Addis Ababa Institute of Technology, Addis Ababa University

African Railway Center of Excellence (ARCE),

Addis Ababa, Ethiopia Issue-05, August, 2022

Evaluating the Evolution and Analysis of axle

fatigue and crack propagation on Railway
Bethel Lemma, Eyuel Mulugeta, Kirubel Moges, Naol Kusa, Mr. Awol Mohammedseid
Student, MSc in Railway Engineering (Rolling stock) [1, 2, 3.4], Addis Ababa Institute of Technology
To Teacher [5], African Railway Center of Excellence (ARCE), Addis Ababa Institute of Technology

ABSTRACT for the future use.overall this paper have four parts
which are: introduction ,literature review ,our gap
This paper review different article about railway axle.
and finally summary
In view of how much axle design is important, how it
is in early stage and the method they used to I. INTRODUCTION
overcome the problem which were triggered by
A railway axle is a single rigid beam and is one of the
accidents and a failure in service are going to be point
most design safety critical components of a bogie, in
out. To import the traditional way of inspection
which the design driver is fatigue strength. The axle
method some even experimented on a way of non-
serves to transmit the driving torque to the wheels,
contact data collection is presented. A series of
maintain the position of the wheels relative to each
experimental investigations was performed on full-
other and supports the vehicle body.
scale railway axles to analyze the fatigue crack
growth behavior of EA4T steel under load spectrum
derived from real operating conditions. Method, like
Finite Element Method numerical and experimental
method carried out based on the commercial software
ANSYS, Abaqus and other way are been assess for
different Authors. Over all this paper review
describes failure scenarios due to fatigue crack The axle load of a wheeled vehicle is the total weight
initiation and propagation and way to investigate, bearing on the roadway for all wheels connected to a
experiment and also, detect to have better (safe) axle
given axle. Axle load is an important design inspection technologies create a growing interest
consideration in the engineering of roadways and towards new solutions and presented the non-
railways, as both are designed to tolerate a maximum destructive inspection method of in-service axles
weight-per-axle (axle load); exceeding the maximum based on non-contact data collection. The NDT
rated axle load will cause damage to techniques for rail- way axles require removing
the roadway or rail tracks. wheels and other components from the axle. The
author arrange to uses non-contact ultrasonic
generation and detection allowing non-contact in-
II. LITERATURE REVIEW service inspection of railway axles at trackside station
through using a numerical method based on explicit
This section of the paper reviews the previous
integration, to investigate a new approach that could
studies, which are to going to be asses and find a gap
be developed for a non-contact in- service inspection
and further suggestion is also been given.
of railway axles at trackside by focusing on the axle
Hoddinott,2004[1] describes five axles that have terminal part, the region between the wheel seat and
failed in service and the consequences of the failures, the axle journal, where it is more difficult to detect
and suggests causes of the failures. Which resulted flaws through traditional methods. Abaqus software
from the slow growth of fatigue cracks across the was used to carry out the numeric simulations.
axles. The problem is that little is known about the
Smith, R. and S. Hillmansen ,2004[3]narrate earliest
maximum loads applied to axles in service or the
accidents by discussion two accidents (the Penistone
range of operating environments, including electrical
accident of 1884 and The Versailles accident of 1842
arcing and corrosion effects where the fatigue cracks
), which leads to considerable progress on empirical
are initiated by Electrical arcing, corrosion and stress
under-standing of the fatigue phenomena, largely
corrosion. The five mid-span failures of axles on
through experimentation. Give a mark to reduce
vehicles that failed in service in the United Kingdom,
dynamic loads from unsprung mass, i.e. mass below
of which four axles fractured completely into two
the suspension of a rail vehicle, which make it
pieces. The method used are the ground theory,
desirable to reduce the mass of axles as far as
narrative analysis, which is supported by past
reasonably possible. A potent way of reducing both
occurred as an evidence.As result,the author remind
bending stresses and axle mass is to employ hollow
that there is a guide issued by British company which
axles.
can help to fill the gap by answering some questions
Rudavskyi, D. ,2020 [4]proposed that failure of axle
to improve standard for axle.
wheelset lead to catastrophic nevertheless the
Mineo,C., D. Cerniglia, and A.
Pantano problem is poorly understood and requires significant
,2014[2],stated that train axles are inspected development especially in the theoretical direction
regularly, but the limits associated to the traditional therefore, in order to prevent possible accidents, there
is a need of reliable prediction of the residual service application of the fracture mechanics methodology
life of wheelset axles. To predict this fracture, one for predicting inspection intervals of railway axles
must be able to calculate the period of subcritical under in-service conditions. These include material
growth of the detected fatigue cracks in the axle, characterization, computational modelling of fatigue
depending on their size and location. Thus, the author crack propagation, effects of variable amplitude
used the energy approach of the fracture mechanics loading and validation experiments on component-
to formulate the calculation model of fatigue crack like specimens. Experimental and Numerical methods
propagation at the wheelset axle surface by using used in validation experiments on component-like
numerical method. And proposed research to create a specimens and the application of fracture mechanics
calculation model of surface fatigue crack growth at tools is currently being introduced as an appropriate
the axle of railway wheelset working under method.
operational loads. The author also formulate the
Beretta, Ghidini et al. 2005 [6]Discussed or
model by finding the period of subcritical crack
addressed the scale effects in fatigue limit and in
growth Ncr from the function of fatigue crack growth
crack growth rate for a high strength steel used for
rate V(S) based on the equation of energy balance in
high speed railway axles. Fatigue limit tests on
thermodynamics (the first law of thermodynamics)
micro-notched specimens led to the determination of
which states as “. The change in a system’s internal
fatigue thresholds for small cracks of the examined
energy is equal to the difference between heat added
steel. This allowed us to successfully analyses the
to the system from its surroundings and work done by
‘scale effect’ in fatigue strength of railway axles in
the system on its surroundings.”
terms of threshold stress for propagation of largest
Luke, Varfolomeev et al. 2011[5], presented Basic non-metallic inclusions, whose probability of
fatigue crack growth data both in the range of stable occurrence increases with component size. A series
crack propagation and near the threshold have been of crack propagation tests on small scale specimens
experimentally determined for the heat-treated led to the definition of an EPFM crack propagation
railway axle steels 25CrMo4 (EA4T) and model which has been successfully compared with
34CrNiMo6+QT under constant and variable propagation data on full-scale components. These
amplitude loading at relevant stress ratios results support the application of the crack
(predominantly fully reversed load cycles, (stress propagation model for the determination of axle
intensity factor) R=-1). For the computational inspection intervals. Used Experimental method to
modelling of fatigue crack propagation, a generally test Scale effects in fatigue limit (Propagation
applicable stress intensity factor solution has been thresholds for small cracks, Analysis of full-scale
derived by finite-element analyses. Additionally, the specimen’s fatigue limits) Scale effects in crack
influence of press fitting on the crack propagation growth rate (Crack growth on full-scale axles) and
behavior in a fillet is discussed. results related to the also Ansys software is used to make stress analysis.
Martin Rieger, Christian Moser [7]discussed the Numerical analyses (FEM and DBEM numerical
experiments carried out and discusses the points that methods), Uncertainty on operating conditions by
have been encountered in the course of the project. using: CGRs and crack paths depend on the load
Crack propagation tests on full-scale railway axles spectrum. The author used different Engineering
are time-consuming. Therefore, the output of such software to perform his project like CAD, MATLAP,
tests should be as high and accurate as possible. The FEM, and BDEM.
existing measurement and testing methods for
Pourheidar, Patriarca et al. 2021 [9]investigates the
railway applications were insufficient in terms of
scissure Growth in all-out Railway Axles on bases of
accuracy and reliability. For this reason, a new
Load Spectra. Within the study, a succession of
measuring principle for crack monitoring in real-
experimental investigations was performed on all-out
time, proposed in the referred article, was pursued.
railway axles to investigate the scissure growth
presents the results of these experimental
behavior. it had been supported by a materials
investigations carried out within the framework of the
category of steel named of EA4T steel underneath
research project “Eisenbahnfahrwerke 2 (EBFW 2)”
load spectrum from real in operation conditions. The
– “Probabilistic Fracture Mechanical Concept for the
study was adopted 2 models: (i) the standard Nasgro
Assessment of Railway Wheel-sets”. Used
equation and (ii) the cyclic R-curve construct
Experimental method (two test benches were
enforced. it had been meant to be tested underneath
optimized, and a method for estimating residual
the three-point bending bench with 250 kN load
stresses at the crack tip during testing was applied) A
capability offered at each Politecnico di Milano and
correlation of crack growth and residual stress state
Lucchini Rs R&D laboratories. The materials that
have been taken into account
axles were fabricated from 25CrMo4 (EA4T) steel
in the results, by measuring residual.
and were quenched and tempered with the properties
V. Giannella [8]discussed a stochastic approach to of tensile. discharge machining (EDM) were wont to
the fatigue crack-growth in a railway axle under notice the spherical crack length and Semi-elliptical
variability of different input parameters. The artificial defects with initial depth a = one.5
approach is used to predict the probability millimeter and ratio a/c = zero.8. the size, instead,
distribution of the residual fatigue life of a railway area unit determined by the minimum
axle in presence of defects taking into account detectabledefect size of the used non-destructive
multiple sources of uncertainty. Experimental and methodology at the side of the statistics obtained.
numerical method are used in the investigation. From the result, he determined that, Nasgro equation
Experimentally (Uncertainty in material properties coincide well with the experimental results with a real
by using: NASGRO formulation description and understatement of the residual period but thirty
stochastic approach to NASGRO formulation, second, whereas the normal Nasgro equation ends up
Uncertainty in fracture algorithms by using: in vital overestimation (≈120%) of the residual period
for load spectra getting ready to the in commission chemical compositions in proportion weight have
scenario. additionally taken for the study.

The study recorded the subsequent conclusion. Plan Topaç, Günal et al. 2009 [11]investigated the predictive
of cyclic R-curve through the changed Nasgro fatigue failure of a prototype rear axle housing. he study
equation tips to shaft life estimates that area unit in used was the take a look at of FEA (using finite part
smart agreement with the all-out experimental, the analysis method). The study was primarily on the failure
contribution to mean stresses area unit enclosed prediction on rear shaft. to predict the rationale of the
within the analysis. At the second half, the standard premature failure, the study was used an in depth solid
Nasgro equation indicates to big overestimation of model of the housing was fashioned in CATIA V5R15
propagation period, by an element a pair of.2, within business software package. By victimization this model,
the real service state of affairs for railway axles once finite part model was created. once formation of the
the bulk of the load categories within the load mode, investigators square measure factory-made a shell
by the stamp-welding method from thick sheets of nine.5
spectrum that is correspond to ∆K worth.
metric linear unit from a small alloyed fine grained, hot
Wu, Liu et al. 2018[10]studied Fatigue performance formable, normalized steel of S460N. within the study
and residual lifetime of intercity railway shafts with varied processes square measure applied to the housing
within axle boxes. the study got to address, Associate material through production, like hardening up to 8000C
in Nursing inspiring downside of the fatigue and hot stamping at ca. 750 0C. 5 specimens were
performance and harm endurance rises within the extracted from shaft housing samples and tests (tensile)
case of defects from varied external sources. The are passed out. Mechanical properties square measure
study was on the hands of broken shaft product of the taken in to account within the analysis to work out the
EA4T steel. Calculated results show that the precise properties of the processed material and also the
important safety region has been transferred to the tests were lead at temperature. From the results of the
shaft center in distinction to classical axles, where a study, showed that the regions, wherever fatigue failure
1.0 mm-depth crack because the scrutiny limit has was initiated throughout vertical fatigue tests, square
less chance to propagate. Besides, the wheelset with measure subjected to worry concentration, which might
within shaft boxes shows the smaller stress price. The cause a premature failure before the anticipated five x
analysis from well-known Kitagawa-Takahashi one hundred and five minimum cycles limit and conjointly
diagram indicates a security risk whereas solely the study was found that the development of the fatigue
victimisation nominal stress approach. It compares lifetime of the housing relies on the decrease of the strain
the quality of European and Chinese high-speed concentration.
railway axles that features the low alloy standardized
of EA4T steel or the quenched and tempered
25CrMo4 with medium strength grade. The minor
III. OUR GAP IV. SUMMARY

This paper is significant because it aim to evaluate the The overall review revolve around how to make
overall rticle written about railway axle which gives railway axle more safe by finding a way to with
information about axle from different percpective, stand to its service life which is approxmatly 40
help to compare and analys where we have to work if years. Therefore, as railway engineering student
we want to design an axle and also it invite or point after reading,anaylsising and discssing about
out where we can have further study. different articles we came up with point we
should consider when designing one.
The overview of this paper deal with the crack
propagation of the axle and how series the  Part on axle like midspan, fitten are place
consequence of on service failure is. Many of the where it highly going to have crack first
researchers express the need and necessity of than other place. Therefore we should
examining theoretical , numerical and experimental take this to our count on making or
method to solve the crack propagation analysis of a design axle to make it better.
railway axle. After reviewing there paper we come up
with the gaps as shown below:

● There are paper that only narrate about the

problem but they didn’t back there statement
with experimental or any software analysis.
● By taking different sample we can do many
experiment but that don’t mean that it is going
to be the same in real world so to make there
work more accurate sometime reseachers
● No all but one article which have a picture of
FEM model but didn’t state which software
they used

● Despite the above gap we found overall the

reacher where done in a professional
manner,information backed with evidence.To
be specific one of the outstanding article,
V.Gianella[8], used diferent software,backed
it with numercal anaylsis to the point us as a
group we cann’t find any gaps.
1. Hoddinott, D., Railway axle failure investigations and fatigue crack growth monitoring of
an axle. Proceedings of the Institution of Mechanical Engineers, Part F: Journal of Rail and
Rapid Transit, 2004. 218(4): p. 283-292.
2. Mineo, C., D. Cerniglia, and A. Pantano, Numerical study for a new methodology of flaws
detection in train axles. Ultrasonics, 2014. 54(3): p. 841-849.
3. Smith, R. and S. Hillmansen, A brief historical overview of the fatigue of railway axles.
Proceedings of the Institution of Mechanical Engineers, Part F: Journal of Rail and Rapid
Transit, 2004. 218(4): p. 267-277.
4. Rudavskyi, D., et al., Assessing surface fatigue crack growth in railway wheelset axle.
Archives of Materials Science and Engineering, 2020. 106(2).
5. Luke, M., et al., Fatigue crack growth in railway axles: assessment concept and validation
tests. Engineering Fracture Mechanics, 2011. 78(5): p. 714-730.
6. Beretta, S., A. Ghidini, and F. Lombardo, Fracture mechanics and scale effects in the fatigue
of railway axles. Engineering fracture mechanics, 2005. 72(2): p. 195-208.
7. Rieger, M., et al., Fatigue crack growth in full-scale railway axles–Influence of secondary
stresses and load sequence effects. International Journal of Fatigue, 2020. 132: p. 105360.
8. Giannella, V., Stochastic approach to fatigue crack-growth simulation for a railway axle
under input data variability. International Journal of Fatigue, 2021. 144: p. 106044.
9. Pourheidar, A., et al., Investigation of Fatigue Crack Growth in Full-Scale Railway Axles
Subjected to Service Load Spectra: Experiments and Predictive Models. Metals, 2021. 11(9):
p. 1427.
10. Wu, S., et al., On the fatigue performance and residual life of intercity railway axles with
inside axle boxes. Engineering Fracture Mechanics, 2018. 197: p. 176-191.
11. Topaç, M., H. Günal, and N. Kuralay, Fatigue failure prediction of a rear axle housing
prototype by using finite element analysis. Engineering Failure Analysis, 2009. 16(5): p.
1474-1482.