International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056

Volume: 03 Issue: 08 | Aug-2016 www.irjet.net p-ISSN: 2395-0072

Finite Element Analysis and Optimization of Weld Distortion in

Automobile Chassis
Akshay Nighot1, Prof. Anurag A. Nema 2, Prof. Anantharma A 3
1M.E.
Student Dhole Patil College of Engineering, Pune, Maharashtra
2AssistantProfessor Dhole Patil College of Engineering, Pune, Maharashtra
3Assistant Professor Dhole Patil College of Engineering, Pune, Maharashtra

---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Chassis is the critical component, the welding process. The welding simulations simulate an
'backbone' of any automobile and it is on this that various actual welding process based on science and physics and
components come and assembly. To facilitate this assembly the tests can be performed inside computer without
the chassis has several bracket welded to it. During welding, wastage of resources and hazardous environment impact.
the process is not properly controlled, and then it may result A hybrid approach involving both FE modeling and
in excessive weld distortion or residual stress, both of which experimental work has proven very beneficial.
harm the long term life of the chassis. Hence it is imperative
that weld distortion be minimized. In this case the chassis 2. LITERATURE REVIEW
assembly is to be analyzed for the existing weld sequence,
and check deficiencies in the process. Then based on results Chang Doo Jang , Yong Tae Kim , Young Chun Jo , Hyun Su
an improvised sequence is to be recommended. The Ryu [1] Welding deformation reduces the dimensional
objectives of the project are, to create analysis of standard accuracy of ship hull blocks and decreases productivity
operating procedure in ANSYS and to study the weld due to the correction work. Prediction and minimizing of
distortion in current weld structure and then optimize the welding distortion at the design stage will lead to higher
welding along weld sequence, weld duration and clamp quality as well as higher productivity. Dragi Stamenković
location respectively. (2009)[2] This work presents the FE model for numerical
simulation of welding residual stresses in high strength
Key Words: backbone, welds distortion, weld carbon steel butt welds. The finite element method is an
sequence, weld duration, and clamp location. efficient technique in analysing residual stresses in
welding processes. Andrea Capriccioli (2009)[3] In order
to reduce the computer time a simplified procedure has
1. INTRODUCTION been developed in the case of TIG welding mechanical
analysis: this procedure lets to run the whole welding
mechanical simulation with a considerable saving in
Welding is widely used in automotive industries to
computer time. Nuraini, A.A. (2013)[4] The results shown
assemble various products. The chassis is considered to be
that the optimum welding parameter able to lower the
one of the significant structures of an automobile. It is the
temperature and residual stress of the weld center area
frame which holds both the car body and the power train.
compare to non-optimum parameters around 8 percent
When structures are manufactured by welding, a non-
for temperature and 1.5 percent for residual stress. E.
uniform temperature distribution is produced. This
Armentani (2007)[5] They studied as transient
distribution initially causes a rapid thermal expansion
temperature distribution in a welded joint is highly
followed by a thermal contraction in the weld and
affected by thermal conductivity. Large tensile
surrounding areas, thus generating inhomogeneous plastic
longitudinal residual stresses are present near the welding
deformation and residual stresses in the weldment when
bead. S. R. Patil, C. A. Waghmare [6] presents the influence
it is cooled. Thus, it is important to evaluate the
of welding parameters like welding current, welding
deformation due to welding. FEA is most dominant
voltage, welding speed on ultimate tensile strength (UTS)
method for calculation and simulation of computer made
of AISI 1030 mild steel material during welding. From this
models. The name FEA comes from the way a complicated
study, it is observed that welding current and welding
model is divided into a model that is built by small
speed are major parameters which influence on the tensile
elements. The traditional manufacturing processes were
strength of welded joint. Chattopadhyay, G. Glinka, M. El-
mainly established from trial-and-error experiments
Zein, J. Qian, R Forams [7] Fatigue analyses of weldments
approach. Such trial-and-error procedure approach
require detailed knowledge of the stress fields in critical
requires tremendous material, energy, labors, as well as
regions. C.M.Cheng, C.P.Chouy, I.K.Lee and H.Y.Lin [8] In
procedure significant waste, fumes and emissions. The
this study they conducted single V-groove butt welding
appropriate control techniques are mandatory with
(GTAW) on three types of heat treatable aluminium alloys
reliability and cost effectiveness for the application of

© 2016, IRJET ISO 9001:2008 Certified Journal Page 656

 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 03 Issue: 08 | Aug-2016 www.irjet.net p-ISSN: 2395-0072

2024-T351, 6061-T6 and 7075-T6 and compared the 4. METHODOLOGY

angular distortion levels of the three aluminium alloys at The proposed work involves the following steps:
different Vee preparation angles with or without restraint. 1. Perform cleaning operation of the model which is
Y. P. YANG, R. DULL, H. CASTNER, T. D. HUANG, AND D. helpful in development of proper meshing. And develop
FANGUY [9] proper method for meshing.
2. Perform preliminary analysis with half model of chassis
3. PROBLEM DEFINITION with reduced number of nodes and hence the solution
time, to check out the result pattern and to develop
Chassis is the critical component, the 'backbone' of any standard operating procedure in workbench.
automobile and it is on this that various components come 3. Considering full model of chassis perform Transient
and assembly. To facilitate this assembly the chassis has Thermal analysis with given input condition.
several bracket welded to it. During welding, the process is 4. Perform Transient Structural Analysis with result
not properly controlled, and then it may result in excessive obtained in thermal analysis as an input named imported
weld distortion or residual stress, both of which harm the body temperature along with the fixed supports to
long term life of the chassis. Hence it is imperative that calculate deformation.
weld distortion be minimized. In this project the chassis 5. Perform above steps with different weld path’s to
assembly is to be analysed for the existing weld sequence, calculate minimum deformation and hence the optimum
and check deficiencies in the process. Then based on weld path.
results an improvised sequence is to be recommended. 6. Calculate the distortion and residual stresses for the
optimum path.
7. Validation of results will be done by comparing
distortion at optimum weld path with same path at
designated working conditions.

5. CLEANING OF MODEL
Much of the hard work in setting up an analysis is related
to creating the desired mesh. However; complicated
geometry can make this a difficult task. Repair utilities
that can be used to clean up complicated CAD geometry.
These simplifications, the various modelling operations
are available in Design Modeller.
Figure-1 Geometric CAD model of commercial vehicle
chassis showing 1- C channel, 2- Rib assembly.

Welding process is carried out in between plates present

on rib six rib assemblies and respective C- channels. There
are total 24 numbers of plates present in model.

Before Cleaning

Figure-2 Shows Rib assembly and 1,2,3,4 are plates After Cleaning
present in assembly Figure-3 Before and after cleaning of same part

© 2016, IRJET ISO 9001:2008 Certified Journal Page 657

 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 03 Issue: 08 | Aug-2016 www.irjet.net p-ISSN: 2395-0072

6. MESHING OF MODEL Table -2: Welding Time required for each plate (included
in full chassis also).
Table -1: Mesh control of a model (* Size varies as per
Welding Time
number of nodes) Sr. No. Plate Number
(in second)
1 1 to 4 90
Element Type SOLID 186
2 5 to 8 60
Method of mesh control Hex Dominant 3 9 to 12 80
size 25* 4 13 to 16 60
Statistics 5 17 to 20 60

No. of nodes 1,82,207 6 21 to 24 60

No. of elements 28,421 Analysis starts with applying different boundary conditions
to a given model basically boundary condition contains,
 Provision of convection
 Provision of heat flux
7. PRELIMINARY ANALYSIS WITH HALF CHASSIS
Result For Transient Thermal Analysis
There are mainly four reasons behind the division of
model into half, they are as follows,
 Development of standard operating procedure for
analysis.
 Computer system limitation.
 Increased number of mesh nodes and elements.
 Increased solution time.

7.1 TRANSIENT THERMAL ANALYSIS

Figure-4 Temperature output for first four plates in
Input Parameters
Transient thermal analysis is 2620°C.
 Type of welding = MIG welding.
 Maximum Temperature reached during welding =
approximately 3,000 °C.
7.2 TRANSIENT STRUCTURAL ANALYSIS
 Welding Speed = 12 inches/min= 5.08 mm/sec.
 Convection film coefficient = 0.5 w/m2. Input Parameter
 Flux required = 5.2w/mm2.
 Result of transient thermal analysis
Time Calculation  Source time and analysis time
Welding time calculation for first four plates of model of Result of Transient Structural Analysis
chassis are shown,

Welding speed

Here, length of weld pool = 445mm

≈ 90 sec
For example, transient thermal analysis for first four
plates is carried out in three numbers of steps. First two Figure-5 Deformation for first four plates in Transient
steps involve welding operation and third step dedicated Structural analysis is 2.9654 mm.
to cooling which equals to 90 seconds. Each step carries
time equal to 90 seconds according to above calculation. From the analysis of half model of chassis standard
Step number three ends at 270 second. operating procedure is properly developed to calculate the
final deformation.

© 2016, IRJET ISO 9001:2008 Certified Journal Page 658

 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 03 Issue: 08 | Aug-2016 www.irjet.net p-ISSN: 2395-0072

8. Analysis of Full Model of Chassis 10. RESULTS

10.1 Results of Path 1
Analyses full model of chassis is analysed for three
different cases of mesh sizes described as follows,
 Approximately 50, 000 nodes.
 Approximately 80,000 nodes.
 Approximately 1, 80,000 nodes.

Note: It is difficult to achieve exact number of nodes in

Ansys therefore approximate values are taken under
consideration.

Result comparison at diffrent number of Figure-6 Deformation due to Transient Structural analysis
nodes for weld path 1 is 4.6345 mm.
20
18
10.2 Results of Path 2
16
14
12
10 50,000 nodes
8
80,000 nodes
6
4 1,80,000 nodes
2
0
0 2 4 6 8
Figure-7 Deformation due to Transient Structural analysis
Chart -1 Graph of deformation vs. no. of plate for weld path2 is 6.6295 mm.

From the graph of number of plates vs. deformation, it is 10.3 Result of Path 3
concluded that the results for different mesh size are
approximately same and variation is within the
considerable limit. Hence further analysis can be done on
approximately 50,000 nodes, as it reduces the solution
time and the results are approximately same compared to
fine mesh.

9. Weld Path Optimization

Analysis are done with full chassis by using different weld
path to achieve minimum deformation, those cases are,
Figure-8 Deformation due to transient structural analysis
 Path 1 linear application of heat flux at all plates for weld path 3 is 4.8095 mm.
serially throughout the length.
 Path 2 alternate application of heat flux to the rows of 10.4 Result of Path 4
plates throughout the length.
 Path3 application of heat flux to the top plates of
consecutive rows simultaneously and then respective
bottom plates simultaneously. In this case 4 plates
getting weld at a time. In this case 4 plates getting
weld at a time.
 Path 4 simultaneous application of heat flux to the top
plates then bottom plates of alternate rows
throughout the length.
According to the weld path discussed above analysis are
carried out at number of nodes equal to 50,000 Figure-9 Deformation due to transient structural analysis
(approximately) and results are discussed as below, for weld path 4 is 4.3606 mm.

© 2016, IRJET ISO 9001:2008 Certified Journal Page 659

 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 03 Issue: 08 | Aug-2016 www.irjet.net p-ISSN: 2395-0072

Table-3 Results of analysis in tabular form. remarks based on the analysis performed on chassis
Sr. no Weld Path Deformation(mm) model.
2. Firstly analysis of half model is done to develop the
1 Weld Path 1 4.3645 standard operating procedure. From the comparison
2 Weld Path 2 6.6295
of results at different mesh sizes it is concluded that
variation in results is within acceptable limit, hence
3 Weld Path 3 4.8095 approximately 50,000 nodes mesh size is fixed for
further analysis.
4 Weld Path 4 4.3606
3. Full model of chassis is analyzed under 4 different
weld paths to find out the minimum distortion, and
hence optimum weld path (Simultaneous application
From the above results, it is seen that Weld Path 4 has
of heat flux to the top and bottom plates of alternate
minimum deformation of 4.3606 mm which is minimum of
rows throughout the length) is decided. Optimum
all the cases stated above, thus weld path 4 (Simultaneous
weld path has minimum distortion of 6.4885 mm
application of heat flux to the top and bottom plates of
alternate rows throughout the length) is considered as the
FEA results and Experimental results are in close
optimum path for welding for given model of chassis.
resemblance and proved that the optimum weld path
decided by FEA analysis is correct and is validated by
11. Distortion due to Welding in Optimized Case
experimental distortion results is 6.8 mm. % error
Welding distortion, a result of the non-uniform expansion between FEA and experimental result’s is 4.5% which is
and contraction of weld metal and adjacent base metal less than the allowable error (10%) in FEA.
during the heating and cooling cycle of the welding
process, is a major concern during the fabrication of a From the above discussion it is concluded that FEA result
welded structure. is reliable and FEA is an ideal tool for such nonstandard
Distortion is nothing but the deformation beyond yield situations and also for research work.
strength, so considering the following properties in
further analysis to calculate Distortion,
ACKNOWLEDGEMENT
 Yield Strength = 2.5E+08 Pa Thanks to Mr. Vinay Patil for his valuable contribution in
 Tangent Modulus = 1.45E+09 Pa developing the research work.

13. References
11.1 Result For Transient structural Analysis
[1]. S. R. Patil, C. A. Waghmare, “Optimization of MIG
welding parameters for improving strength of welded
joints” E-ISSN2249–8974 Int. J. Adv. Engg. Res. Studies /
II/ IV/July-Sept., 2013/14-16

[2]. A .Chattopadhyay, G. Glinka, M. El-Zein, J. Qian, R

Forams, “STRESS ANALYSIS and FATIGUE of welded
structures, 1106_0037_WELDING_7_8_2011.indd 20

[3]. Chang Doo Jang, Yong Tae Kim, Young Chun Jo, Hyun
Su Ryu, “Welding Distortion Analysis of Hull Blocks Using
Equivalent Load Method Based on Inherent Strain”,10th
Figure-10 Distortion due to transient structural analysis International Symposium on Practical Design of Ships and
for optimum weld path is 6.4885 mm Other Floating Structures Houston, Texas, United States of
America © 2007 American Bureau of Shipping.
12. Conclusions
[4]. C.M.Cheng, C.P.Chouy, I.K.Lee and H.Y.Lin, “Distortion
1. The Project basically focused on an Analysis and Analysis of Single V-groove Butt Welding on Heat
Optimization of weld distortion in automobile chassis. Treatable Aluminium Alloys”, J. Mater. Sci. Technol., Vol.21
The analysis of chassis model was done in ANSYS 15.0 No.5, 2005
Workbench. The results were supported with an
experimental validation for verifying the actual
distortion and FEA results. Following are concluding

© 2016, IRJET ISO 9001:2008 Certified Journal Page 660

 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 03 Issue: 08 | Aug-2016 www.irjet.net p-ISSN: 2395-0072

[5]. Y. P. Yang, R. Dull, H. Castner, T. D. Huang, and D.

Fanguy, “Material Strength Effect on Weld Shrinkage and
Distortion”, NOVEMBER 2014 / WELDING JOURNAL 421-s

[6]. N. R. Mandal and C. V. N. Sundar, “Analysis of Welding

Shrinkage”, WELDING RESEARCH SUPPLEMENT I 237-s
JUNE 1997
[7]. Z. M. Liu, C. S. Wu, Y. K. Liu, and Z. Luo, “Keyhole
Behaviour’s Influence Weld Defects in Plasma Arc Welding
Process”, SEPTEMBER 2015 / WELDING JOURNAL 281-s

[8]. Dean Deng, Ninshu Ma, and Hidekazu Murakawa,

“Finite Element Analysis of Welding Distortion in a Large
Thin-plate Panel Structure”, Transactions of JWRI, Vol.40
(2011), No. 1

[9]. Nuraini, A.A, Zainal A.S.M. and Azmah Hanim, M.A,

“The Effect of Welding Process Parameter on Temperature
and Residual Stress in Butt-Joint Weld of Robotic Gas
Metal Arc Welding”, Australian Journal of Basic and
Applied Sciences, 7(7): 814-820, 2013 ISSN 1991-8178

[10]. Z. Barsoum, “Determination of Weld Loads and

Throat Requirements Using Finite Element Analysis with
Shell Element Models--A Comparison with Classical
Analysis”, Welding research supplement April 1999

[11]. Dragi Stamenkovic, Ivana Vasovic, “Finite Element

Analysis of Residual Stress in Butt Welding Two Similar
Plates”, Scientific Technical Review,Vol.LIX,No.1,2009

[12]. K.T. B. Padal, K. Ajay Kumar, “Optimization and

Control of Welding Distortion in Hatch Cover Girders”,
IOSR Journal of Mechanical and Civil Engineering (IOSR-
JMCE) e-ISSN: 2278-1684, p-ISSN: 2320-334X, Volume 11,
Issue 2 Ver. VII (Mar- Apr. 2014), PP 22-29

© 2016, IRJET ISO 9001:2008 Certified Journal Page 661