Optimization of Gas Metal Arc Welding Process Parameters for Welding Dissimilar Steels. Santhosh .R.V.1, Dr. R. Sivarmakrishnan2 1 Department of Production Technology, Madras Institute of Technology, Anna University, Chennai – 600044, India Phone: 09003149467, Email: santhrekha@gmail.com 2 Associate Professor, Department of Production Technology, Madras Institute of Technology, Anna University, Chennai Abstract In this work, optimisation of process parameters used for gas metal arc welding of dissimilar metal joint involving duplex stainless steel (DSS) to grade ASTM/UNS S32205 and weathering steel CORTEN-A to grade ASTM A242 is attempted. Such dissimilar metal joint finds application in transportation industry, especially in rail car fabrication. Square butt joint between 2mm thick sheets welded using gas metal arc welding (GMAW) process with CO2 as shielding gas and flux cored wire of grade 309L as filler material has been considered for the study. L9 Taguchi array has been used for optimisation with tensile strength of the resulting weld joint as the required quality characteristic. The process parameters of GMAW process, such as voltage, wire feed rate and welding speed at three levels are considered for optimisation. Influence of each factor has been examined with ANOVA. Optimum set of parameters were arrived at and it was also found that, voltage is the most influential factor. Confirmation test has been carried out to verify the results. Keywords: CORTEN Steel, Dissimilar welding, Duplex Stainless Steel, GMAW process, Optimization. 1. Introduction Duplex stainless steel which has advantages like high strength to weight ratio, easy to form, good weldability characteristics is presently under consideration for rail coach fabrication [1]. DSS, owing to its appealing aesthetic and other qualities mentioned above, find wide application in the side wall, end wall and roof assemblies of rail car. CORTEN steel is well known for its atmospheric corrosion resistance properties and is widely used for fabrication of under gear and floor side assemblies of the rail car. Hence joining of these dissimilar steel finds lot of application in rail coach fabrication. Also, joining dissimilar metal is indispensable in manufacturing, constructing advanced equipments, and machinery. Fusion Welding is a widely used method for joining dissimilar steels. Optimization of weld parameters for such fusion weld joints will help in achieving a sound weld joint free from defects [1]. Since such kinds of joints are widely used, this study assumes a lot of significance in terms of safety, quality and life cycle improvement of the product. GMAW is widely used in fabrication activity as a semi mechanized, fully mechanized or automatic process. It is expected that GMAW will continue to evolve to allow better arc control, better bead contour control, better deposition control and higher productivity. GMAW process will retain its dominant position and the latest research supports the application and further development of these processes [2]. A Successful weld between dissimilar metals is one that is as strong as the weaker of the two metals being joined, i.e., possessing sufficient tensile strength and ductility so that the joint will not fail in the weld. The dissimilar weld joint showing higher resistance than the base metal in the tensile test and good performance in the bend test shows soundness of the joint [3]. The study employs Taguchi Orthogonal Array Technique for arriving optimum value of process parameters with higher the better characteristics for the quality characteristic. The experimental design proposed by Taguchi involves use of orthogonal arrays to organize the parameters affecting the process and the levels at which they should be varied. Instead of having to test all possible combinations like the factorial design, Taguchi method tests pairs of combinations. This allows for the collection of the necessary data to determine which factors most affect product quality with a minimum amount of experimentation, thus saving time and resources [4, 5, and 6]. The objective of this project work is to study and optimize the process parameters for dissimilar welding involving Duplex stainless steel and Weathering Steel to grade CORTEN-A using GMAW process to get a sound weld joint. An Orthogonal array, signal to noise (S/N) ratio and analysis of variance (ANOVA) are employed to arrive at the optimum value of the parameters. 3. Experimental details 3.1. Experimental setup Square butt weld joint using 2mm thick Base Metals was studied in this experiment. A KEMPPI (Finland) make GMAW power source of 300Amps capacity with mechanized torch movement was used for making the weld joint. The mechanized torch movement ensured precise control over the weld speed. Voltage and wire feed rate are set at the GMAW power source which has provision to set these values to one decimal place precision. Flux Cored Wire Electrode of diameter 1.2mm of specification AWS5.22 E309L T1-1 suitable for this dissimilar welding was used as filler material with CO2 as shielding gas. Welding was done in down hand position. Gas flow rate of 15 LPM, stick-out of 20mm and a root gap of 1 to 1.1mm were maintained. 3.2. Materials Chemical composition study for both DSS and CORTON steel samples were carried out using Atomic Emission Spectrophotometry. The Chemical Composition of the samples obtained is as tabulated below in table-1. Element Present C Mn Si S P Cu Cr Ni V Mo Nh N ASTM-A 242 0.066 0.343 0.463 0.005 0.076 0.403 0.424 0.298 <0.001 <0.001 0.0007 - ASTM-S 32205 0.026 1.570 0.676 0.001 0.017 0.158 22.56 5.59 - 2.9 0.015 0.16 Table-1, Chemical composition of ASTM-A 242 and ASTM-S 32205 samples 3.3. Preparation of weld coupon Tensile strength and ductility of a weld joint can be measured by subjecting the standard specimens drawn out from the test coupons to tensile test as per ISO 4136:2001 and bend tests as per ISO 5173:2009 respectively using Universal Testing Machine. In the case of dissimilar weld under tensile test, separation is expected to happen only at the weaker of the two parent materials and not on the weld joint. Bend test determines ductile behaviour of the specimen over a given radius and provides insight into the modulus of elasticity and the bending strength of the material. Hardness of different regions in the weld joint is measured using a Vickers hardness test as per ISO 9015-1 & 9015-2 standards. Vickers Hardness Test is used to find out the hardness of the material which can be correlated to the strength of the material. Sample sizes of weld coupon required for testing these mechanical strength properties was finalized based on Tensile & Bend test requirements which are illustrated in Figure. 5 & 6 respectively. Accordingly weld pad of size 80mm wide, 300mm long and 2mm thick with run-on and run off plates was prepared. Figure.5, Test sample for Transverse tensile test Figure.6, Test sample for Bend Tests 4. Design of Taguchi’s orthogonal array The Taguchi method is best used when there are an intermediate number of variables (3 to 50), few interactions between variables, and when only a few variables contribute significantly. The arrays are selected by the number of parameters (variables) and the number of levels (states). Benefits of Orthogonal Array method are (a) Conclusions valid over the entire region spanned by the control factors and their settings, (b) Large saving in the experimental effort, and (c) Analysis is easy. 4.1. Selection of process parameters for the study Process Parameters that influences quality of weld joint is dealt in EN ISO 15609-1:2004(E) standards. Based on this and field expert opinion the parameters which are variable and hence to be considered for optimization for a GMAW process are welding current, arc voltage, wire feed rate, and travel speed. GMAW process is having constant voltage characteristics in which the welding current is decided by wire feed rate. Thus arc voltage, wire feed rate, and travel speed are selected as the process parameters that are to be optimized for the present study. 4.2. Experiments based on Taguchi’s orthogonal array As explained above, process parameters to be optimized for the present study are Voltage [V], Wire feed rate [F], and welding speed [S]. Transverse Tensile Strength is selected as the quality characteristic for performance measure. Strength of the weld joint can be found out using Transverse Tensile Test. During trial welding of the dissimilar samples, ductile failure occurred on the CORTEN-A material but beyond weld and heat affected zone which indicates that weld metal is stronger than the weaker of the two base metals. The above joint, arrived at after many screening trials was made with following process parameters, i.e., Voltage [V] = 25volts, Wire Feed rate [F] = 5.2 meter/minute with corresponding current of 102 Amps and Welding speed[S] = 60 meter / hour. Design matrix of three welding parameters (V, F & S) each at three levels was selected for this optimization study in order to obtain a weld joint having highest possible tensile strength value. The three levels for these process parameters (factors) are tabulated in table-2 below. Experimenting with these three parameters each with 3 levels will call for a full factorial array with 27 possibilities. But using Taguchi Orthogonal array method, same can be analyzed with nine experiments instead of 27 possibilities. The degrees of freedom of the orthogonal array should be greater than or at least equal to the degrees of freedom of all the process parameters. The interaction effect between the parameters is not considered. The total degrees of freedom of all process parameters are 8. Hence, L9 [3(3)] orthogonal array was chosen which has 8 degrees of freedom. Process Parameters Level 1 Level 2 Level 3 Voltage [V] 23.0 25.0 27.0 Wire Feed rate [F] 4.8 5.2 5.7 Welding speed [S] 55.0 60.0 65.0 Table-2, Process Parameters and its levels for the experiment 4.1. Orthogonal array used for the study. Each of the 9 experiments is conducted with a pre-specified combination of voltage, wire speed and welding speed as shown in table-3 below. Table-4 shows the actual values used and Table-5 shows the results of the experiment. Exp. No Voltage (V) Wire feed rate (Meter/Min) Weld speed (Meter/hr) 1 1 1 1 2 1 2 2 3 1 3 3 4 2 1 2 5 2 2 3 6 2 3 1 7 3 1 3 8 3 2 1 9 3 3 2 Exp. No Voltage (V) Wire feed rate (F) (Meter/Min) Weld speed (S) (meter/hr) 1 23 4.8 55 2 23 5.2 60 3 23 5.7 65 4 25 4.8 60 5 25 5.2 65 6 25 5.7 55 7 27 4.8 65 8 27 5.2 55 9 27 5.7 60 Table-3, Orthogonal array used Table-4, L9 array with parameter values used for the experiment. Exp. No. UTS (MPa) Location of Failure Result of Face bend test Result of Root bend test 1 494.32 HAZ CS Satisfactory Satisfactory 2 505.01 Parent metal CS Satisfactory Satisfactory 3 492.56 Parent metal CS Satisfactory Satisfactory 4 510.01 Parent metal CS Satisfactory Satisfactory 5 535.45 Parent metal CS Satisfactory Failed at HAZ CS 6 555.25 Parent metal CS Satisfactory Satisfactory 7 453.32 HAZ CS Failed Satisfactory 8 488.66 Fusion line & HAZ Cavity opened Cavity opened 9 471.70 HAZ CS Satisfactory Failed HAZ CS Table-5: Results of Tensile and Bend Test using orthogonal array 4.2. Loss function and s/n ratio. As discussed, the weld strength belongs to the higher-the-better quality characteristic. The signal to noise ratios (S/N), which are log functions of desired output, serve as the objective functions for optimization that help in data analysis and the prediction of the optimum results. Generic expression for S/N ratio as per TAGUCHI method is n = -10 log10 (Ci) ………….(1) where Ci is the mean of sum of squares of reciprocals of measured data. S/N ratio, its overall mean value, deviation from mean and square of deviation etc., are tabulated as below in table-6. Exp. No [V] [F] [S] Meas-ured Data [UTS] mean of square of reciprocal [Ci] S/N Ratio [n] overall mean value for S/N [m] Square of S/N Deviation of S/N from mean (I) square of deviation S/N to mean 1 23 4.8 55 494.32 0.00000409 53.88 53.916 2903.07 2923.13 2899.73 -0.097 0.009 2 23 5.2 60 505.01 0.00000392 54.06 0.089 0.008 3 23 5.7 65 492.56 0.00000412 53.84 -0.128 0.016 4 25 4.8 60 510.01 0.00000384 54.15 2932.39 2978.36 3012.88 0.175 0.030 5 25 5.2 65 535.45 0.00000349 54.57 0.597 0.357 6 25 5.7 55 555.25 0.00000324 54.89 0.913 0.833 7 27 4.8 65 453.32 0.00000487 53.12 2822.59 2833.41 2859.39 -0.849 0.721 8 27 5.2 55 458.66 0.00000475 53.23 -0.197 0.039 9 27 5.7 60 471.70 0.00000449 53.47 -0.503 0.253 Total 485.24 2.267 Table-6,L9 matrix array table. 4.3. Factor effects The effect of a factor level is defined as the deviation it causes from the overall mean. Using the S/N ratio data available in Table- 5 the average of each level of the three factors is calculated. These average values are shown in table- 7 below. They are separate effect of each factor and are commonly called main effects which are graphically represented in figure-7 below. Factor level 1 Level 2 Level 3 Factor A [V] 53.932 54.540 53.270 Factor B [A] 53.720 53.960 54.070 Factor C [S] 54.000 53.897 52.650 Table 7: Average of each level of the three factors Figure 7,S/N ratio graph 4.4 Analysis of variance Different factors affect the quality characteristic, i.e., the tensile strength, in this study to a different degree. The relative magnitude of the factor effects are listed in Table- 6 above. A better feel for the relative effect of the different factors is obtained by the decomposition of variance, which is commonly called as analysis of variance (ANOVA) and is tabulated in table-8 below. . factor Level 1 Level 2 Level 3 degree of freedom sum of squires Mean Squire= Sum of squires/degree of freedom % ge contribution Factor A [V] 53.93 54.54 53.27 2 1.752 0.876 77.3 Factor B [A] 53.72 53.96 54.07 2 0.305 0.152 13.4 Factor C [S] 54.00 53.89 52.65 2 0.195 0.097 8.6 Error 2 0.015 0.7 TOTAL 8 2.267 100.0 Table.8: ANOVA table 5. Confirmation tests Tests are now carried out to verify the improvement of quality of characteristics using the optimal level of welding process parameters such as welding voltage at level 2, wire feed rate at level 3 and welding speed at level 1. The resultant UTS value is around 555N/mm2. Hence optimal inputs determine greater tensile strength of the weld. The root and bend tests also revealed that the failure does not occur at weld joint. Fig-9, Tensile Test Specimens Fig-10,Test specimens for bend tests Figure-11, Test Specimens showing failure at CS side away from weld and HAZ Figure-12, Test Specimens showing satisfactory weld properties 5.1. Hardness survey. The hardness survey on the above referred joint carried out using a Vickers Hardness Testing Machine with a load of 10kg has given the values as mentioned in Figure-13. The hardness value is observed to be in accordance with the failure pattern showing that ductile failure under tensile load has occurred at the area of minimum hardness. Figure-13, Hardness profile at various regions Figure-15, Macrograph of the weld Joint 6. Conclusions Optimum welding parameters for welding dissimilar metal joints of 2mm thick steel sheets of DSS-2205 and Corten-steel were arrived at. The optimum values of GMAW process parameter using 1.2mm diameter flux cored electrode of 309L grade wire with mechanized torch travel for obtaining a sound dissimilar metal joint between 2mm thick DSS 2205 and CORTEN-A steel sheets with square butt joint using CO2 as shielding gas are found to be Voltage = 25Volts, Wire Feed = 5.7 meter per minute and Welding Speed = 55 meter per hour. When tested for tensile strength test samples with optimum welding parameters failed at the weaker of the two parent metals, i.e., in the Corten-A steel and this failure occurred much away from the weld joint and HAZ. During bend test also there was no failure at the weld joint. Hence the weld joint obtained was strong and sound. Acknowledgment. Authors are thankful to management and staff of Advanced Welding Training Institute of Integral Coach Factory, Chennai-38 and Madras Institute of Technology, Chennai-44 for extending their support in carrying out this work. The cooperation extended by Mr.V.Gnanamurthy, CMS/AWTI, and Mr. N. Hari Hara Swaminathan, Instructor/AWTI, is highly appreciated. References [1] Piotr Gotkowski et al. “Duplex Steel Welding in Construction of Railway Vehicles”,Welding International-2013, selected from Bileteyn Instytytu Spawalinictiva 2011(5), pp 43-47, 2013. [2] Kah.P and Martikainen. J, “Current Trends In Welding Processes And Materials: Improve In Effectiveness”, Rev. Adv. Mater. Sci. Lappeenranta University of Technology, Lappeenranta, Finland, pp-189-200, 2012. [3] Bravo Ivan Mendoza et al. “Dissimilar Welding of Super Duplex Stainless Steel/HSLA Steel for Offshore Applications Joined by GTAW” Scientific Reserch, published online July 2010 (http://www.SciRP.org/journal/eng), pp-520-528, 2010. [4] Anoop.C.A, Pawan Kumar (2013), ‘Application of Taguchi Methods and ANOVA in GTAW Process Parameters Optimization for Aluminium Alloy 7039’, International Journal of Engineering and Innovative Technology (IJEIT) Volume 2, Issue 11, pp-54-58, 2013. [5] Lenin N., et al. “Process Parameter Optimization in ARC Welding of Dissimilar Metals”, Thammasat International Journal of Science and Technology. Vol. 15, No. 3, pp 1-7, 2010. [6] Taban. E, et al. “ Evaluation of Dissimilar Welds between Ferritic Stainless Steel Modified 12% Cr and Carbon Steel S355”, Welding Research Journal of American Welding Society, pp 291S - 297S , 2008 [7] Elin M. Westin, ‘Microstructure and properties of welds in the lean duplex stainless steel LDX 2101’, Scientific thesis for the degree of Doctorate of Engineering Royal Institute of Technology, Department of Materials Science and Engineering,SE-100 44 Stockholm, Sweden, pp 10-59,2010. [8] Tewari .S. P et al. “Effect of Welding Parameters On The Weldability of Materials: Improve in Effectiveness”, International Journal of Engineering Science and Technology Vol. 2(4), pp 512-516, 2010. [9] Sapakal.S.V, Telsang M. T., “Parametric Optimization of MIG Welding Using Taguchi Design Method”, International Journal of Advanced Engineering Research and Studies, Vol. I/ Issue IV/July-Sept., pp 28-30, 2012. [10] Suban.M, Tusek J, “Dependence of melting rate in MIG/MAG welding on the type of shielding gas used”, Journal of Material Science Processing Technology, Welding Institute, Lubljana, Slovania, pp-185-192,119 (2001). [11] Sukhomay Pal et al. “Optimization of quality characteristics parameters in a pulsed metal inert gas welding process using grey-based Taguchi method” International Journal on Advanced Manufacturing Technology, issue-44, pp-1250–1260, 2009. [12] Weglski.M.St, et al. “Effect of welding current on metal transfer in GMAW” , International Science Journal of World Academy of Materials and Manufacturing Engineering, Volume 33, Issue 1 pp 49-56,2008 [13] Labanowski. J., “Mechanical properties and corrosion resistance of dissimilar stainless steel welds” International Scientific Journal of the Polish Academy of Sciences, Archives of Materials Science and Engineering Volume 28, Issue 1, pp 27-33, 2007. [14] ZHANG Xue-yong et al. “Study on the Microstructure, Mechanical Properties and Corrosion Resistance of a Novel HSLA Steel”, National High Technical Research and Development Programme of China (No.2003AA331020), 1977. . Proceedings of National Conference on Recent Advances on Materials and Manufacturing Engineering RAMM 2014