International Journal of Scientific & Engineering Research, Volume 3, Issue 11, November-2012 1

ISSN 2229-5518

Analysis of Railway Wheel to study Thermal and

Structural Behaviour

Pramod Murali Mohan

AbstractWheel-Rail systems are intensively subjected to damage caused by rolling contact and slip behaviour between the wheel and the rail. This is caused because of the brake block pushed against the wheel tread to create friction through which energy is dissipated. Friction also generates heat which results in distortion of the wheel as well as thermal stresses.In order to accurately predict this phenomena an accurate understanding of the thermal and structural loading on the wheel is required. In this paper this type of problem is being discussed. Two applications are presented. In the first the behaviour of wheel due to thermal and structural loading are dealt and in the second the combined loading is being discussed.

Index TermsFailure, Fatigue, Fracture, Heat, Railway Wheel, Structural load, Thermal load, Wear

1 INTRODUCTION

—————————— ——————————
Rail systems exploit friction to transmit power, therefore high cyclic loads, dynamic loads; heat generation and wear are in- evitable. Premature rail replacement, re-profiling of wheels, increased noise, reduced performance and, in the worst case, failure are the consequences of this phenomenon.
The fracture occuring on the railway wheels, are caused by thermal load. Many research institutes such as Indian Institute of Mechanical and Electrical Engineering, Rail Research Insti- tute (ERRI) are carrying out research aiming to define new design of wheel which is as little as possibly sensitive to ther- mal and structural combined loading.
Due to high thermal loads acting on the wheel due to long-
term braking for maintaining constant train speed on lower
side or in some cases the wheels are locked.
This paper is intended towards analysing a CJ36 Griffin
Freight Car wheel subjected to thermal, structural and com-
bined loading.
Analysis is carried out by finite element method (FEM). A
detailed discussion is being carried out on how to analyse
railway wheel. Further, on basis of results obtained, a method
to carry out combined loading and the thermal loads effect on
the wheel is discussed.

2 ASSUMPTIONS AND CALCULATION

2.1 Assumptions

Due to intensive braking, there will be an increase in train traction consumed energy and thermal load. This is done to maintain the train speed constant.

————————————————

Pramod Murali Mohan, Bachelor of Enginering in mechanical engineering in BMSCE, VTU, India. E-mail: pramodmchintu@gmail.com

Let us consider a typical case of a 4 wheel bogie that carries a load of 220KN. This bogie is travelling at 80KM/h is brought to rest through one brake shoe on each wheel in 30s.
Following are the assumptions considered during analysis:
 CJ36 Griffin freight car wheel as shown in Fig.1 has
been used for modeling.[1]
 Heat generated is uniformly distributed around the periphery of the wheel.
 Apart from the thermal load generated due to brak- ing, the wheel is also subjected to a vertical load and horizontal load of 320KN and 160KN.
 The bogie load is equally distributed to the four

wheels of the freight car.

Fig.1. Cross section of CJ36 Griffin railway wheel

2.2 Calculation

Load acting on for wheels = 220KN = 22426.0958 Kg. Load acting on one wheel (m) = 5606.5239 Kg. Velocity of the bogie (v) = 80 KM/h = 22.222 m/s. Time the bogie brought to rest = 30s.
Kinetic energy generated at wheel = 0.5*m*v2 = 1384326.8 J. Power generated = Kinetic energy / time taken = 46144.226W. Heat flux generated at the rim = Power generated / area.
The diameter of the wheel from the hub is 500mm. The width of the rim that is in contact with the rail is approximate- ly 80mm for the selected wheel.
The surface area which is in contact with the rail is: Area = π*diameter of wheel * width = 125714.285 mm2. Heat flux generated = 0.36720411 W/mm2.

IJSER © 2012

http://www.ijser.org

International Journal of Scientific & Engineering Research, Volume 3, Issue 11, November-2012 2

ISSN 2229-5518

3 ANALYSIS

Griffin wheels are heat treated to improve the wear resistance, and imbibe the circumferential residual compressive stresses in the wheel’s upper rim. These are meant to prevent fatigue cracks. This also induces axial tensile stress which could cause rapid fracture added with impact and thermal load.

Fig.3. Thermal load and boundary condition applied on the wheel.

Fig.2. FEA Model of CJ36 Griffin railway wheel

Fig. 2 shows the mesh generation for cross section of the railway wheel. The mesh is generated using Hypermesh. The parameters considered during mesh generation are element size is 2 and type of element is QUAD element. The triangular element is also called as constant strain triangle (CST) element which is commonly used for most of the analysis. This element is not being employed due to stress tensor being constant throughout. In practical situations, the stresses wil not be con- stant in the element, because of which Quad element is used. The element size 2 is optimized to maintain a aspect ratio within 3-4.
The plane 55 element has been chosed which suits axisym- metric with a two dimensional thermal conduction as well as applicable to steady state thermal analysis. The Fig. 3 shows the thermal load and boundary conditions applied on the wheel. The boundary conditions applied are hub of the wheel is considered to be maintained at ambient temperature, the edges of the plate are subjected to convective film coefficient as air is in contact with the outer surface and the rim edges which is in contact with the rail undergoes heat generation due to friction, this is applied as heat flux load to the wheel.

TABLE 1

MATERIAL PROPERTIES


The material selected for the analysis is AAR class U, Asso- ciation of America Railroads AAR M 107/208, EN 1326, BS
5892 part 4. The Table 1, Depicts the material property used for analysis. [1]
A solid wheel is used for analysis which presents the domi-
nate influence of thermal loads. Model is taken as axissym-
metric because the convection and conduction process on
wheel surface, during high wheel revolution is considered as
uniform.
One of the major steps in analysis is determination of ther- mal balance, i.e. wheel areas where thermal energy exchange
happens with the environment.

Fig.4. Structural loads acting between the rail and wheel.


Fig.4 shows the wheel and rail interactions where the hori- zontal and vertical loads are acting. The points in the figure represent the contact points where volume of each other is intersecting. Usually contact point problems are addressed by dynamic analysis to estimate attitude of axle and effect of normal forces on the contact surface. To carry out a detailed dynamic analysis, axle and rail is to be accounted. Here we assume only the horizontal and vertical loads acting on the wheel. [2]
Fig.5 shows the vertical and horizontal load applied on the
wheel under consideration. The plane 182 element is used for
structural analysis since the geometry is axisymmetric and
thickness can be assigned for plane stress problem.

IJSER © 2012

http://www.ijser.org

International Journal of Scientific & Engineering Research, Volume 3, Issue 11, November-2012 3

ISSN 2229-5518

Fig.5. Structural load and boundary condition applied on the wheel.

at the rim portion of the wheel. The material AAR M107 class U is ductile in nature and it is a high carbon steel with a car- bon percentage of 0.6-0.79 percent. Due to this von mises stress is plotted. The maximum stress is observed at the bot- tom of the plate with a stress of 46.34 N/mm2. This is due to the contact load acting on the major portion of the rim. The bottom portion of the plate’s radius is less than upper portion of the plate. This along with the pre-stressed state of the plate has induced the maximum stress in this region.


The wheel hub is constrained so that the model restrains to rigid body motion. The rim which is in contact with rail is sub- jected to vertical and horizontal load of 320KN and 160KN.

4 RESULTS OF COMPUTATION

The steady state heat transfer through a rail road wheel is con- sidered for analysis. The steady model includes heat transfer by conduction through the web of the wheel and convection heat transfer via the plates, i.e. the exposed surfaces of the wheel. The web is treated as an annular disc of uniform thick- ness, connected to the wheel tread and hub by contact con- ductance and transferring heat to the surroundings by convec- tion.

Fig.7. Structural displacement and von mises stress distribution due to structural loads.

Fig.6. Temperature distribution of the wheel.

Fig.6 shows the steady temperature distribution in the uni- form annular disc. Since the rim of the wheel is in contact with the rail, due to friction between rim and rail during brake ap- plied heat generation is more. The maximum temperature at- tained is at the rim with a temperature of 343.44K. Due to con- vection around the plate of the wheel the temperature reduces even though it is conducted towards the hub. If effective cool- ing is not employed and due to excessive braking, residual thermal stress will get induced which will lead to wear and fatigue failure of the wheel. [3,4]
To reduce the complexity of structural analysis, the vertical and horizontal loads are acting on the rim. This results in the simplicity of point loads acting on the surface of the wheel.
Fig.7 shows the structural displacement and von mises stress distribution for the load acting on the wheel. It is ob- served that the maximum deflection of the wheelis 0.2196mm

Fig.8. Combined loading acting on the wheel.

Fig.8 shows the combined loading acting on the wheel. The element plane 182 is retained for combined loading because the temperature distribution over the cross section of the wheel is applied as body loads. In this type of element the volumetric strains at the gauss integration points are replaced by the average volumetric elements to achieve reduced inte- gration method. This helps to prevent volumetric mesh lock-

IJSER © 2012

http://www.ijser.org

International Journal of Scientific & Engineering Research, Volume 3, Issue 11, November-2012 4

ISSN 2229-5518


sidual stresses in order to prevent fracture.
Following process can be carried out to monitor:
 Determination of paint burns and therefore to choose
the paint sensitive to high temperatures.
 Determination of increase in the distance of the inter-
nal side surfaces of the wheel.
Owing to these damages, wheel reprofiling has to be car-
ried out. In order to prevent wheel fractures caused due to
thermal loads best method is to test the residual stress of
wheel rim of different material to determine the allowable level.[5] This along with intensive research by carrying out
vibration analysis, transient analysis along with rail and axle would help to prevent the wheel damage at the initial stage.

Fig.9. Displacement and von mises stress distribution due to combined loading.

Fig.9 shows the displacement and von mises stress distribu- tion due to combined loading. Since temperature is applied as body load, thermal stresses are developed in the wheel. This along with the vertical and horizontal load has led to 1.084mm of deflection of the wheel as compared to 0.2196mm of deflec- tion due to structural loading. If the excessive braking of the wheel is excercised continuously over a period of time, plastic deformation occurs on the surface of the wheel causing ratch- eting across the periphery of contact surface of the wheel.[7,8]
This also leads to pressure driven vertical cracks in the
wheel. As the temperature of rim increases, there will be a
built up of the residual stress. This eventually leads to increase
in deflection of the wheel, causing the wheel damage called
“Wheel flat” formed by a locked wheel sliding on a rail.
The maximum stress attained is 148.982N/mm2 as com- pared to 46.34 N/mm2 during structural loading. Large
amount of stress is induced due to the thermal load. During selection of material for wheel design, the factor of safety for yield stress should be considered greater than 2 to avoid dam- age that may occur due to combined loading.
Due to these damages it causes high impact load resulting in cracking, wear, noise and discomfort.

5 CONCLUSION

The paper is intended to outline a simple first stage analysis of railway wheel. The analysis result depicts the behavior of wheel for varying loading conditions. It is observed that ex- cessive braking of wheel leads to thermal overloading which results in fatigue, crack propagation leading to fracture and wear.
In order to prevent damages, measures are to be taken for consistent wheel monitoring process and examination of re-

REFERENCES

[1] Kexiu Wang , “The Probabilistic Study of Heat Treatment Process for

Railroad Wheels Using ANSYS/PDS”, Griffin Wheel Company.

[2] M. Bozzone, E. Pennestri, P. Salvini , 8th International Conference on Contact Mechanics and Wear of Rail/Wheel Systems (CM2009), Firenze, It- aly, September 15-18, 2009, COMPLIANCE BASED METHOD FOR WHEEL-RAIL CONTACT ANALYSIS.

[3] Dusan Milutinovic-Railway Transport Company Belgrade, Aleksan- dar Radosavljevic-Traffic Institute CIP Belgrade, Vojkan Lucanin- Associate Professor University of Belgrade Faculty of Mechanical Engineering “Temperature and Stress State of the Block-Bracked Solid Wheel in Operation on Yugoslav Railways, FME Transac- tions (2003).

[4] Matthew Rudas, John Baynham, Robert A Adey “Simulation of wheel-rail

Damage” .(Unpublished paper)

[5] Coenraad Esveld, Professor of Railway Engineering and Valery L.

Markine, Assistant Professor of Railway Engineering and Ivan Y. Shev- tsov, Researcher of Railway Engineering, TU Delft, “Shape Optimization of a Railway Wheel Profile”.

[6] Tanel Telliskivi, “Wheel-Rail Interaction Analysis”, PhD dissertation,

Dept. of Machine Design, Royal Institute of Technology, 2003. (Thesis) [7] K. D. Cole, C. M. Tarawneh, A. A. Fuentes,B. M. Wilson,and L. Navarro,

“Thermal models of railroad wheels and bearings”, Published in Interna- tional Journal of Heat and Mass Transfer 53:9-10 (April 2010), pp. 1636 -

1645.

[8] Yongming Liu, Liming Liu, Brant Stratman, Sankaran Mahadevan, “ Multiaxial fatigue reliability analysis of railroad wheels”, Reliabil- ity Engineering and System Safety 93(2008) 456-467, published at science direct.

IJSER © 2012

http://www.ijser.org