CORDIS - Résultats de la recherche de l’UE
CORDIS

EURAXLES: Minimizing the risk of fatigue failure of railway axles

Final Report Summary - EURAXLES (EURAXLES: Minimizing the risk of fatigue failure of railway axles)

Executive Summary:
EURAXLES aims to bring the risk of failure of railway axles to such a minimum level that it will no longer be considered as a significant threat to the safe operation of the European interoperable railway system; at the same time, it shall keep the cost of maintenance to a reasonable level and minimise the risk of service disruption.

The global concept approach for axle design, production and maintenance includes:
• A design approach development, including a risk analysis method which could offer a simple design route by combining loads with difference occurrences including loading specificity of vehicles and service conditions together with the axles resistances, including new materials and methods in order to predict the ‘failure probability’.
• New developments will also include:
o improved axle protection against corrosion, including protection of already corroded axles
o improved adhesion of coatings with a study of the roughness influence (adhesion and fatigue behaviour)
o new, innovative coating solutions. The new solutions will also aim to fulfil environmental requirements to avoid or limit VOC emissions.
• New/improved NDT inspection methods will allow the in-service inspection of axles in order to guarantee safe service conditions with a low impact on the vehicle availability.
• A RAMS/LCC analysis of the solutions will be carried out. The railway transportation system requires a risk analysis of the safety components. Activities will improve design validation and inspection technologies of axles which will optimise costs, safety and environmental compliance to be shown with RAMS/LCC analyses.

The EURAXLES global concept will not only guarantee the current level of safety, but improve it in an interoperable network at optimized cost.

Project Context and Objectives:
WP2: New axle fatigue design method
The main objectives of WP2 were:

o To develop a method to characterize the severity of real axle in-service loads, taking into account the variability of the load amplitudes and the variability of axles’ usages.
First of all, a database of on-line axle load measurements was set up, in order to gain a better knowledge of the variability of real in-service loads (Deliverable D2.1): data from the Dynotrain and Hembot projects were made available for the project duration and a test campaign was led by SNCF in order to gather additional data on a passenger coach running in France. From this load data, two simplified representative load spectra were extracted and distributed to all WP2 partners for further analyses.
In parallel, a method for the analysis of the axle fatigue loads was developed. Parametric analyses were performed and they showed, amongst others, the significant sensitivity of the results to the chosen damage law. This damage law was therefore accurately modelled and characterized, thanks to experimental results from WP3 on small specimen under variable amplitude loads. Then, a method to estimate the real distribution of the load severities was proposed by SNCF, based on the definition of elementary situations of life and the generation of virtual uses of axles. Finally, the method was applied to the SNCF passenger coach and the severity of the standardized load (defined in EN13103/13104) was estimated according to the distribution of load severity (deliverable D2.2).

o To introduce the use of the Finite Element Method in the modelling of axles in order to close some open points of the standards EN13103/13104, to give some recommendations for numerical modelling when using the FEM and to develop a commonly accepted numerical validation process of axles.
After defining the detailed geometries of the axles and the benches to be tested and simulated in WP2 and WP3, the numerical models were generated and the analyses were performed, for both trailer and motor axles. Similar results were obtained by all partners and the comparisons with the tests were satisfactory. The two types of bench (Minden and Vitry) also gave similar results. It was also observed that the concentration factors related to transitions defined in the EN13103/13104 are lower than those obtained by FEM calculations. Stresses in grooves derived from FEM were also found different from those obtained following the EN1310X rules. But in the meantime, experimental results from WP3 showed that the fatigue limits were underestimated in the standards, so that the current design process remains safe (as it can also be stated from return of experience from the field). A parametric analysis performed with transitions enabled the team to propose formulas giving the stress concentration factors depending on the axle geometrical parameters. Simulations of whole wheelsets were also performed. Finally, recommendations on how to use FEM in the validation process of axles and how to generate relevant FEM models were given (deliverables D2.3 D2.4).

o To develop a method to estimate the real in-service reliability (probability of failure) of axles, taking account of the variability of the loads and the scatter of the fatigue limits of steel grades A1N and A4T.
First, a semi-probabilistic approach, based on the Eurocode standard, was proposed by Polimi (deliverable D2.5) as an approach to validate an axle design. This method uses a “representative load spectrum” coming from measurements and enables one to propose minimum safety coefficients to be defined in the validation process, depending on the scatter of the material fatigue limits and simple assumptions made on the load uncertainties. A second method, fully-probabilistic, based on the Stress and Strength Interference Analysis, was also proposed by SNCF in order to estimate the probability of failure of an axle, considering the fatigue limit scatter but also the real distribution of the load severity, as defined in D2.2. The SNCF passenger coach trailer axle was taken as an application and its probability of failure was estimated, using the material characteristics identified in WP3 and the load measurements carried out in WP2. The calculated probability of failure is relatively small and its order of magnitude seems rather realistic (deliverable D2.6).

WP3: New testing methods of railway axle fatigue limit assessment
The WP3 scope was to experimentally estimate fatigue limits of axles; such information is a main input in the design. The WP considered axles made in standard material (A1N and A4T) and defined a standard method for testing and analyzing the obtained data from a statistical point of view in order to be able in the future to apply the same method in the characterization of new materials or new surface treatments.
Axle conditions considered for the actual testing were axles in standard surface finishing for which axle body (free surface) was evaluated separately from axles seats where wheel, brake disc or bearing press fits take place; in this later case the coupling of the components can generate, when high bending is applied to the axle, relative micro sliding and derived friction forces that end up in local wear and possible micro cracks of the seat side surface; the phenomenon is known as fretting corrosion and the result is that the fatigue limit is substantially lower than on the body. The severity of this phenomenon depends more on the geometry of the axle (Seat-body diameter ratio) than on the material itself; for this reason, different axle geometries were defined and tested. This part of the testing activity was complemented by task 3.3 dedicated to the theoretical modeling of fretting fatigue providing criteria for evaluating both the possibility for crack initiation and assessment of the crack propagation. The aim was to provide tools for optimizing the design.
An important subject that was treated in WP4 concerns the study of possible solutions that can improve the axles surface protection from corrosion and for this some surface treatment techniques were proposed (shot pinning, increase roughness ecc); as similar solutions can have an effect on the surface fatigue limit; some solutions were verified in WP3 trough simple 1/3 scale axles and then in full scale conditions.
WP3 also tested corroded axles to verify the actual reduction of the fatigue limit.
A common testing procedure was defined in order that similar tests performed in different laboratories could produce comparable results.
A list of small scale and full scale tests and the relative drawings were defined and the wheelset manufacturing partners prepared the materials for the tests. All tests were performed by 7 partners of the Project in their laboratories.

WP4: Tools, technologies and surface protection systems minimizing the negative influence of corrosion
The main issue addressed by research activities in WP4 is that existing painting systems for wheel-set–axles, that meet the environmental requirements, often fail to meet all the requirements defined in the existing standard EN 13261 (Axles Product Requirements, e.g. chapter 3.9). Indeed, for environment protection reasons, legislation is moving from solvent-based paint systems to water-based paint systems.
The fatigue resistance of railway axles depends on the surface conditions. WP4 aims to advance beyond the state-of-the-art by resolving the problems associated with the existing surface coating methods used until now (corrosion, damage) through improved adhesion and new innovative coating and treatment processes, while considering the real service conditions and the environmental requirements.
The main objective of WP4 is to develop practical solutions for axle protection systems to avoid corrosion and damages with respect to the design/ calculation method, the product requirements, the inspection and maintenance requirements and the requirements of environmental legislation.
Given their experience, project partners analysed the current standards and determined their limitations in terms of testing and validation of protective coatings. Particular attention was given to the operating conditions of wheel-sets in service. Where necessary, alternative quality test methods are proposed with regard to operating conditions and reasonable costs.
The expected results of WP4 were new surface preparations and procedures for the design method of wheel-set axles, including new fatigue limits tested in WP3, associated to the new axle surface conditions (roughness or unpainted solutions). WP4 also aimed to investigate innovative and alternative protection systems, eliminating the need for paint application entirely, such as the improvement of hardness and corrosion resistance of wheel-set-axles by innovative treatment/ coating processes.
Regarding the possibility to have a design method without any painting system with defined corrosion level (SNCB is using paint-free axles for freight cars and passenger coaches) unpainted axles from service should be analysed in WP 4 in terms of surface aspect, corrosion, roughness and chemical composition. The objective was also to make fatigue test on about 10 axles from service without painting to determine a new fatigue limit.
The tasks of WP4 finally should include preparing recommendations to improve the existing standard for the product requirements of axles, recommending guidelines and also improved or new quality test methods.

WP5: Non destructive testing (NDT) and verification of the reliability of axles in service
WP5 is focused basically in the study of the NDT methods applied by railway operators and maintenance companies in railway axles, to verify the detectability of these in order to guarantee safe service conditions with a low impact on the vehicle availability.
The main objectives of the Work Package are as follows:
Review of the current practice NDT techniques used in preventive maintenance in railway axles, to highlight the weakness and strengths of each method and to find possible points of improvement and solve potential risks which are not addressed by these methods.
The study of a new inspection method based on a new on-board continuous measurement technique. This technique describes a new methodology of diagnosis and classification of flaws in order to develop a new robust NDT method for axle inspection that could be classified as a condition based maintenance technique. This method is based in a sophisticated signal processing procedure that uses vibration signals obtained during rotation of the axle.
Verification of the influence of surface damage and corrosion in service using standard electrochemical and other NDT techniques. Different NDT techniques are reviewed for monitoring of corrosion and cracking in train axles without the need of disassembly.
The investigation of a novel crack detection method based on the change of the elastoresistive behaviour of an adhesive plug using electrochemical techniques, to detect cracks in railway axles.

WP6: RAMS and LCC taking into account market uptake
In WP2 to WP5, the technical parameters for implementation of the design solutions (coated/non coated surfaces, type of protection, coating thickness, effect of corrosion, etc.) together with adequate inspection technologies to detect defects on axles in real service conditions have been investigated.
Technical aspects of the studied technologies must be complemented by cost analysis in order to assess the translation from research to the application in real operation. Hence, the main objective of WP6 is to review and analyze the market uptake of different solutions developed during the present project.
The overall safety and economic assessment will follow RAMS/LCC methodologies, a recognized method for assisting optimization processes in engineering systems. A common RAMS/LCC tool will be first defined between the participants and afterwards applied for the analysis of the different solutions in order to identify the preferred solutions for future applications and to understand the RAMS and LCC implications for an optimal market uptake. The analysis requires the data collected from the return of experience provided by the partners and collaborators which is used as the reference to evaluate the innovative solutions developed in the project.
The common approach to RAMS/LCC analysis defined in EURAXLES can serve as a starting point for unified safety and cost evaluation of railway systems in general and wheelsets in particular within the European railway industry.

Project Results:
please see attachment
Potential Impact:
A high impact from the different solutions and recommendations from the project is expected in the following technical areas, highlighted at the inception of the project.
• Optimised axle design
The project has proposed a new axle design methodology based on modern analysis techniques and appropriate safety limits derived from calculated and tested loading scenarios of vehicles in service. The new optimised design method will enable the axle to adequately withstand the applied loads leading to reduced axle degradation, thus reducing the risk of axle failure and vehicle accidents.

This specific activity carried out in the project has been presented in the spring 2014 to CEN (European Committee for Standardization). The committee responsible for the standardization of railway rolling stock, in more specifically in the field of Wheels and Wheelsets had decided to officially start working on the publication of a CEN “Technical Report”. This report will use the key results of this activity in order to prepare the grounds for future standardization work.

• New coating solutions
Fatigue cracks causing axle failures are a major cause of accidents, and two recent events (in Italy and Germany) highlight this link. Current organic based coatings are particularly susceptible to degradation from the environment and ballast impact. The project will develop a more robust coating system (through the investigation of various innovative coatings) ensuring axle projection and hence improved fatigue resistance. This will significantly reduce the likelihood of axle failure and axle failure induced accidents, thus reducing the number of fatalities and injuries caused.

• Better detection of defects
Improved detection of defects will be enabled through an improved/innovative NDT methodologies which will permit more accurate measurements of the axle condition and detection of defects. The technology will detect possible defect-induced failures before they occur, thus leading to fewer axle failures and accidents. An improved knowledge of the fatigue properties of used steel and understanding of crack creation and propagation will enable an optimised regime for in service inspection by NDT, MPI, and possibly other new methods. These measures lead directly to a lower probability of crack creation, axle failure and vehicle accidents.


In addition, the common approach to RAMS/LCC analysis defined in EURAXLES can serve as a starting point for unified safety and cost evaluation of railway systems in general and wheelsets, in particular within the European railway industry. Considering the fragmentation in this field in the rail sector, this activity also allowed to advance the RAMS/LCC approach.

The project’s contributions to the standardization process will allow a direct influence of its results onto the different actors of the rail sector, especially at the design level and during the maintenance process. The results will ensure an ever high level of safety of rail transport, and will ensure an improved and cost effective maintenance process, which will have an impact on the competitiveness of the sector as a whole (operators and manufacturers).

The complementarity of the EURAXLES consortium is a strength for the research aspect but will also allow an efficient and effective dissemination of the project results.
They key members of the CEN WG11, responsible for the standardization of railway wheels and wheelsets are also major partners of the project. This will ensure the smooth transition of the project results to standardization, which was one of the main goals of the project.

In addition, UNIFE, as coordinator and dissemination leader, and UIC as dissemination partner, will enable the good communication around the project. The two sector associations are well placed in terms of membership and relations with the key stakeholders to ensure the afterlife of EURAXLES.

EURAXLES was already represented in major events since 2010:
• WCRR 2011, Lille, France and WCRR 2013, Sydney, Australia.
• Transport Research Arena (TRA) 2012 in Athens, Greece and TRA 2014 in Paris, France.
• Dedicated session at the occasion of the 17th International Wheelset Congress (IWC) in Kiev, Ukraine, September 2013.
• The 1st European Forum on Running Gears, Madrid, Spain, in June 2011. The project results were again presented after the official end of the project in June 2014 at the occasion of the 2nd European Forum on Running Gear, Madrid, Spain.
List of Websites:
http://www.EURAXLES.eu