Available online at www.amse.org.

cn
Acta Metall. Sin.(Engl. Lett.)Vol.23 No.4 pp312-320 August 2010

Eect of process parameters of pulsed current tungsten inert gas welding on weld pool geometry of titanium welds
M. Balasubramanian V. Jayabalan
Department of Manufacturing Engineering, Anna University, Guindy, Chennai 600 025, Tamilnadu, India

Department of Mechanical Engineering, Velammal Engineering College, Surapet, Anna University, Guindy, Chennai 600 025, Tamilnadu, India

V. Balasubramanian
Centre for Materials Joining & Research, Department of Manufacturing Engineering, Annamalai University, Annamalai Nagar-608 002, Tamilnadu, India Manuscript received 8 July 2009; in revised form 12 April 2010

The eects of pulsing current parameters on weld pool geometry namely front height, back height, front width and back width of pulse current gas tungsten arc welded (GTAW) titanium alloy was analysed. Four factors, ve levels, central composite design were used to develop empirical relationships, incorporating pulsed current parameters and weld pool geometry. KEY WORDS Gas tungsten arc welding; Pulsed current; Titanium alloy; Design of experiments

1 Introduction Welding thin sheets is quite dierent from welding thick sections, because during welding of thin sheets many problems are experienced. These problems are usually linked with heat input. Fusion welding generally involves joining of metals by application of heat for melting of metals to be joined. Almost all the conventional arc welding processes oer high heat input, which in turn leads to various problems such as burn-through or melt-through, distortion, porosity, buckling, warping & twisting of welded sheets, grain coarsening, evaporation of useful elements present in coating of the sheets, joint gap variation during welding, fume generation from the coated sheets etc. Use of the proper welding process, procedure and technique is one tool to address this issue[15] . GTAW is a good process for joining thin sheets but it suers with low welding speed and high equipment cost. Pulsed current gas tungsten arc (PCGTA) welding, developed in the 1950s, is a variation of GTAW process that involves cycling the welding current at a selected regular frequency. The maximum current is selected to give adequate penetration and bead contour, while the minimum is set at a level sucient to maintain a stable arc[6,7] . This

Corresponding author. Professor; Tel: +91 044 23751817; Fax: +91 044 24819579

E-mail address: [email protected] (M. Balasubramanian)

 313 permits arc energy to be used eciently to fuse a spot of controlled dimensions in a short time producing the weld as a series of overlapping nuggets. By contrast, in constant current welding, the heat required to melt the base material is supplied only during the peak current pulses allowing the heat to dissipate into the base material leading to a narrower heat aected zone (HAZ)[8] . Advantages include improved bead contours, greater tolerance to heat sink variations, lower heat input requirements, reduced residual stresses and distortion, renement of fusion zone microstructure, and reduced width of HAZ. There are four independent parameters that inuence the process are peak current, background current, pulse frequency, and pulse on time[7,8] . Experimental results have shown that the front height, front width, back height and back width of the weld pool in the GTA welding of stainless steel are greatly improved by using this approach[9,10] . Earlier AC was used to weld aluminium and magnesium alloy only. But recent study revealed that, pulsed AC TIG can be used for welding of titanium alloy to greater eect. Grain renement in titanium alloy was stronger in AC PCTIG than DC welding. However, no thorough study has been reported so far on pulsed current GTA welding of titanium alloys and hence the present investigation was carried out. 2 Experimental Set-up 2.1 Materials and methods The titanium alloy (Ti-6Al-4V) sheet of 1.6 mm thick with yield strength (at 0.2% oset) of 910 MPa, ultimate tensile strength of 998 MPa, and elongation of 10% was single pass autogenously welded. High purity argon gas (99.99%) was used as a shielding gas and as trailing gas right after welding to prevent absorption of oxygen and nitrogen from the atmosphere. Since the sheet was very thin, no distortion was found and hence backing plate was not necessary. There was also no need of using purging gas because of very thin sheet. From the literature four important factors of pulsed current GTA welding as presented in Table 1 were chosen. The joints were established with 2 mm tungsten electrode under the welding conditions presented in Table 2. A large number of trial runs were carried out using 1.6 mm thick mill annealed sheets of titanium (Ti-6Al-4V) alloy to nd out the feasible working limits of pulsed current GTAW process parameters. Due to wide range of factors, it was decided to use four factors, ve levels, rotatable central composite design matrix to perform a number of experiments for investigation. 2.2 Measurement of weld bead geometry Three metallographic weld bead samples were cut from each joint, with the rst sample being located at 15 mm behind the trailing edge of the crater at the end of the weld. The transverse face of the samples were surface-ground using 120 grit size belt with the help of a belt grinder, polished using grade 1/0 (245 mesh size), grade 2/0 (425 mesh size), and
Table 1 Important factors and their levels Series. 1 2 3 4 Factor Peak current Base current Pulse frequency Pulse-on-Time Notation p b f t Unit A A Hz s -2 60 20 0 35% -1 70 30 3 40% Levels 0 1 80 90 40 50 6 9 45% 50% 2 100 60 12 55%

 314 grade 3/0 (515 mesh size) sandpaper. The specimens were further polished by using aluminum oxide initially and then by utilizing diamond paste and velvet cloth in a polishing machine. The polished specimens after cleaning with alcohol were macro-etched by using kroll s solution to reveal the geometry of the weld bead (Fig.2). Several critical parameters, such as bead height and bead width (Fig.1) were measured. The bead geometry was measured with toolmakers microscope. The front height varied between 0.035 and 0.210 mm, back height varied between 0.019 and 0.106 mm, front width varied between 5.100 and 8.286 mm and back width varied between 3.967 and 7.478 mm. The experimental design matrix and the mathematical modeling of the corresponding results were already discussed elsewhere[10] .
Table 2 Welding conditions Power source Polarity Arc voltage Electrode Electrode diameter Shielding gas Gas ow rate Torch position Operation Welding speed Lincoln, USA AC 22 V W+2% Thoriated (alloy) 2 mm Argon 10 L/min Vertical Automatic 300 mm/min

Fig.1 Typical weld pool geometry.

Fig.2 Macrographs of weld pool.

3 Empirical Relationships The response function representing any of the weld pool dimensions is expressed as Y =f (p, b, f, t)[11,12]

 315 The second order polynomial (regression) equation used to represent the response Y is given by
k k

Y = b0 +
i=1

bi xi +
i=j
i=1

bii x2 i

+
i,j =1

bij xi xj

(1)

and for four factors, the selected polynomial could be expressed as

+b12 IP IB + b13 IP F + b14 IP t + b23 IB F + b24 IB t + b34 F t

(2)

where b0 is the average of responses and b1 , b2 , , b34 are the co-ecients that depend on respective main and interaction eects of the parameters. The value of the co-ecients has been calculated using the expressions[12] . All the co-ecients were tested for their signicance at 90% condence level applying student s t-test using the SPSS statistical software package. After determining the signicant co-ecients, the nal relationships were developed[10] . Front height (Fh ) Fh = {0.050288 0.0145p + 0.008583b 0.01108f + 0.006917t + 0.00655p2 + 0.002802b2 +0.0288f 2 + 0.0096t2 + 0.00175f t 0.00425pb 0.0066pt + 0.0018bf + 0.0035bt}mm (3) Back height (Bh ) Bh = {0.02543 0.00575p + 0.0045b 0.00342f + 0.00466t +0.00805p2 + 0.00568b2 + 0.0193f 2 + 0.008059t2 + 0.001125bf }mm Front width (Fw ) Fw = {5.3793 0.3678p + 0.1765b 0.01875f + 0.2255t + 0.3472p2 + 0.0873b2 +0.7135f 2 + 0.348t2 0.1475f t + 0.1611pb + 0.1213f t}mm Back width (Bw ) Bw = {4.5535 0.3678p + 0.1765b + 0.225t + 0.3472p2 + 0.71349f 2 +0.34799t2 + 0.16112pb + 0.294pt + 0.1158bf }mm 4 Results and Discussion The empirical relationships developed above can be employed to predict the geometry of weld bead and shape relationships for the range of parameters used in the investigation by substituting their respective values in coded form. Based on these models, the main and the interaction eects of the process parameters on the bead geometry were computed and plotted as depicted in Figs.3a3f. The results show the general trends between cause and eect. (6) (5) (4)

 316

0.10
(a)

0.10
Front height Back height

6.8
(b)

6.2
Front width Back width

0.09
Front height / mm

0.09 0.08 0.07 0.06 0.05 0.04 0.03 0.02 3.0

6.6 6.4
Front width / mm

6.0 5.8 5.6

0.07 0.06 0.05 0.04 0.03 0.02 -2.0 -1.5 -1.0 -0.5

Back height / mm

0.08

6.2 6.0 5.8 5.6

5.4 5.2 5.0 4.8 4.6

5.4 5.2 -1.0

4.4 4.2 3.0

0.0

0.5

1.0

1.5

2.0

2.5

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

Base current / A

Base current / A

0.20
(c) Front height Back height
Front height / mm

0.20

12
(d)

12
Front width Back width

11 0.15 10
Front width / mm

11 10 9 8 7 6 5 4 3.0

Back height / mm

0.15

9 8 7 6 5

0.10

0.10

0.05

0.05

0.00 -2.0 -1.5 -1.0 -0.5 0.0

0.5

1.0

1.5

2.0

2.5

0.00 3.0

4 -2.0 -1.5 -1.0 -0.5 0.0

0.5

1.0

1.5

2.0

2.5

Pulse frequency / Hz

Pulse frequency / Hz

0.12
(e) Front height

8.0
(f) Front width Back width

7.5 7.0 6.5 6.0

0.10
Front height / mm

Back height
Back hieght / mm
Front width / mm

7.5

7.0

0.08

6.5 5.5 6.0 5.0 4.5 4.0 3.0

0.06

0.04

5.5

0.02 -2.0 -1.5 -1.0 -0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

5.0 -2.0 -1.5 -1.0 -0.5 0.0

0.5

1.0

1.5

2.0

2.5

Peak current / A

Peak current / A

Fig.3 Eect of pulsing current parameters on weld pool geometry: (a) eect of base current on height; (b) eect of base current on width; (c) eect of pulse frequency on height; (d) eect of pulse frequency on width; (e) eect of peak current on height; (f) eect of peak current on width.

4.1 Direct eects of process variables on the bead geometry In contrast to constant current welding, in pulsed current gas tungsten arc welding, heat energy is supplied only during peak current pulses, allowing it to dissipate into the

Back width / mm

 317 base metal during the background current and thus lowering heat build up in the adjacent base material, thus leading to a narrower heat aected zone. From Fig.3a and 3b it is apparent that the back height decreases up to 35 A of base current and increases thereafter. The front height behaved constant up to 25 A and then increased thereafter. With respect to bead width, there is a slight dip in the front width up to 30 A and then increases as the base current increases. Back width gradually increases from the lower level (20 A) to higher level (60 A) of base current. From Fig.3c and 3d it is apparent that the back height decreases up to 6 Hz of pulse frequency and increases thereafter. Same trend was observed in the case of front height too. With respect to bead width, there is a sharp dip in the front width up to 6 Hz and then increases as the pulse frequency increases. Similar trend was observed in the back width also. This may be due to dierence of heat input caused by variation in pulse frequency. From 0 to 6 Hz of pulse frequency the interval between pulses is high and hence the heat input which enters the system at a moment increases and bead width and height is high, but when the frequency increases the heat input decreases and hence the width and height decreases (Fig.4). Similar eect was observed in the case of pulse-on-time also. The bead height and width displayed two dierent trends. It followed a decreasing trend up to 45% of pulsing time and then thereafter it was found to be increasing as seen in Fig.5.
0.25
(a) Peak current 60 A Peak current 70 A

0.35
(b)

0.30 0.25 0.20 0.15 0.10 0.05

Peak current 60 A Peak current 70 A Peak current 80 A Peak current 90 A Peak current 100 A

0.20
Back height / mm

Peak current 80 A

0.15

Peak current 100 A

0.10

0.05

Front height / mm

Peak current 90 A

0.00 -2.0 -1.5 -1.0 -0.5 0.0

0.5

1.0

1.5

2.0

2.5

3.0

0.00 -2.0 -1.5 -1.0 -0.5 0.0

0.5

1.0

1.5

2.0

2.5

3.0

Pulse frequency / Hz

Pulse frequency / Hz

16 15 14
Back width / mm

14
(c) Peak current 60 A Peak current 70 A Peak current 80 A Peak current 90 A Peak current 100 A

(d)

13 12
Front width / mm

Peak current 60 A Peak current 70 A Peak current 80 A Peak current 90 A Peak current 100 A

13 12 11 10 9 8 7 6 5 4 3 -2.0 -1.5 -1.0 -0.5

11 10 9 8 7 6 5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

4 -2.0 -1.5 -1.0 -0.5 0.0

0.5

1.0

1.5

2.0

2.5

3.0

Pulse frequency / Hz

Pulse frequency / Hz

Fig.4 Interaction eect of pulsing current parameters on weld pool geometry: (a) pulse frequency and peak current vs back height; (b) pulse frequency and peak current vs front height; (c) pulse frequency and peak current vs back width; (d) pulse frequency and peak current vs front width.

 318

0.15 0.14 0.13 0.12

8
(a) Front height Back height

8
(b) Back width Front width

0.14 0.12 0.10 0.08 0.06 0.04 0.02 0.00 2.0

7
Front height / mm

7
Back height / mm

0.11 0.10 0.09 0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0.00 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5

3 -2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

3 2.0

Pulse-on-time / s

Pulse-on-time / s

10
(c) Peak current 60 A Peak current 70 A Peak current 80 A
Back width / mm Front width / mm

12
(d)

11 10 9 8 7 6

Peak current 60 A Peak current 70 A Peak current 80 A Peak current 90 A Peak current 100 A

Peak current 90 A Peak current 100 A

5 5 4 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 4 -2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

Pulse-on-time / s

Pulse-on-time / s

0.10
(e) Peak current 60 A Peak current 70 A Peak current 80 A
Front height / mm

0.20 0.18 0.16 0.14 0.12 0.10 0.08 0.06 0.04 0.02
(f) Peak current 60 A Peak current 70 A Peak current 80 A Peak current 90 A Peak current 100 A

0.08
Back height / mm

Peak current 90 A

0.06

Peak current 100 A

0.04

0.02

0.00 -2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

0.00 -2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

Pulse-on-time / s

Pulse-on-time / s

Fig.5 Eect of pulsing current parameters on weld pool geometry: (a) pulse-on-time vs front height; (b) pulse-on-time vs front width; (c) peak current and pulse-on-time vs back width; (d) peak current and pulse-on-time vs front width; (e) peak current and pulse-on-time vs back height; (f) peak current and pulse-on-time vs front height.

 319 4.2 Interaction eects of process variables on the bead geometry From Fig.4a to Fig.4d it is apparent that the bead height and bead width decreases up to 6 Hz of pulse frequency and increases thereafter. This may be due to dierence of heat input caused by variation in pulse frequency. From 0 to 6 Hz of pulse frequency the interval between pulses is high and hence the heat input which enters the system at a moment increases and bead width and height is high, but when the frequency increases the heat input decreases and hence the width and height decreases. After certain level of pulse frequency say 6 Hz in this case, the pulse frequency is increasing and seems to be like a continuous current which causes more heat input and hence the width and height increases. The same trend is observed in the case of peak current also. Initially as the peak current is less the bead width and height is more and as the peak current increases, the height and width reduced up to say 70 A and then it increased for dierent values of pulse frequency. From Fig.5 it is evident that there exist an interaction eect between the peak current and pulse-on-time as the peak current increases from 60 A. A pulsing frequency of 6 Hz is found to produce optimum results. At low frequencies, the eect of succeeding pulses on a solidifying bead is only minimal. On the other hand, at very high frequencies the amplitude of the vibrations induced in the weld pool and of the temperature oscillations are considerably reduced. Thus there exists an optimum frequency at which the greatest eect is produced. In the current investigation, the minimum bead width and height has been achieved at the middle level of the process parameters. 5 Conclusions (1) A ve level factorial technique was employed for developing empirical relationships for predicting important weld bead dimensions of pulsed current GTA welded titanium alloy. (2) The empirical relationships developed can be employed easily in automated or robotic welding in the form of a program, for obtaining the desired weld bead dimensions. (3) Out of the four process variables considered, pulse frequency had a signicant positive eect on most of the important bead parameters.

AcknowledgementsThe authors sincerely acknowledge the help and facilities extended to them by the Department of Manufacturing Engineering, Annamalai University, Annamalai Nagar, TamilNadu, India. The authors are grateful to Mr.K. Anbazhagan, Chennai for making necessary arrangements to procure the base metal for investigation. The authors are also grateful to the management of Velammal Educational Trust for rendering their full support. The authors wish to thank Mr.S. Babu, DRDO Project Associate, Annamalai University for rendering helping hand to carry out the statistical analysis.

REFERENCES [1] T. Ueyama, H. Tong, S. Harada, R. Passmore and M. Ushhio, Weld J 84(2) (2005) 40. [2] G. Mathers, Welding of Aluminium and Its Alloys (England: Wood Head Publishing Limited, 2002) p.4.

 320
[3] P. Praveen and P.K.D.V. Yarlagadda, J Mater Process Technol 164-165 (2005) 1106. [4] M. Brace and J. Brook, Weld J 81(3) (2002) 23. [5] G. Madhusudhan Reddy, Proceedings of ISTE Summer School on Recent Developments in Materials Joining (Annamalai University, India, 2001). [6] K. Prasad Rao, Proc of the National Conference on Recent Trends in Materials Processing (RAMP, Annamalai University, 2001) p.176. [7] G. Madusudhan Reddy, A.A. Gokhale and K. Prasad Rao, Mater Sci Technol 14 (1993) 61. [8] M. Balasubramanian, V. Jayabalan and V. Balasubramanian, J Mater Sci Technol 22(6) (2006) 821. [9] S.C. Juang and Y.S. Tarang, J Mater Process Technol 122 (2002) 33. [10] M. Balasubramanian, V. Jayabalan and V. Balasubramanian, J Mater Eng Performance 18(7) (2009) 871. [11] Y.S. Tarng, H.L. Tsai and S.S. Yeh, Int J Mach Tools Manuf 39(9) (1999) 1427. [12] D.C. Montgomery, Design and Analysis of Experiments, 3rd Ed. (NewYork: John Wiley and Sons, 1991).