EXPERIMENTAL INVESTIGATION OF MULTIPASS TIG WELDING ...

242

Int. J. Mech. Eng. & Rob. Res. 2013 Sreejith S Nair, 2013

EXPERIMENTAL INVESTIGATION OF MULTIPASSTIG WELDING USING RESPONSE SURFACE

METHODOLOGY

Sreejith S Nair1*

*Corresponding Author: Sreejith S Nair,[email protected]

Stainless steel as a weld bead joint has very important applications in industry. TIG welding is animportant joining process used in manufacturing industries. The major needs in the welding arehigh tensile strength, good surface finish and hardness. In order to produce any product withdesired quality by welding, proper selection of process parameters is essential. The review ofliterature reveals that only few works have been reported on optimizing the welding of stainlesssteel. Therefore, this project is aimed at evaluating the optimal process environment whichcould simultaneously satisfy the requirements of both quality and strength. In this work Centralcomposite Design Methodology of Response surface methodology is used to conduct theexperiments. Analysis of variance is used to analyze the influence of parameters duringmachining. The results of the present work indicate that welding parameters have significantinfluence on tensile strength, hardness and penetration. The optimal process parameters soobtained have been verified by confirmatory experiments. In this project TIG welding of stainlesssteel was studied at different values of current, and electrode diameter but keeping electrodematerial, voltage and welding speed as constant. ATIG welding machine was used to weld 6mmthick stainless steel plates. The photographs of the welded seams were observed and tensiletesting of specimens was done to evaluate the mechanical properties of the welded joint.

Keywords: TIG welding, Response surface methodology, Central composite design, Hardness,Weld penetration, Tensile strength

INTRODUCTIONGas Tungsten Arc Welding (GTAW) is a widelyused process for metal joining. Its arc isestablished between the tip of a no

ISSN 2278 – 0149 www.ijmerr.comVol. 2, No. 3, July 2013

© 2013 IJMERR. All Rights Reserved

Int. J. Mech. Eng. & Rob. Res. 2013

1 Department of Mechanical Engineering, Dr. Mahalingam College of Engineering &Technology, Pollachi, Tamil Nadu, India.

consumable tungsten electrode and the workpiece with a shielding gas applied to protectthe arc and the weld pool area. The GTAWprocess can be used in welding a wide variety

Research Paper

243


of metals. It is typically used for root passeson pipes and thin-gauge materials. Its arc isvery stable and can produce high-quality andspatter-free welds without requiring much postweld cleaning. A typical GTAW system consistsof a power supply, a water cooler, a weldingtorch, cables, etc. For most of its applications,Direct Current Electrode Negative (DCEN)polarity is used and approximately 70% of thearc heat is applied into the work piece.Opposite to the Direct Current ElectrodePositive (DCEP) polarity, the DCEN polarityproduces a relatively narrow and deep weld.In order to achieve desirable welds, filler metalsare typically required during GTAW. Currently,there are two commonly used approaches forfilling the joint: cold-wire GTAW process andhotwire GTAW process. In the cold-wire GTAWprocess, the filler metal is directly added asis. To melt the wire faster, in the hot-wire GTAW,the filler metal is preheated by a resistive heatwhile it is being fed into the weld pool. Thisresistive heat is generated by a separatecurrent (typically an alternating current, AC)supplied to the filler metal that flows from thewire directly into the weld pool.

The current is properly adjusted so thatideally the temperature of the filler metal canreach its melting point as soon as it enters theweld pool. In comparison with the cold-wireGTAW, the hot-wire GTAW process is morecomplicated and has a higher cost with theadditional power supply, but it can provide ahigher deposition rate. Unfortunately, despitethe increased temperature of the filler metalwhen it enters into the weld pool, the wiremelting is still finished by the heat generatedfrom the weld pool during the hot-wire GTAWprocess. That is, part of the heat used to meltthe filler metal is essentially absorbed from the

weld pool. To melt the wire faster, the arc wouldhave to establish a larger weld pool. Increasingthe melting or deposition rate is at the expenseof an increased weld pool. The arc energy anddeposition rate are thus coupled. This couplingreduces the process controllability to providedesirable arc energy and deposition rate freelyto meet the requirements from differentapplications. For overhead welding where themaximal mass of the weld pool is restricted,this coupling also directly reduces the amountof the filler metal that can be added in eachpass. The productivity is directly reducedbecause of this coupling or undesirableprocess controllability.

TIG Welding ProcessThe necessary heat for Gas Tungsten ArcWelding (TIG) is produced by an electric arcis maintained between a non-consumabletungsten electrode and the part to be welded.The heat affected zone, the molten metal, andthe tungsten electrode are all shielded fromthe atmosphere by a blanket of insert gas fewthrough the GTAW torch. Insert gas is thatwhich is inactive, or deficient in active chemicalproperties, the shielding gas serves to blanketthe weld and exclude the active properties.

The shielding gas serves to blanket theweld and excluded the active properties inthe surrounding air. It does not burn, andadds nothing to or takes anything from themental. Inert gases such as argon andhelium do not chemically react or combinewith other gases. They possess no order andare transparent, permitting the weldermaximum visibility of the arc.

In some instances a small amount ofreactive gas such as hydrogen can be addedto enhance travel speeds.

244


The GTAW process can producetemperatures of up to 35,000 °F/19,426 °C.The torch contributes only heat to the workpiece.

If filler metal is required to make the weld, itmay add in the oxyacetylene welding process.There are also a number of filler metal feedingsystems available to accomplish the taskautomatically.

MATERIALS AND METHODSIn this analysis tungsten inert gas welding isused. It is a process which yields coalescenceof metals by heating with a welding arcbetween a continuous filler metal electrode andthe work piece. Firstly, specimens ofdimensions 60 mm × 50 mm× 6 mm areprepared, then closed butt joint are made bythese specimens. Before welding, edges ofthe work pieces are suitably prepared. Theedges and the area adjoining them is clearedof dust using wire brush. Afterwards, the workpieces to be welded were positioned withrespect to each other and welding processwas performed under constant voltage andcurrent in flat (down hand) position. But thewelding speed varies for each test.

During the welding process, following dataare chosen in Table 1.

defined as a steel alloy with a minimum of10.5% to 11% chromium content by mass.

Stainless steel does not readily corrode,rust or stain with water as ordinary steel does,but despite the name it is not fully stain-proof,most notably under low oxygen, high salinity,or poor circulation environments. It is alsocalled corrosion-resistant steel or CRES whenthe alloy type and grade are not detailed,particularly in the aviation industry. There aredifferent grades and surface finishes ofstainless steel to suit the environment the alloymust endure. Stainless steel is used whereboth the properties of steel and resistance tocorrosion are required.

Work Piece DetailsThe work piece used for this test is SS 304.

Dimension of the specimen: 60 mm x50 mm x 6 mm

Specimen showed in Figure 1.

Filler Rod Dia (mm) 1.6 2.3 3

Welding Current (A) Ampere 140 160 180

Table 1: Welding Parameters and Levels

LevelsUnitsFactors

–1 0 +1

Material Used – Stainless SteelIn metallurgy, stainless steel, also known asinox steel or inox from French “in oxydable”, is

Figure 1: Specimens Prepared for Welding

Response Surface MethodologyResponse surface methodology usesstat ist ical models, and thereforepractitioners need to be aware that even the

245


best statistical model is an approximationto reality. In practice, both the models andthe parameter values are unknown, andsubject to uncertainty on top of ignorance.Of course, an estimated optimum point neednot be optimum in reality, because of theerrors of the est imates and of theinadequacies of the model.

The application of RSM to designoptimization is aimed at reducing the cost ofexpensive analysis methods (e.g., finiteelement method or CFD analysis) and theirassociated numerical noise. The problem canbe approximated as described with smoothfunctions that improve the convergence of theoptimization process because they reduce theeffects of noise and they allow for the use ofderivative-based algorithms.

ANOVAThe ANOVA procedure performs analysis ofvariance (ANOVA) for balanced data from a

wide variety of experimental designs. Inanalysis of variance, a continuous responsevariable, known as a dependent variable, ismeasured under experimental conditionsidentified by classification variables, known asindependent variables. The variation in theresponse is assumed to be due to effects inthe classification, with random error accountingfor the remaining variation.

Analysis of variance is most important toolfor calculating responsible factors, whichsignificantly affects mechanical properties.For determining these affect on processparameters, F-test was performed. Resultsof ANOVA and percentage contributions byeach process parameters are tabulated inTables 2, 3 and 4.

Confirmation TestOptimal level of process parameters waspredicted using response graph and ANOVA.Process parameters and their levels which

Source Df Sum of Squares Mean of Squares F P % of Contribution

Filler Wire Diameter 2 0.5788 0.1166 0.18 0.839 1.67

Current 2 28.9951 14.49 22.32 0.001 83.41

Error 8 5.1954 0.64

Total 12 34.76

Table 2: Analysis of Variance for Hardness

Note: R-Sq = 85.06%.


Filler Wire Diameter 2 0.5788 0.1166 0.18 0.839 1.67

Filler wire diameter 2 0.457 0.222 108.94 0.00 94.8

Welding Current 2 0.00814 0.00407 2 0.198 1.68

Error 8 0.01632 0.002

Total 12 0.48197

Table 3: Analysis of Variance for Penetration

Note: R-Sq = 96.61%.

246


affect mechanical properties are pulse currentat level 2, background current at level 5, pulsefrequency at level 4, pulse duty cycle at level 2and percentage of Helium in Argon at level 3.The obtained results were verified for theimprovement in multiple quality characteristics


Filler wire diameter 2 7510.6 3626.2 11.35 0.005 69.6

Current 2 721.2 360.6 1.13 0.370 6.68

Error 8 2555.0 319.4

Total 12 10786.7

Table 4: Analysis of Variance for Tensile Strength

Note: R-Sq = 76.31%.

Filler Wire Diameter Welding Current Penetration (mm) Hardness (HRA) Tensile Strength N/mm²

3 140 1.09 56 434.9

3 140 1.1 55 433.3

3 140 1.08 57 435

Mean 1.09 56 434.4

Table 5: Confirmation Test

by conducting a confirmation test based onresults obtained in Table 5. The optimizationplot developed using MINITAB Software isshown in the Figure 2.

% of Experimental Error = ((Actual Value –Predicted Value)/Predicted Value)

Figure 2: Optimization Plot

Model Calculation% of Experimental Error (Penetration) =

((1.09 – 1.077)/1.077) × 100 = 1.2

% of Experimental Error (Hardness) = ((56– 55.28)/55.28) × 100 = 1.3

% of Experimental Error (Tensile strength)= ((434.4 – 430.9)/430.9) × 100 = 0.81

Welding Parameters

Welding SpeedSpeed of welding is defined as the rate oftravel of the electrode along the seam or therate of the travel of the work under the electrodealong the seam. Some general statements canbe made regarding speed of travel. Increasingthe speed of travel and maintaining constantarc voltage and current will reduce the width ofbead and also increase penetration until anoptimum speed is reached at whichpenetration will be maximum. Increasing the

247


speed beyond this optimum will result indecreasing penetration.

In the arc welding process increase inwelding speed causes:

• Decrease in the heat input per unit lengthof the weld.

• Decrease in the electrode burn off rate.

• Decrease in the weld reinforcement.

If the welding speed decreases beyond acertain point, the penetration also will decreasedue to the pressure of the large amount of weldpool beneath the electrode, which will cushionthe arc penetrating force.

Calculations: Speed of welding is defined asthe rate of travel of the electrode along theseam or the rate of travel of the work under theelectrode along the seam.

Speed of welding = Travel of electrode/Arctime mm/min.

Heat input rate or arc energy = V × I × 60/vjoules per mm

where, V is arc voltage in volts, I is weldingcurrent in ampere and v is speed of welding inmm/min.

Welding CurrentWelding current is the most influential variablein arc welding process which controls theelectrode burn off rate, the depth of fusion andgeometry of the weld elements.

Welding VoltageThis is the electrical potential differencebetween the tip of the welding wire and thesurface of the molten weld pool. It determinesthe shape of the fusion zone and weldreinforcement. High welding voltage produces

wider, flatter and less deeply penetrating weldsthan low welding voltages. Depth ofpenetration is maximum at optimum arcvoltage.

TIG Welding Machine Set-UpWhen setting up a TIG welder there are twomain settings. They are amperage and gasflow. Amperage settings vary depending on thetype and thickness of the metal to be welded.Gas flow rates also vary depending on draftconditions, cup size, and sometimes the

Figure 3: TIG Welding Set-Up

Figure 4: TIG Welding Machine

248


position of the weld. The gas flow rate couldrange from 5 CFH to 60 CFH for a large cupand drafty conditions. When choosing the gasto weld it is almost always assumed that youwill be using pure Argon. The TIG welding setup is shown Figure 3. The photograph of TIGwelding machine is shown in Figure 4.

Hardness TestHardness is defined as the resistance of amaterial to plastic deformation usually byindentation. It also refers to stiffness orresistance to scratching. Indentationhardness refers to number related to the areaor depth of the impression made by anindenter of fixed geometry under a knownstatic load.

There are many methods to determine thehardness, among those Brinell and Rockwellhardness tests are frequently used.

In our testing procedure, first test wasHardness test. For hardness test, we madedifferent test specimen. We cut a piece of 60mm from the length of the welded pieces. Likethat we made pieces of weldment for hardnesstest.

Before hardness test, we need to makesmooth the surface of piece which we got aftercutting.

We used 220, 320 400 and 600 gradedemery papers for getting the smoothnessrequired, after getting appropriatesmoothness, we have done hardness test forparent material by using which are showed inFigure 5 tested at welded zone.

Weld PenetrationIt is the depth to which the base metal and fillermaterial have melted and mixed during welding

Figure 5: Rockwell Hardness Machine

process. It depends on the weld processparameters used and can vary for differentplate conditions.

To measure the depth of penetration thefollowing processes were carried out on thespecimens: (1) sectioning, (2) grinding, (3)polishing, (4) etching, and (5) profile tracing.

Sectioning: The transverse sections of eachweld were cut using a bans saw. Care wastaken to avoid deformation of the sensitiveaustenitic grade material.

Grinding: Grinding was performed in orderto remove the cold work from cutting and wasdone at speeds of approximately 300 RPM.

Polishing: After grinding the specimens wererough polished by hand. In order to obtainbetter edge flatness, the specimens werepolished using silicon carbide abrasive papersof grades 100, 220, 400, 600 and 800,respectively. The specimens were thenpolished using an abrasive-slurry of alumina(Al2O3) and water (H2O) on a polishingmachine.

249


Etching: After polishing, the specimensunderwent etching. Etching was necessary forexamining the microstructure of the weldbead. The etchant used was Marble’sreagent which is a mixture of HCL (50 ml),CuSO4 and H2O (50 ml). The polished facesof each specimen were swabbed for about50-60 seconds with the etchant in order toreveal the weld bead.

Profile Tracing: The bead profiles of thespecimens were traced using a reflective typeoptical profile projector. The profile projectorused is shown in Figure 6. The traced beadprofiles were scanned in order to determinethe depth of penetration. The depth wasmeasured with the help of SOLID WORKSsoftware. The photograph of welded specimenis shown in the Figure 7.

Tension TestMechanical testing plays an important role inevaluating fundamental properties ofengineering materials as well as in developingnew materials and in controlling the quality ofmaterials for use in design and construction. Ifa material is to be used as part of anengineering structure that will be subjected toa load, it is important to know that the materialis strong enough and rigid enough to withstandthe loads that it will experience in service. Asa result engineers have developed a numberof experimental techniques for mechanicaltesting of engineering materials subjected totension, compression, bending or torsionloading.

The most common type of test used tomeasure the mechanical properties of amaterial is the Tension Test. Tension test iswidely used to provide basic designinformation on the strength of materials and isan acceptance test for the specification ofmaterials. The tension test machine used inthis project is shown in Figure 8.

RESULTS AND DISCUSSIONThis chapter deals with the results anddiscussions of the experimental findings ofwelded joints prepared at constant current,

Figure 6: Profile Projector

Figure 7: Welded Jointed

250


voltage, electrode size and welding technique(Down hand welding). The welded specimensprepared under varying current, filler wirediameter, and constant welding speed ishaving different effects. The experimentalresults were displayed below in the Table 6.

The Effect of Filler Diameter onPenetrationReadings of penetration is obtained from themathematical relation and variations in thepenetration are analysed with the help of graph

Figure 8: Tension Test Machine

S. No. Filler Rod Dia (mm) Welding Current (A) Penetration (mm) Hardness (HRA) Tensile Strength N/mm²

1. 1.6 140 1.40 57 326.3

2. 3.0 140 1.10 55 421.5

3. 1.6 180 1.40 52 401.2

4. 3.0 180 0.92 52 420.7

5. 1.6 160 1.50 52 384.4

6. 3.0 160 1.01 53 429.2

7. 2.3 140 0.98 56 375.4

8. 2.3 180 0.97 52 365.8

9. 2.3 160 0.99 52 363.3

Table 6: Experimental Results

which is plotted between filler wire diameterand penetration. Voltage (24 v) and speed (95mm/min) are taken constant and filler diameteris varied during the welding of specimens.Increasing the filler rod diameter andmaintaining constant arc voltage andpenetration. High penetration will improve themechanical properties of the welded joint. Lowheat input will decrease the defects of thewelded joints. So it can be concluded fromexperimental analysis that for the SS304specimen having dimension 60 mm × 50 mm× 6 mm, the weld penetration is decreasing tosome extent then later it is increasing whenthe filler wire diameter is chosen as 2.3 mm.

The influence on filler wire diameter onpenetration is shown as graph in Figure 9.

The Effect of Filler Diameter onHardnessReadings of hardness is obtained from theBrinell hardness testing machine andvariations in the hardness are analysed withthe help of graph which is plotted betweenwelding speed and hardness of the weldedjoint. Voltage (24 v) and different current aretaken and welding speed is constant duringthe welding of specimens. The hardnessdecreases with increase filler diameter upto2.3 mm which was optimum value to obtain

251


maximum hardness. So it can be concludedfrom experimental analysis that for the SS304specimen having dimension 60 mm × 50 mm× 6 mm, optimum weld ability can be achievedby considering the welding parameters as withdifferent current, filler wire diameter and voltage24 V. It is shown as graph in Figure 10.

in the tensile are analysed with the help ofgraph which is plotted between filler roddiameter and tensile strength of the weldedjoint. It is shown in Figure 11. Voltage (24 v)and different current are taken and weldingspeed is constant during the welding ofspecimens. The tensile strength slowlydecreases with increases filler wire diameterwhich was optimum value to obtain maximumtensile strength at filler wire diameter as 3 mm.

Figure 9: Effect of Filler Wire Diameteron Weld Penetration

Effects of Filler Diameter onTensile StrengthReadings of tensile strength is obtained fromthe universal testing machine and variations

Figure 10: Effect of Filler Wire Diameteron Hardness

Figure 11: Effect of Filler Wire Diameteron Tensile Strength

The Effects of Welding Current onPenetrationIncreasing the welding current and maintainingconstant arc voltage and so that arc energywill increased. High arc energy will reduce themechanical properties of the welded joint. Lowheat input will decrease the defects of thewelded joints. So it can be concluded fromexperimental analysis that for the SS304specimen having dimension 60 mm × 50 mm× 6 mm, optimum weld ability can be achievedby considering the welding parameters as withdifferent current, filler wire diameter and voltage24 V. It is denoted in Figure 12.

252


The Effect of Welding Current onHardnessReadings of hardness is obtained from theBrinell hardness testing machine. Voltage (24v) and different current are taken and weldingspeed is constant during the welding ofspecimens. The hardness decreases withincrease in welding current up 160 Amperescurrent and suddenly increases. The influenceof welding current on hardness is shown inFigure 13.

Effects of Welding Current onTensile StrengthReadings of tensile strength is obtained fromthe universal testing machine and variationsin the tensile are analysed with the help ofgraph which is plotted between welding currentand tensile strength of the welded joint. Thetensile strength increases with increaseswelding current which was optimum value toobtain maximum tensile strength. The graphshowing the effect of welding current on tensilestrength is shown as graph in Figure 14.

Figure 12: Effect of Welding Currenton Weld Penetration

Figure 13: Effect of Welding Currenton Hardness

Figure 14: Effect of Welding Currenton Tensile Strength

CONCLUSIONThe effects of welding parameters like currentand filler diameter on tensile strength ,hardness and penetration of SS 304 steel hasbeen studied. The optimum condition inwelding operation and effect of weldingparameters using Response surfacemethodology has been found out.

As per our experimental work the mostimportant factor affecting the tensile strengthand penetration is found to be filler wire

253


diameter whereas for hardness the majorfactor is welding current.

According to the order of importance, theparameters affecting the penetration, tensilestrength and hardness are: Welding currentand filler wire diameter.

The optimum process parameters for TIGwelding of SS 304 steels are found to be:

Wire Diameter = 3 mm

Welding Current = 140 Ampere

REFERENCES1. Ahmed Khalid Hussain, Abdul Lateef,

Mohd Javed and Pramesh T (2010),“Influence of Welding Speed on TensileStrength of Welded Joint in TIG WeldingProcess”, International Journal ofApplied Engg Research, Vol. 1, No. 3.

2. Atanda P, Fatudimu A and Oluwole O(2010), “Sensitization Study ofNormalized 316L Stainless Steel”, Journalof Minerals & Materials Characterization& Engineering, Vol. 9, No. 1, pp. 13-23.

3. Bandhita Plubin, NarongchaiSathavornvichit and PutipongBookkamana (2006), “Central CompositeDesign in Optimization of the Factors ofAutomatic Flux Cored Arc Welding”,Proceedings of the 2nd IMT-GT RegionalConference on Mathematics, Statisticsand Applications, pp. 13-15.

4. Farhad Kolahan and Mehdi Heidari (April2010), “Modeling and Optimization ofMAG Welding for Gas Pipelines UsingRegression Analysis and SimulatedAnnealing Algorithm”, Journal ofScientific & Industrial Research, Vol. 69,pp. 259-265.

5. Gadewar S P, Peravli Swaminadhan,Harkare M G and Gawande S H (2010),“Experimental Investigation of WeldCharacteristics for a Single Pass TIGWelding with SS304”, InternationalJournal of Engineering Science andTechnology, Vol. 2, No. 8, pp. 3676-3686.

6. Howard B Cary and Scott Helzer (2004),Modern Welding Technology, 6th Edition,AMAZON.

7. Jeffus (2013), Welding: Principles andApplications, AMAZON.

8. Kishore K, Gopal Krishna P V, VeladriK and Kiran Kumar G (2010), “Analysisof Defects in Gas Shielded ArcWelding of AA 6351 Using TaguchiMethods”, International Journal ofApplied Engineering Research, Vol. 5,pp. 393-399.

9. Kumanan S, Edwin Raja Dhas andGowthaman K (June 2007),“Determination of Submerged ArcWelding Process Parameters UsingTaguchi Method and Regression Analysis”,Indian Journal of Engineering & MaterialSciences, Vol. 14, pp. 177-183.

10. Lakshminarayanan A K andBalasubramanian V (2009), “Comparisonof RSM with ANN in Predicting TensileStrength of FSW of AA7039”,Transactions of Nonferrous MetalsSociety of China, Vol. 19, pp. 9-18.

11. Lenin N, Sivakumar M and VigneshkumarD (2010), “Process ParameterOptimization in ARC Welding ofDissimilar Metals”, Thammasat Int. J. Sc.Tech., Vol. 15, No. 3.

254


12. Scott Funderburk R (1999), “A Look atHeat Input: Key Concepts in WeldingEngineering”, Welding Innovation,Vol. XVI, No. 1.

13. Senthil Kumar T, Balasubramanian V,Sanavullah M Y and Babu S (2007), “Effectof Pulsed Current TIG WeldingParameters on Pitting Corrosion Behaviorof AA6061 Aluminum Alloy”, Journal ofMaterial Science Technology, Vol. 23,No. 2.

14. Sureshkumar L, Verma S M,Radhakrishna Prasad P, Kiran Kumar P

and Siva Shanker T (2011),“Experimental Investigation for WeldingAspects of AISI 304 & 316 by TaguchiTechnique for the Process of TIG & MIGWelding”, International Journal ofEngineering Trends and Technology,Vol. 2, No. 2.

15. Ugur Esme, Melih Bayramoglu, YugutKazancoglu and Sueda Ozgun (2009),“Optimization of Weld Bead Geometry inTIG Welding Process Using Grey RelationAnalysis and Taguchi Method”, Materialsand Technology, pp. 143-149.

EXPERIMENTAL INVESTIGATION OF MULTIPASS TIG WELDING ...

Documents