20120140506017

International Journal of Advanced Research in Engineering and Technology (IJARET), ISSN 0976 –
6480(Print), ISSN 0976 – 6499(Online) Volume 5, Issue 6, June (2014), pp. 112-120 © IAEME
112
OPTIMIZATION OF MIG WELDING PROCESS PARAMETERS TO
CONTROL ANGULAR DISTORTION OF A FILLET WELD IN AN
EARTHMOVING EQUIPMENT MANUFACTURING PLANT
Mamatha.K1
, Mr.H.V.Vasuki2
, Mr.Jagadish Mogaveera.B3
, Dr.C.K.Nagendra Guptha4
1
Student, II year M.Tech (MEM), Department of IEM, RVCE, Bangalore
2
Assistant General Manager, Quality Department, Earthmoving Equipment Manufacturing
Company, Bangalore
3
Senior Engineer Support Services, Fabrication Department, Earthmoving Equipment Manufacturing
Company, Bangalore
4
Associate Professor, Department of IEM, RVCE, Bangalore
ABSTRACT
Distortion is the major problem faced by the fabrication engineers. Departure from initial
dimensional specifications in a fabricated structure or component as a consequence of welding is
termed as welding distortion. When distortion exceeds the acceptable limits, rework of the fabricated
components occurs thus leading to increased rework time and cost. The aim of this study was to
investigate the optimization process parameters for Metal inert gas welding (MIG) to control the
angular distortion measured in terms of Deck height. The experiment chosen was fractional factorial
design matrix and the process parameters studied were welding speed, welding current and voltage.
Response Surface Methodology (RSM) was applied to optimize the MIG welding process parameters
to reduce the angular distortion of the fillet welds. Analysis of variance (ANOVA) is also applied to
identify the factors that affect the response. The model was validated using the optimized parameter
obtained from the analysis and the results obtained were found to be within the acceptable limits.
Keywords: Analysis of Variance, Angular distortion, Design of Experiment, Fillet weld, Response
Surface Methodology.
I. INTRODUCTION
Gas Metal Arc Welding (GMAW), sometimes referred to by its subtypes Metal Inert Gas
(MIG) welding or Metal Active Gas (MAG) welding, in which a continuous and consumable wire
INTERNATIONAL JOURNAL OF ADVANCED RESEARCH
IN ENGINEERING AND TECHNOLOGY (IJARET)
ISSN 0976 - 6480 (Print)
ISSN 0976 - 6499 (Online)
Volume 5, Issue 6, June (2014), pp. 112-120
© IAEME: http://www.iaeme.com/IJARET.asp
Journal Impact Factor (2014): 7.8273 (Calculated by GISI)
www.jifactor.com
IJARET
© I A E M E

113
electrode and a shielding gas are fed through a welding gun. A wide range of materials may be
joined by Gas metal arc welding—similar metals, dissimilar metals, alloys, and nonmetals. GMAW
welding is used because of its advantages over other welding techniques like high welding speeds,
less distortion, no slag removal required, high weld metal deposition rate, high weld quality, precise
operation, etc [1].
The problem encountered in a welding process of model where rework of the Left hand (LH)
and Right hand (RH) Deck products occur due to nonconformance of height with respect to the
required standard specifications which is caused due to higher heat input leading to welding
distortion. The time spent on rework of the products is very high. This increases costs and lowers
customer satisfaction for internal and external customers.
The optimization of MIG welding process parameters on alloy steel work piece using grey
relational analysis method has been discussed. The objective function was chosen in relation to
parameters of MIG welding bead geometry. The ANOVA is applied and identified that the welding
current was the most significant factor by Dinesh Mohan Arya, et al. (2013) [2]. An investigation
was carried out by using Taguchi’s Parameter Design methodology for Parametric Study of Gas
Metal Arc Welding of Stainless Steel & Low Carbon Steel to predict tensile strength & hardness.
Pawan Kumar, et al. (2013) [3], concluded that arc Current significantly affects the hardness. The
optimization of weld process parameters such as weld current, root gap, argon gas flow rate and weld
speed with Taguchi approach L8 orthogonal array for the transverse distortion control applied to MS
structures of 3 mm thickness with TIG weld process. ANOVA was applied for the optimization of
weld parameters control. S. Akella, et al. (2013) [4] concluded from these experiments that Root gap
has a major contribution of 43% and Weld current of 36% influence on distortion.
The major gap found was application of the optimization of MIG welding parameters to
determine the optimal settings of the process factors to control or reduce the angular distortion of the
single pass fillet welds using DOE and RSM, when used could result in solving the problem more
accurately.
The present work is aimed to study the influence of MIG process parameters and their
optimization for the angular distortion control which is caused due to higher heat input and is
measured in terms of Deck height. However, the weld process control parameters optimization with
Response surface method towards weld distortion for the single pass fillet weld studies has been
rarely reported in the literature. The purpose of this work is to optimize the MIG weld process
parameters to control weld distortion as major output measured in terms of Deck height using RSM
optimization technique and the significant factors that affect the response is identified using
ANOVA.
II. EXPERIMENTAL WORK
The basic experimental design matrix is of fractional factorial: 2×3×3= 18 trials, which
indicate three input variables. One input has two levels, and the other two each have three levels with
single replication. The experiment is conducted at the workstation where the Frame along with the
Deck - LH and RH are full welded on the positioner. These experiments were conducted as per the
design matrix using Fronius manually operated welding equipment. Copper-coated steel wire of 1.2
mm diameter (E70S6 type solid wire), in the form of coil was used, with a shielding gas of Argon
(85%) and CO2 (15%). Medium carbon steel (grade IS – 2062) specimens of length and width of
Deck is 3769 mm * 2700+/-3 mm and thickness of LH Deck is 172mm, and RH Deck is 215 mm
respectively. The length of the single pass fillet weld joint considered in this study is LH= 260mm
and RH= 180 mm. The gas flow rate is 17 to 19 liters per minute.

114
III. PLAN OF INVESTIGATION
The research work is carried out in the following steps [5].
Identification of important process variables and finding their upper and lower limits (i.e.
Range).
Design matrix and experiments conduction
Recording responses; angular distortion measured in terms of Deck height-LH and RH.
Development of mathematical models.
Checking adequacy of developed models.
Optimizing the process parameters using RSM.
3.1 Identification of factors and responses
The factors are chosen based on the rate of heat input during the welding process. In this
study, the experimental plan has three variables, namely, Welding Voltage, Welding Current, and
Welding Speed. The responses chosen were angular distortion measured in terms of Deck height –
LH and RH respectively.
3.2 Finding the limits or Range of the process variables
Working ranges of all selected factors are fixed by conducting trial run. Working range of
each process parameters was decided upon by inspecting the bead for smooth appearance without
any visible defects [5]. The upper and the lower limits were selected based on the Range of operation
of the welding process variables in order to find out the angular distortion measured in terms of Deck
height- LH and RH respectively within that ranges. The chosen welding parameters and their levels
with their units are given in Table 1.
Table 1: Welding parameters and their levels
Parameters Factor levels
Unit Notation Low level Medium
level
High level
Welding
speed
mm/sec X1 3-4 4-6 -
Current Amps X2 275-300 300-325 325-350
Voltage Volts X3 28-30 30-32 32-34
3.3 Design matrix
Design matrix chosen to conduct the experiments was fractional factorial design.18
experimental trails were conducted that make the estimation of linear quadratic and two way
interactive effects of process parameters on Deck height – LH and RH as shown in the Table 2 and
Table 3.

115
3.4 Experiments conduction
The experiments were conducted as per the design matrix at random, to avoid the bias of
conducting the experiments. The experiments are conducted for LH and RH Deck separately by two
operators respectively.
3.5 Recording the Responses
The angular distortion measured in terms of Deck height was measured using Tape with the
help of a straight edge. The measurements of both LH and RH Deck are carried out separately. The
measured values are given in Table 2. The welded joint specimen is shown in the Fig. 1.
Table 2: Design matrix and observed values of LH and RH Deck height
Experiment
no
Welding speed
(mm/sec)
Current
(amps)
Voltage
(volts)
LH Deck
height
(mm)
RH Deck
height
(mm)
1 3.3 295 29.5 44 44
2 4.4 280 28.5 46 41
3 3.4 305 28.5 44 42
4 5.3 305 28.5 46 45
5 3.7 350 29 47 42
6 5.2 350 29 46 42
7 3.6 300 30 44 42
8 4.4 300 30 47 44
9 3.7 325 30.5 45 43.5
10 5 325 30.5 45 43
11 3.7 350 30 45 41
12 4.9 350 30 45 42
13 3.1 300 32 44 43
14 4.4 300 32 47 41
15 3.5 320 32 46 39
16 5 320 32 46 44
17 3.5 350 32 42 43
18 4.4 350 32 41 45

116
Fig.1: Welded joint specimen
3.6 Developing the Mathematical model
The second degree polynomial model may adequately describe the curvature of the response
surface as a function of the input factors. The response function representing Deck height LH and
RH can be expressed as follows (1):
Y = f (X1, X2, X3) (1)
Where,
Y= Response variable, X1 = Welding speed, X2=Current, X3=Voltage
The second-order surface response model fit to the Deck height LH and RH response is as
follows (2):
Y=d0+d1X1+d2X2+d3X3+d11X1^2+ d22X2^2+ d33X3^2+d12X1X2+ d13 X1 X3+d23 X2 X3 (2)
Where d0 is the free term of the regression equation, the coefficient d1, d2 and d3 are linear
terms, coefficients d11, d22 and d33 are quadratic terms, and the coefficients d12, d13, and d23 are
the interaction terms.
The coefficients were calculated using Minitab 17 software package. The mathematical
models are developed after determining the coefficients. The developed mathematical models are
given as follows (3) and (4):
Deck height LH = -349 + 27.4 Weld speed + 1.556 Current + 6.5 Voltage – 1.29 Weld speed ×
Weld speed – 0.000250 Current × Current + 0.118 Voltage × Voltage – 0.0363 Weld speed ×
Current – 0.138 Weld speed × Voltage – 0.04209 Current × Voltage (3)
Deck height RH = -242 - 7.6 Welding speed + 0.322 Current + 16.4 Voltage + 1.18 Welding speed
× Welding speed – 0.00084 Current × Current – 0.336 Voltage × Voltage - 0.0249 Welding speed ×
Current + 0.204 Welding speed × Voltage + 0.0101 Current × Voltage (4)

117
3.7 Model adequacy checking
ANOVA technique was used to determine the significant factors that affect the response
variable and test the adequacy of the fitted model. The level of significance of the parameter is
assessed by magnitude of “p” value associated with it. If p values of factors are lesser than level of
significance 0.05, then this indicates that it has statistically significant effect on the response Deck
height. If p values of the factors are greater than 0.05, then this indicates that it has statistically
insignificant effect on the response Deck height. The Analysis of Variance for Deck height LH is
given in Table 3.
Table 3: Analysis of Variance output for Deck height LH using Minitab Software 17
Source DF Adj SS Adj MS F-Value P-Value Significant/
Insignificant
Model 9 38.6369 4.2930 4.66 0.021
Linear 3 8.9113 2.9704 3.23 0.082
Weld speed 1 0.0388 0.0388 0.04 0.843 Insignificant
Current 1 6.0247 6.0247 6.55 0.034 Significant
Voltage 1 0.0003 0.0003 0.00 0.986 Insignificant
Square 3 1.5475 0.5158 0.56 0.656
Weld speed*Weld
speed
1 1.3118 1.3118 1.43 0.267 Insignificant
Current*Current 1 0.0910 0.0910 0.10 0.761 Insignificant
Voltage*Voltage 1 0.2719 0.2719 0.30 0.602 Insignificant
2-Way Interaction 3 23.1492 7.7164 8.38 0.007
Weld speed*Current 1 2.8189 2.8189 3.06 0.118 Insignificant
Weld speed*Voltage 1 0.1740 0.1740 0.19 0.675 Insignificant
Current*Voltage 1 21.4401 21.4401 23.29 0.001 Significant
Error 8 7.3631 0.9204
Total 17 46.0000
From the above analysis of variance, p-value < .05 indicates that the two way interaction of
Current * Volt and Current has statistically significant effect on the response LH Deck height. The
residual plots for Deck height LH are shown in the Fig2.

118
11110000----1111
99999999
99990000
55550000
11110000
1111
Residual
Percent
44448888....000044446666....555544445555....000044443333....555544442222....0000
1111
0000
----1111
Fitted Value
Residual
1111....55551111....00000000....55550000....0000----0000....5555----1111....0000
4444
3333
2222
1111
0000
Residual
Frequency
11118888111166661111444411112222111100008888666644442222
1111
0000
----1111
Observation Order
Residual
Normal Probability Plot VersusFits
Histogram VersusOrder
Residual Plotsfor Deck height LH
Fig.2: Residual plots output for Deck height LH using Minitab Software 17
From the above, normality plot of the residuals follows a normal distribution as all the data
points lie on the fitted line. Both plot of residuals versus fitted values and plot of residuals versus run
order do not show any pattern. Thus both constant variance and independence assumptions are
satisfied.
Similarly the Analysis of Variance was used for Deck height RH and found that none of the
p-values was below .05, thus indicating that all the factors are statistically insignificant effect on the
RH Deck height. The normality plot of the residuals follow a normal distribution as all the data
points lie on the fitted line. Both plot of residuals versus fitted values and plot of residuals versus run
order do not show any pattern. Thus both constant variance and independence assumptions are
satisfied.
IV. OPTIMIZATION OF PROCESS PARAMETER USING RESPONSE SURFACE
METHODOLOGY
Response Optimizer helps identifying the combination of input variable settings that jointly
optimize a single response or a set of responses. The objective of the response optimization is to
target the response of deck height (45.5mm) with the upper (48mm) and lower specification limits
(43mm) for the response to control the welding angular distortion measured in terms of Deck height.
The optimum parameters setting conditions for achieving control of welding angular distortion to
meet the required standard specifications of Deck height are obtained using Minitab Software 17.
4.1 Results of Response optimization
Target the Response of Deck height LH. Optimum process parameters are
Welding speed= 4.5mm/sec, Current = 320.782 amps, Voltage = 32 volts.
Target the Response of Deck height RH. Optimum process parameters are
Welding speed= 5.5935 mm/sec, Current = 280 amps, Voltage = 28.50 volts.

119
V. RESULTS AND DISCUSSION
5.1 Validating the model using optimized parameter setting
Experiments were conducted to verify the optimized results for Deck height LH and RH.
Three weld runs were used at optimized values of welding speed, welding current, welding voltage.
The results obtained found to be satisfactory and are within the standard specifications of height
reducing the welding distortion and the results presented in Table 5 and Table 6.
Table 5: LH Deck Validation model using the optimized parameter setting
Experiment
no
Welding
speed
(mm/sec)
Current
(amps)
Voltage
(volts)
LH Deck
height (mm)
1 4.3 320 32 43
2 4.4 320 32 44
3 5.9 320 32 47
Table 6: RH Deck Validation model using the optimized parameter setting
Experiment
no
Welding
speed
(mm/sec)
Current
(amps)
Voltage
(volts)
RH Deck
height (mm)
1 5.8 280 28.5 43
2 3.6 280 28.5 44
3 4.1 280 28.5 45
VI. CONCLUSION
A Mathematical model has been developed to predict Deck height as a function of parameters
that can be measured and controlled independently in MIG welding . The optimization of welding
input parameters leads to determining the best settings and tolerances for Xs to optimize Ys, thus
reducing the welding angular distortion of fillet weld of Deck height. The optimized parameters are
found to be satisfactory and LH, RH Deck products occur within required standard specifications of
height. The time and cost spent on rework of the products LH and RH Deck is reduced. The
optimized parameter setting has been applied to a particular joint and this can be extended to other
joints of a particular model.

120
REFERENCES
[1]. Ajit Hooda, Ashwani Dhingra and Satpal Sharma, Optimization of MIG welding process
parameters to predict maximum yield strength in AISI 1040, IJMERR, 1(3), 2012, ISSN 2278
– 0149.
[2]. Dinesh Mohan Arya, Vedansh Chaturvedi, Jyoti Vimal, Parametric optimization of MIG
process parameters using Taguchi and Grey Taguchi Analysis, IJREAS, 3(6), 2013, ISSN:
2249-3905.
[3]. Pawan Kumar, Dr.B.K.Roy, Nishant, Parameters Optimization for Gas Metal Arc Welding of
Austenitic Stainless Steel (AISI 304) & Low Carbon Steel using Taguchi’s Technique,
International Journal of Engineering and Management Research, 3(4), 2013, ISSN No.: 2250-
0758, 18-22.
[4]. S.Akella, B. Ramesh Kumar, Distortion Control in TIG Welding Process with Taguchi
Approach, Advanced Materials Manufacturing & Characterization, 3(1), 2013.
[5]. P.Sreeraj, T. Kannan, Subhasis Maji, Optimization of weld bead geometry for stainless steel
cladding deposited by GMAW, American Journal of Engineering Research (AJER), 2(5),
2013, e-ISSN: 2320-0847 p-ISSN : 2320-09, 178-187.
[6]. V. Velmurugan and V.Gunaraj, Effects of process parameters on angular distortion of Gas
Metal Arc Welded Structural Steel Plates, Supplement to the welding journal, 2005.
[7]. Aniruddha Ghosh and Somnath Chattopadhyaya,, “Conical Gaussian Heat Distribution for
Submerged Arc Welding Process”, International Journal of Mechanical Engineering &
Technology (IJMET), Volume 1, Issue 1, 2010, pp. 109 - 123, ISSN Print: 0976 – 6340,
ISSN Online: 0976 – 6359.
[8]. P.Govinda Rao, Dr.CLVRSV Prasad, Dr.D.Sreeramulu, Dr.V.Chitti Babu and M.Vykunta
Rao, “Determination of Residual Stresses of Welded Joints Prepared under the Influence of
Mechanical Vibrations by Hole Drilling Method and Compared by Finite Element Analysis”,
International Journal of Mechanical Engineering & Technology (IJMET), Volume 4, Issue 2,
2013, pp. 542 - 553, ISSN Print: 0976 – 6340, ISSN Online: 0976 – 6359.
[9]. P.Govinda Rao, Dr.Clvrsv Prasad, Dr.S.V.Ramana and D.Sreeramulu, “Development of
GRNN Based Tool for Hardness Measurement of Homogeneous Welded Joint Under
Vibratory Weld Condition”, International Journal of Advanced Research in Engineering &
Technology (IJARET), Volume 4, Issue 4, 2013, pp. 50 - 59, ISSN Print: 0976-6480,
ISSN Online: 0976-6499.
[10]. Harshal K. Chavan, Gunwant D. Shelake and Dr. M. S. Kadam, “Finite Element Model to
Predict Residual Stresses in MIG Welding”, International Journal of Mechanical Engineering
& Technology (IJMET), Volume 3, Issue 3, 2012, pp. 350 - 361, ISSN Print: 0976 – 6340,
ISSN Online: 0976 – 6359.
[11]. L.Suresh Kumar, Dr.S.M.Verma and Dr.V.V.Satyanarayana, “Impact of Voltage on
Austentic Stainless Steel for the Process of TIG and MIG Welding”, International Journal of
Mechanical Engineering & Technology (IJMET), Volume 1, Issue 1, 2010, pp. 60 - 75,
ISSN Print: 0976 – 6340, ISSN Online: 0976 – 6359.
[12]. A. Chennakesava Reddy, “Studies on the Effects of Oxidation and its Repression in MAG
Welding Process”, International Journal of Advanced Research in Engineering & Technology
(IJARET), Volume 3, Issue 1, 2012, pp. 48 - 54, ISSN Print: 0976-6480, ISSN Online:
0976-6499.
[13]. Harshal K. Chavan, Gunwant D. Shelake and Dr. M. S. Kadam, “Effect of Heat Input and
Speed of Welding on Distortion in MIG Welding”, International Journal of Industrial
Engineering Research and Development (IJIERD), Volume 3, Issue 2, 2012, pp. 42 - 50,
ISSN Online: 0976 - 6979, ISSN Print: 0976 – 6987.

20120140506017

More Related Content

What's hot (20)

Viewers also liked (20)

Similar to 20120140506017 (20)

More from IAEME Publication (20)

Recently uploaded (20)

20120140506017