
   • Resolution: This refers to the number of lines in the FFT spectrum. The standard provides guidance on choosing a resolution sufficient to separate closely spaced frequency components, which is critical for identifying sidebands around a gear mesh frequency or distinguishing between closely spaced running speeds in a multi-shaft machine.
   • Averaging: The standard explains the use of signal averaging to improve the signal-to-noise ratio and provide a more stable, repeatable measurement. It describes different types of averaging, such as RMS averaging and peak hold, and when to apply them.
   • Windowing: This explains the necessity of applying a windowing function (like a Hanning window) to the time data before performing the FFT to minimize an error known as spectral leakage.
4. 4. Data Acquisition Procedures:

   This chapter moves from setup to execution, providing a rigorous procedure for the act of data collection itself. The primary focus is on ensuring that every measurement taken is comparable to all past and future measurements. It places a strong emphasis on documenting the machine’s operating conditions at the time of the test, including its rotational speed, load, temperature, and any other relevant process variables. This is critical because a change in these conditions can significantly alter a machine’s vibration signature, and without this context, a change in vibration could be misinterpreted as a developing fault. The standard also provides a checklist for verifying the integrity of the measurement chain before data collection, ensuring that the sensor is properly mounted, the cable is in good condition, and the data collector’s settings are correct.

5. 5. Data Analysis and Evaluation:

   Once high-quality data has been collected, this chapter provides the framework for its interpretation. It formalizes the two-pronged approach to evaluation first introduced in standards like ISO 10816-1. The first method is **absolute limit comparison**, where the measured broadband vibration value is compared against predefined severity charts (e.g., from the ISO 10816 series) to determine if the machine is in a “Good”, “Satisfactory”, or “Unsatisfactory” condition. The second, and more powerful, method is **trend analysis**. This involves plotting measurement values over time to establish a stable baseline and then looking for significant deviations from that baseline. The standard emphasizes that detecting a change is often more important than the absolute value. It provides the methodology for setting data-driven “Alert” and “Trip” alarm levels—for example, setting an Alert if the vibration doubles (a 100% increase) and a Trip if it quintuples (a 400% increase) from its normal baseline, even if the absolute values are still within an acceptable zone.

6. 6. Basic Fault Identification:

   This final chapter serves as an introduction to the diagnostic process. While the primary focus of Part 1 is on data acquisition and detection, this section bridges the gap to diagnostics by explaining the fundamental principle that different mechanical and electrical faults generate unique, recognizable patterns in the vibration data. It introduces the concept of correlating specific frequencies in the FFT spectrum to their physical sources on the machine. For example, it explains that a high peak at exactly one-times the running speed (1X) is typically indicative of unbalance, while a high peak at 2X running speed often points to misalignment. It also describes how high-frequency, non-synchronous peaks can be associated with bearing defects. This chapter provides the foundational knowledge needed for an analyst to begin the process of root cause analysis, which is the subject of more advanced standards in the ISO 13373 series.

Key Concepts

• Consistency and Repeatability: The central theme of the standard. A monitoring program is useless if the data is not collected in a consistent way. ISO 13373-1 provides the rules to achieve this.
• Data Quality: The standard places a strong emphasis on factors that influence data quality, especially transducer mounting and the selection of appropriate measurement settings (e.g., frequency range, resolution).
• Foundation for a Program: This standard is not a diagnostic guide that tells you how to identify specific faults. Instead, it is the essential first step that tells you how to properly *collect the data* that will be used for diagnostics (which is covered in other standards, like ISO 13373-2 and -3).

← Back to Main Index