Why EN 15610 Changes European Rail Noise Control
Discover EN 15610: the essential standard for measuring rail and wheel roughness. It’s key to controlling rolling noise, ensuring quieter, more sustainable railway operations.

EN 15610: A Technical Guide to Rail and Wheel Roughness Measurement for Noise Control
EN 15610, titled “Railway applications – Acoustics – Rail and wheel roughness measurement related to noise generation,” is a fundamental European standard that specifies the method for measuring the surface roughness of railway rails and wheels. Its primary purpose is to provide a standardized, reproducible procedure for quantifying the roughness that directly contributes to the generation of rolling noise at the wheel-rail interface.
In the context of railway acoustics, rolling noise is the dominant noise source for conventional trains at speeds typically between 40 km/h and 300 km/h. This noise is generated by the vibration of the wheel and rail, which are excited by the surface roughness of both components. By standardizing the measurement of this roughness, EN 15610 provides a critical tool for noise assessment, mitigation, and the validation of track and vehicle maintenance procedures.
Core Principles and Technical Objectives
The standard is built on the principle that small-scale imperfections on the running surfaces of wheels and rails act as a continuous excitation source. As the wheel rolls over the rail, these imperfections force high-frequency vibrations in the wheel and rail structures, which then radiate sound into the environment. The characteristics of this noise (level and frequency content) are directly linked to the amplitude and wavelength of the roughness.
Key Objectives of the Standard
- Standardization: To establish a uniform methodology for data acquisition, processing, and reporting, ensuring that roughness measurements from different organizations, equipment, and locations are comparable.
- Relevance to Acoustics: To focus specifically on the roughness wavelengths that are most influential in generating audible noise. This typically covers a wavelength range from approximately 3 mm to 1 m.
- Data for Modelling: To provide reliable input data for acoustic prediction models (e.g., TWINS – Track-Wheel Interaction Noise Software), which calculate rolling noise levels based on track and vehicle parameters, including roughness.
- Performance Assessment: To enable the objective evaluation of track maintenance techniques like rail grinding and milling, or wheel maintenance like lathing, by quantifying their effect on acoustic roughness.
Measurement Methodology and Data Processing
EN 15610 outlines a detailed technical procedure, from the physical measurement on the track or wheel to the final data presentation. The process is designed to convert a raw surface profile into a meaningful acoustic roughness spectrum.
1. Data Acquisition
The measurement of the surface profile is performed using high-resolution profilometers. These devices can be contact-based (using a stylus) or non-contact (typically using laser or other optical sensors). The measurement is taken longitudinally along the centre of the running band of the rail or wheel. Key parameters defined by the standard include:
- Measurement Length: A sufficient length must be measured to capture the relevant low-frequency (long wavelength) components accurately.
- Sampling Interval: The distance between consecutive data points must be small enough to resolve the shortest wavelengths of interest (e.g., a sampling interval of 0.5 mm or less).
- Surface Condition: The surface must be clean and dry to avoid measurement errors.
2. Data Processing and Analysis
The raw profile data (a series of height values versus distance) is not directly used. It must be processed to isolate the roughness and present it in a spectral format. The typical steps are:
- Detrending: The long-wavelength shape of the rail or wheel (the nominal profile) is removed from the data to isolate the small-scale roughness. This is often done using a high-pass filter.
- Windowing: A mathematical window function (e.g., a Hanning window) is applied to segments of the data to minimize spectral leakage during the subsequent Fourier analysis.
- Fourier Transform: A Fast Fourier Transform (FFT) is applied to the windowed data segments. This converts the signal from the spatial domain (roughness amplitude vs. distance in meters) to the frequency domain (roughness amplitude vs. spatial frequency in cycles/meter).
- Spectral Averaging: The spectra from multiple segments are averaged to produce a more stable and representative roughness spectrum for the entire measurement length.
- Presentation: The final result is presented as a one-third octave band spectrum. The roughness level is expressed in decibels (dB) relative to a reference amplitude of 1 micrometer (1 µm). This logarithmic format is convenient as it aligns with how human hearing and acoustic levels are perceived.
Practical Applications in the Railway Sector
The data obtained according to EN 15610 is invaluable for a wide range of applications within the railway industry aimed at noise management and operational excellence:
- Noise Acceptance Testing: Used to verify if new or renewed railway lines meet specified acoustic performance targets.
- Track Maintenance Planning: Monitoring the evolution of rail roughness helps infrastructure managers decide when and where to perform rail grinding to control noise emissions.
- Validation of Maintenance: Measuring roughness before and after rail grinding or wheel lathing provides a quantitative measure of the effectiveness of the procedure for acoustic purposes.
- Vehicle Homologation: Wheel roughness data can be part of the acoustic certification of new rolling stock.
- Research and Development: Provides a basis for studying the degradation of wheel and rail surfaces and developing new, quieter wheel and track designs.
Comparison: Acoustic Roughness vs. General Surface Finish
It is important to distinguish the acoustic roughness measured by EN 15610 from general engineering metrics of surface finish (like Ra or Rz). The following table highlights the key differences:
| Feature | Acoustic Roughness (per EN 15610) | General Surface Finish (e.g., Ra) |
|---|---|---|
| Primary Goal | To quantify surface imperfections that generate rolling noise. | To quantify the overall texture of a surface for mechanical purposes (e.g., friction, wear, lubrication). |
| Key Parameter | Roughness level spectrum in one-third octave bands (dB re 1 µm). | A single value representing the average deviation from the mean line (e.g., Ra, Rq, Rz). |
| Domain of Analysis | Wavelength / Spatial Frequency Domain. | Amplitude / Spatial Domain. |
| Relevant Wavelength Range | Typically 3 mm to 1 m. | Typically much shorter, in the micrometer to millimeter range. |
| Primary Application | Railway acoustics, noise prediction, and noise mitigation assessment. | Manufacturing quality control, tribology, and material science. |
Conclusion: A Cornerstone of Railway Noise Management
EN 15610 is more than just a measurement procedure; it is a foundational standard that enables the effective management of railway rolling noise. By providing a common “language” for describing the acoustic state of the wheel-rail interface, it allows infrastructure managers, vehicle operators, regulators, and researchers to communicate clearly and work towards a common goal: a quieter and more sustainable railway network. Its rigorous and scientifically-grounded methodology ensures that noise control strategies are based on accurate, reliable, and comparable data.





