See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/281079832

Stress Distribution Analysis of Rear Axle

Housing by using Finite Elements Analysis

Article · October 2014

CITATION READS

1 1,102

1 author:

Khairul Akmal Shamsuddin

University of Kuala Lumpur
15 PUBLICATIONS 18 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Automotive Ergonomics View project

All content following this page was uploaded by Khairul Akmal Shamsuddin on 07 August 2017.

The user has requested enhancement of the downloaded file.

The International Journal Of Engineering And Science (IJES)
|| Volume || 3 || Issue || 10 || Pages || 53-61 || 2014 ||
ISSN (e): 2319 – 1813 ISSN (p): 2319 – 1805

 Stress Distribution Analysis of Rear Axle Housing by using

Finite Elements Analysis
1
Khairul Akmal Shamsuddin, 2Mohd Syamil Tajuddin,3Megat Mohd Amzari
Megat Mohd Aris,4Mohd Nurhidayat Zahelem
1
Mechanical Section, Universiti Kuala Lumpur (UniKL), Malaysian Spanish Institute (MSI)
2
Mechanical Section, Universiti Kuala Lumpur (UniKL), Malaysian Spanish Institute (MSI)
3
Mechanical Section, Universiti Kuala Lumpur (UniKL), Malaysian Spanish Institute (MSI)
4
Mechanical Section, Universiti Kuala Lumpur (UniKL), Malaysian Spanish Institute (MSI)

----------------------------------------------------------ABSTRACT-----------------------------------------------------------
A premature failure that occurred due to the higher loading capacity of the heavy vehicle is studied. To
determine the reason of the failure, a CAD model of the housing was developed. The mechanical properties of
the housing material were determined from the manufacturer manual. By using the given data, stress
distribution analysis performed by using finite element software and fatigue life are predicted. Design
enhancement solutions were proposed to increase the fatigue life of the housing.

Keywords - Rear axle housing, Stress Distribution, Finite Element Analysis and Fatigue.
---------------------------------------------------------------------------------------------------------------------------------------
Date of Submission: 07 October 2014 Date of Publication: 20 October 2014
---------------------------------------------------------------------------------------------------------------------------------------

I. INTRODUCTION
The automobile was considered merely a mode of transportation. Today, the automobile is
considered more and more as one of life’s necessities. Many countries around the world have immersed interest
and investment in the automobile industry which affects their industries a great deal. Also, they are focusing
on the development of vehicles with improve performance, increased stability, and enhanced driver pleasure
[1]. A premature failure that occurs prior to the expected load cycles during life of rear axle housing as shown
in figure 1 is studied. Due to their higher loading capacity, solid axles are typically used in the heavy
commercial vehicles [2]. During the service life, dynamic forces caused by the road surface roughness produce
dynamic stresses and these force lead to fatigue failure of axle housing, which is the main load carrying part of
the assembly. Therefore it is vital that the axle housing resist against the fatigue failure for a predicted service
life. In automobiles, axle shafts are used to connect wheel and differential at their ends for the purpose of
transmitting power and rotational motion. In operation, axle shafts are generally subjected to torsional stress
and bending stress due to self-weight or weights of components or possible misalignment between journal
bearings. Thus, these rotating components are susceptible to fatigue by the nature of their operation and the
fatigue failures are generally of the torsional, rotating-bending, and reversed (two-way) bending type [3].

Figure 1: Rear axle housing

www.theijes.com The IJES Page 53

 Stress Distribution Analysis of Rear…

II. LITERATURE REVIEW

A. Introduction
Stress distribution analysis of rear axle housing is analysis to determine the stress distribute around
the axle housing after load applied to the housing. In order to determine the location of the stress distribute or
the failure point on the axle housing, the axle housing dimension was measured and convert to CAD software
and this design was imported to FE analysis software to analyze the housing by given loading on it. The
loading point is referring to the real loading condition on the vehicle and other condition of axle housing is
considered to reach this aim.

B. Stress Concentration
In reality, bars often have holes, grooves, notches, keyways, shoulders, threads, or other abrupt
changes in geometry that create a disruption in the otherwise uniform stress pattern. These discontinuities in
geometry cause high stresses in very small regions of the bars, and these high stresses are known as stress
concentration which is shown in figure 2. The discontinuities themselves are known as stress raisers [4].

Figure 2: Stress Concentration Region

An object is strongest when force is evenly distributed over its area, so a reduction in area, e.g. caused
by a crack, results in a localized increase in stress. The crack was occurred due to overload load carried by the
vehicle. Which is the vehicle carry a load more than the axle can support. A material can fail, via a
propagating crack, when a concentrated stress exceeds the material's theoretical cohesive strength. The real
fracture strength of a material is always lower than the theoretical value because most materials contain small
cracks that concentrate stress. Fatigue cracks always start at stress raisers, so removing such defects increases
the fatigue strength. In this test, crack originated at the high stress concentration region.

C. Fatigue Failure
Fatigue is the progressive and localized structural damage that occurs when a material is subjected to
cyclic loading [5]. The nominal maximum stress values are less than the ultimate tensile stress limit, and may
be below the yield stress limit of the material. Fatigue occurs when a material is subjected to repeat loading and
unloading. If the loads are above a certain threshold, microscopic cracks will begin to form at the surface.
Eventually a crack will reach a critical size, and the structure will suddenly fracture [6]. The shape of the
structure will significantly affect the fatigue life; square holes or sharp corners will lead to elevated local
stresses where fatigue cracks can initiate as shown in figure 3. Round holes and smooth transitions or fillets
are therefore important to increase the fatigue strength of the structure.

Figure 3: Crack initiation

www.theijes.com The IJES Page 54

 Stress Distribution Analysis of Rear…

III. METHODOLOGY
A. Finite Element Analysis
i. CAD models
A full scale of rear axle housing model was developed by using Catia software as shown in figure 4
below. There are two main part body was develop which is consist of main part of the housing and dome. The
housing length in X-direction is 1180 mm, in Y-direction is 207.5 mm and in Z-direction is 264.5 mm. The
housing essentially consists of thin walled shells, which is having uniform thickness of 5 mm. Mass body for
the housing is 61.189 kg and it volume is 7.7948 x mm². On the front side, a mounting ring is welded
with cap to increase rigidity of the housing. This cap was replacing drive shaft position where the drive shaft
transmits the rotation to make the rear tires rotate while the vehicle moves. In the other side, a dome is welded
to the rear side for sealing reasons. Element A and B represent for the force acting on the axle. The distance
between element A and B from the mounting ring is the same [7]. Support 1 and 2 stand for the wheel road
contact. The distance between the support housing contact points is equal to the wheel road contact.

Figure 4: Complete CAD model of housing

ii. Finite Element Model

The FE model used in stress and fatigue analyses is shown in figure 5. From the Catia file, the model
was converted to IGES file before run the simulation on FE software. In order to run the simulation, part must
be converted to IGES file. This is the rule while the part was developed from the separate software. To build
the finite element model, housing was meshed a higher order three-dimensional solid element, which has a
quadratic displacement behavior and is well suited to model irregular meshes. The element size is defined by
10 mm completely bonded contact was chosen as the contact condition for all welded surfaces. By using small
size for the element mesh, the result shown have more accurate than using a greater size of element. As a
result for this housing, the total FE model consisted is 53245 nodes which are consist of 30118 element.

Figure 5: Finite Element model of the housing

B. Housing Material
Housing are manufactured by the stamp welding process from 7 mm thick sheets made from a micro
alloyed fine grained, hot formable normalized structural steel S460N. Material number is 1.8901, EN 10025 –
3: 2004 technical delivery conditions for normalized / normalized rolled weld able fine grain structural steels.
Weld able fine grain structural steels are in compliance with the EN 10025-3:2004 standard, which defines
four levels of mechanical properties. Each grade can be obtained with guaranteed impact properties at -20°C
(N grades), or for low-temperature applications at -50°C (NL grades) [8]. Weld able fine grain structural steels
are either normalized or obtained by normalizing co-lamination and thus keep their mechanical properties if
the welded parts are normalized. These steels are used for welded parts that have to withstand high levels of
strain. In order to take effects on the applied process on mechanical properties into FE analysis, each value of

www.theijes.com The IJES Page 55

 Stress Distribution Analysis of Rear…

mechanical properties of this material was defined before simulation. The material was defined as structural
steel with constant variable. The mechanical properties of this material as shown in figure 6.
Material Modulus of Poisson's Yield strength, Ultimate strength,
elasticity, E (GPa) ratio, v Sy (Mpa) Sut (Mpa)

S460N ( 1.8901 ) 208.5 0.3 497.5 643

Figure 6: Mechanical Properties

C. Loading Condition
Load applied to the FE model was chosen according to the loading range used in real vehicle condition.
Tests were performed on a 5 metric tons loading capacity. Housing for a rear axle, which is supported by two
air springs in real life but in FE analysis support was defined as in fixed condition. The load pressed on the
element A and B was divided with same loading condition which is each element was given starting 0 tons up
to 2.5 tons load in 12 seconds as shown in figure 7 below. The load given was used to determine the maximum
loading condition can be supported by the housing before failure. During the loading press, the bending
moment occurs before the housing failure. The maximum loading can be supported by the housing before
failure determined on this analysis is while loading at 4224.755 kg.

Figure 7: Schematic for loading condition

IV. RESULT AND DISCUSSION

A. Equivalent von-Mises Stress
This result based on the equivalent tensile stress known as von Mises theory where this theory state
that the failure at a particular location occurs when the energy of distortion reaches the same energy for failure
in tension. In this case, yielding occurs when the equivalent stress reaches the yield strength of the material in
simple tension. By applying the force data before in 12 seconds, the result of the stress was shown in figure 9
below.

Fig. 8: Graph for Equivalent von-Mises Stress vs Time Fig. 9: Equivalent von-Mises Stress

www.theijes.com The IJES Page 56

 Stress Distribution Analysis of Rear…

From the result, the graph for equivalent von-Mises stress versus time was constructed. Starting from 0
second until 12 second, the stress was increase based on the load applied on it. The stress increase because of
the loading applied is increase. This is the relationship between the load applied and stress occurs. Based on
the graph shown in figure 8, the result exposed that the minimum equivalent von-Mises stress is 0.0036 MPa
when time taken reaches 1 second. The minimum values of this stress means that the load applied still do not
achieve the maximum load applied. When time is increase, the stress also increase directly proportional to the
time until reach the maximum stress before the axle housing failure. After 12 second, the stress reach the
maximum stress where the stress obtained from this analysis is 318.77 MPa. This means after 318.77 MPa the
failure is initiated and these maximum stresses occur on the expected failure region where the failure initiated.

B. Normal Stress

Fig.10: Graph for Normal Stress versus Time Fig 11: Normal Stress

From the result as shown in figure 11 above, the graph for normal stress versus time was constructed. Initial
value from 0 second until 12 second, the stress was raise based on the load acting on it. The stress increase
because of the loading applied is increase. The relationship between the normal stress and time is directly
relative where stress is increasing when time is increase. Based on the graph shown in figure 10, the result
exposed that the minimum normal stress is 0.082 MPa when time initially at 1 second. The minimum values of
this stress means that the load applied still do not achieve the maximum load applied. The stress increase
proportional to the time until reach the maximum stress before the axle housing failure. After 12 second, the
stress reach the maximum stress where the stress obtained from this analysis based on normal stress is 25.613
MPa. This means after 25.613 MPa the failure is initiated and these maximum normal stresses occur on the
expected failure region where the failure initiated.

C. Principal Stresses
The principal stresses are defined as the algebraically maximum and minimum values of the normal
stresses and the planes on which stress act. On this stress component analysis, the stress on rear axle housing
was determined by using finite element analysis. There are two types of result for principal stress which is
divided by maximum and the minimum principal stress. The maximum shear stress or maximum principal
shear stress is equal to one-half the difference between the largest and smallest principal stresses, and acts on
the plane that bisects the angle between the directions of the largest and smallest principal stresses.

Figure 12: Graph for Maximum Principal Stress versus Time Figure 13: Graph for Minimum Principal Stress versus Time

www.theijes.com The IJES Page 57

 Stress Distribution Analysis of Rear…

Figure 14: Maximum Principal Stress Figure 15: Minimum Principal Stress

Based on these two types of principal stress (maximum and minimum) as shown in figure 14 and 15 above, the
value of the principal stress obtained from the analysis result shown that the maximum principal stress has the
higher stress compare to the minimum principal stress. The value obtain fulfill the theory statement which is
the maximum must be higher than the minimum value. These two graph which are shown in figure 12 and 13
were had a similar trend line where the stress is directly proportional to the time. From the analysis, the
maximum principal stress obtained is 298MPa and the minimum principal stress is 42 MPa. The maximum
value for principal stress obtained at time 12 second. The stress occurs at the critical region of the axle housing
where the failure initiated.

D. Reason of the Premature Failure and Maximum Loading Capacity

Premature failure occurs when the axle housing reach the limit of fatigue strength. The axle is no
longer can prevent the failure if the stress acting is higher than the true fatigue strength. From this analysis,
while the force acting on the axle is 5 tons (5000 kg), the axle became failure. In order to predict the maximum
loading capacity, the axle was tested in 12 second where starting from 1 second until 12 second; the value of
load adding is increased by 5000 N. So, at time 12 second the total load acting on axle housing is 50000 N ≈ 5
tons. From the analysis result show that the maximum loading capacity is 4224.755 kg ≈ 42000 N where
for equivalent von-Mises stress is 318.77 MPa.

E. S-N Diagram and Fatigue Life Prediction

By using Finite element software the S-N curve for axle housing material was developed directly after
the mechanical properties for the housing was defined. The cycles developed for this diagram is up to
cycles. According to the theoretical, when the alternating stress is decrease, the fatigue life for the material is
increase. As a result, the curve as shown in figure 16 below fulfills the requirement for the structural steel
curve behaviour. Fatigue life from this experiment is 1 x cycles. The true fatigue strength from the
calculation is 361.0290 MPa. At this point the obtained minimum value of n from this analysis is 1.45. The
minimum value of n from calculated is 1.56 where the maximum stress concentration was observed. This mean
at critical region, fatigue crack can initiate before 1 x cycles as observed in this analysis.

Figure 16: Graph S-N curve for S460N material Figure 17: Factor of Safety

www.theijes.com The IJES Page 58

 Stress Distribution Analysis of Rear…

i. Fatigue Strength Calculation

Since the rear axle housing is actually loaded with dynamic forces during the service, fatigue analysis
was also performed. An estimation of the stress life endurance limit is given as:

= 0.504 . (1)
Where,

= Endurance limit

= Ultimate strength

for steels with ultimate strengths of less than 1400 MPa [12]. This represents the fatigue strength at
cycles and more. For the fatigue life prediction of the part in the range of – cycles, the S–N curve of the
housing material was predictable by means of a practical method given by FE software. stands for the stress
life endurance limit of ideal laboratory samples. To predict the true fatigue strength for a mechanical
component has to be multiplied by several modifying factors which represent various design,
manufacturing, and environmental influences on the fatigue strength. is given as:

= (2)

Where,
= surface factor

= size factor

= load factor

= temperature factor

= fatigue strength modifying factor

where a surface factor which depends on surface finish is given as:

(3)

Where,
= factor

= exponent

Since the roughness of the shell surfaces is similar to hot rolled sheet after hot stamping, recommended values
are a = 57.7 and b = - 0.718. The calculated value of = 0.5557 for = 643 MPa. In addition, shot peening,
a well-known process to introduce favorable residual stresses in the material surface of a component is also
applied on the housing surfaces after hot stamping to increase the fatigue life of the part. Hence is used in
the fatigue analysis as 0.959. For non-round sections, size factor can be assumed as 0.75 for the values of
the depth of the cross-section h, which is greater than 50 mm. Load factor is given as 1 for bending and
temperature factor is 1 for the range of the ambient temperature of T = 0– 250 °C. By means of static FE
analysis, it is observed that there are stress concentrated regions on critical area and arm transition areas.
Therefore, in addition to the modifying factors mentioned, a fatigue strength modifying factor must be taken
into account by means of the static stress concentration factor that is related to fatigue stress concentration
factorc . Hence is calculated as:

(4)
For safety reasons, can be assumed as to be equal to . Because of the dimensions and shape complexity of
the housing cannot be derived from data in the standard literature [2]. On the other hand is defined as:
= (5)

www.theijes.com The IJES Page 59

 Stress Distribution Analysis of Rear…

where is the peak stress and at the root of the notch and the nominal stress which would be
present if a stress concentration did not occur was used as the value obtained from static FE analyses as
= 318.77 MPa. To calculate , the rear axle was assumed as a simple beam which has a uniform
box profile cross-section X–X of critical region along the longitudinal axis Y and subjected to pure bending .
was computed by means of the model given as:
= (6)

Z= (7)

where M is bending moment and Z is the section modulus of the critical cross-section. M was obtained as
5724.6 kN.mm. Section modulus Z was calculated as 39543.96 mm³. Hence was computed as 144.765
MPa. ≈ is found as 2.202 and = 0.454. The S–N curve plotted regarding the modifying factors was
defined in the FE software user interface. Stress-life approach was used to determine the fatigue life of the
housing material. All fatigue analyses were performed according to infinite life criteria (N = cycles). Von
Mises stresses obtained from finite element analyses are utilized in fatigue life calculations. Since the loading
has a sinusoidal fluctuating characteristic (mean stress, > 0), modified Good- man approach given as:

(8)

was used. Here n stands for factor of safety (as shown in figure 17). Stress amplitude is given as:

(9)
and the mean stres can be expressed as:

(10)

Here, both corresponding to maximum 381.13 MPa and and matching a minimum 0.0035924 MPa
of vertical load were obtained via FE analysis. Distribution of n at the lower shell can be seen in factor of safety
distribution region. In the light of the fatigue analysis results, it was estimated that crack initiation can occur at
the region E1 and E2 of the outer shell surface. The value of n is obtained.

ii. Factor of Safety

Factor of safety is used to provide a design margin over the theoretical design capacity to allow for
uncertainty in the design process. Factor of safety to be used in this design is essentially a compromise the
benefit of increased safety and reliability. From the analysis, the value of safety factor of the axle is 1.56. By
using manual calculation, the value of safety factor is 1.45. So, the error between both result is 7% still can be
accepted because less than 10% error.

F. Rear Axle Model Solution

By increasing the thickness sheet metal from 5 mm to 7 mm at critical region as shown in figure 18
below, the rigidity and safety of the axle housing can be improved. From the theoretical, when the area is
increase, the value of stress is decrease. When the stress is decrease, the axle can prevent from failure. So, in
order to reduce the value of stress acting on the axle, the new design with more thickness than before can be
used to solve the axle problem and the axle can be used longer than before.

Fig. 18: Geometry of rear axle housing

www.theijes.com The IJES Page 60

 Stress Distribution Analysis of Rear…

V. CONCLUSION
Stress distribution analysis of rear axle housing was investigated by using finite element analysis. The
model of the axle was developed by using CAD software. The dimension of the axle housing follow the real
dimension based on the data collected. In order to run the simulation in the FE software, the model must be
meshed. By using small values of the element size, the result given will more accurate. All the parameter and
boundary condition for the rear axle was defined in the FE analysis before run the simulation. In the analysis,
in which the stress are distribute, stress concentration from the load given to the axle housing make the axle
housing failure. From the analysis, the result for stress distribution for equivalent von-Mises stress, normal
stress, maximum principal stress, and minimum principle stress was analyzed and determined.The reason
failure occurs is because the axle housing no longer can prevent the load given onto it. From the several load
given, the maximum load for the housing can stand was determined by using FE analysis. The result show that
the maximum load can be carried by this rear axle housing is 4224.755 kg ≈ 42000 N.

In this FE analysis, the fatigue life cycle for the axle material also was computed and the result shown
that the axle material can stand up to 1x cycles before the crack initiated on the critical region. The factor
of safety for the axle housing was determined from the analysis and manual calculation. The values from the
analysis is n = 1.56 and from the calculation is 1.45 can be accepted because only has 7% of the error. In order
to increase the fatigue life of the housing, new design of rear axle model was developed with increasing the
thickness of sheet metal to 7 mm used to make the housing can stand longer and increase the rigidity of the
axle housing. But other consideration such as material changing and redesigning also can be used to improve
the problem
REFERENCES
[1] Woongsang Jeong, Jinhee Jang, Changsoo Han. Modeling & Dynamic Analysis for Four Wheel Steering Vehicle.
[2] M.M. Topac, H. Gunal, N.S. Kuralay (2008). Fatigue failure prediction of a rear axle housing prototype by using
finiteElement analysis.
[3] Osman Asi (2006). Fatigue failure of a rear axle shaft of an automobile.
[4] Carol Francois (2011). What is an axle, http://www.wisegeek.com/what-is-an-axle.htm, 21 March 2011.
[5] James M. Gere (2002). Mechanics of Materials. 5th SI Edition. Nelson Thornes Ltd, pp 139.
[6] Richard G. Budynas & J. Keith Nisbett (2008). Fatigue Failure Resulting from Variable Loading, 8th Edition In SI
Units, pp 257-319.
[7] Higgins, Raymond Aurelius (2001). The Properties of Engineering Material, 2 nd Edition. Raymond A. Higgins, pp.
384-407.
[8] Yung-Li Lee, Jwo Pan, Richard Hathaway, Mark Barkey (2005). Fatigue Testing and Analysis. Elsevier Butterworth-
Heinemann, pp 107-176.
[9] Yasushi Ito, Alan M. Shih and Bharat K. Soni. Reliable Isotropic Tetrahedral Mesh Generation Based on an
Advancing Front Method.
[10] ANSYS Theory References. ANSYS Release 14.0 ANSYS, Inc; 2011.
[11] Shigley JE, Mischke C (1989). Mechanical Engineering Design. 8th Edition. New York, McGraw-Hill, pp 286.
[12] Li-Ping Lei, Jeong Kim, Beom-Soo Kang (2000). Analysis and design of hydroforming process for automobile rear
axle housing by FEM.
[13] Hong SU, Ph.D. Automotive CAE Durability Analysis Using Random Vibration Approach
[14] M. Davis, Y.S. Mohammed, A.A. Elmustafa, RF. Martin, C. Ritinski (2010). Designing for Static and Dynamic
Loading of a Gear Reducer Housing with FEA.
[15] E. Makevet, I. Roman (2001). Failure analysis of a final drive transmission in off-road vehicles.
[16] K.J. Anusavice and B. Hojjatie (1987). Stress Distribution in Metal-Ceramic Crowns with a Facial Porcelain Margin.
[17] Chris Dolling (2006). Material Selection
[18] F.P. Brennan and W.D. Dover (1995). Stress Intensity Factors for Threaded Connections.
[19] Pilkey WD, Pilkey DF. Peterson’s (2008). Stress Concentration Factor. 3rd Edition. John Wiley & Sons, pp 159.
[20] Collins, J.A. (Jack A.) (1993). Failure of Materials in Mechanical Design, pp 34-37.
[21] Walter D. Pilkey (2005). Formula for Stress, Strain and Structural Matrices. 2nd Edition. John Wiley & Sons, Inc, pp
137-344.
[22] Schijve J (2001). Fatigue of structures and materials. Dordrecht, Netherlands: Kluwer Academic Publishers, pp. 68-
72.

www.theijes.com The IJES Page 61

View publication stats