Torsion strength of TIG welded similar and dissimilar ...

Torsion strength of TIG welded similar and dissimilar metal jointsof ASME SA213 Gr.T11 and BS3059:1987 PT1 ERW320

SEWA SINGH1,* , VIKAS CHAWLA1 and GURBHINDER SINGH BRAR2

1Department of Mechanical Engineering, IKG-Punjab Technical University, Kapurthala, Punjab, India2Department of Mechanical Engineering, Anand College of Engineering and Management, Kapurthala, Punjab,

India

e-mail: [email protected]; [email protected]; [email protected]

MS received 14 March 2021; revised 9 September 2021; accepted 22 September 2021

Abstract. Tungsten Inert Gas (TIG) welding has been learnt to be the most widely used technique among the

fusion welding techniques. Welding of different components of boilers is preferably accomplished by TIG

welding, due to the process capabilities of the technique to produce sound joints, even in case of Dissimilar

Metal Joints (DMJ). DMJs owing to the techno-economic advantages, find vast area of application especially in

boiler fabrication industry. It has been learnt that the quality of welded joints is signified by mechanical

properties of the joint. The present paper is focused on the behavioural aspects of similar and dissimilar metal

joints of ASME SA 213 GR. T11 and BS 3059:1987 PT 1 ERW 320, the boiler steam tube materials, prepared

by TIG welding under torsional loading. It has been observed that the dissimilar metal joint has performed better

than the individual similar metal joints under torsional loading by achieving 85% of the torsion strength of one

of the parent metal, whereas in the case of individual similar metal joints of ASME SA213 Gr. T11 and

BS3059:1987 PT1 ERW320, 78% and 68% of the torsion strength of respective parent metals has been

observed. ANOVA statistics have validated the existence of significant impact of input process variables on the

output quality measure of the welded joint at 95% level of confidence. The predictive models for optimum

torsional strength have been proposed by regression analysis.

Keywords. Dissimilar metal joints; Torsion strength; TIG welding; ANOVA; Multiple regression.

1. Introduction

Fusion Welding (FW) has been considered as the suit-

able and oldest method of joining two metallic components

and the versatility of the technique, deems it fit for several

industrial applications wherein nearly all kind of metallic

joints are being fabricated under the one umbrella of FW.

Different techniques of FW have been in practice, and the

thoughtful selection of most suitable technique for a par-

ticular material combination renders sound joint [1], certain

metallurgical issues however are associated with the FW,

especially in case of dissimilar metal joints [2], reduction in

the mechanical strength has also been found, when com-

pared with the base metals [3].

Arc welding in manual mode, Gas Metal Arc Welding

(GMAW), Gas Tungsten Arc Welding (GTAW) or TIG are

frequently preferred welding methods [4]. It is quite obvi-

ous that being a fusion welding process, the performance of

the TIG welded joint be marginally inferior in comparison

to the parent metal as well as the joint prepared by solid

state welding, the reported results in the literature advocate

the statement wherein it has been mentioned that the

accumulated plastic strain is higher in the TIG welded joint

than that in friction stir welded joint of aluminium alloy

AW 1050 [5]. However, literature has suggested that being

capable of joining large number of metals, TIG welding is

considered as most versatile and popular method of

metallic joint fabrication [6, 7], it has also been learnt that

using TIG welding, effective and efficient production of

sound joints is feasible irrespective of the welding position

[8]. Fox et al have stated that deeper penetration in single

pass has been found feasible using TIG welding [9].

Numerous researchers have discussed TIG welded joints

from different perspectives in the literature published so

far. Here, few of them have been quoted, discussing the

weldment quality of TIG welded joints and the extent of

influence of the process variables on it. TIG welding has

been observed better than MIG welding in several aspects

[10], the effect of TIG welding parameters on the

mechanical properties and the corrosion resistance of

aluminium alloy AA 6061 T6 has been reported in the

literature [11], it has been found that the TIG welded

specimen experienced ductile fracture in tensile as well as

torsion loading. However, the fracture location has been*For correspondence

Sådhanå (2021) 46:231 � Indian Academy of Sciences

https://doi.org/10.1007/s12046-021-01750-w Sadhana(0123456789().,-volV)FT3](0123456789().,-volV)

noticed in heat affected region [12]. The torsion test of TIG

welded SS 304 and Ni-Ti joint revealed that on an average

the joint resisted a torque of 52 Nm before fracture, it has

been reported that the heat affected zone in the said TIG

welded joint extended to a distance of 125 lm [13].

The welding current has been reported to be an important

factor influencing the output weldment quality of the TIG

welded joint [14], size of filler wire, flow rate of gas and

welding speed have also been presented as influencing

factors affecting the weldment quality [15], speed of

welding and mechanical properties of the weldment have

been reported to be negatively associated [16]. The welding

current affects the heat input at the weld interface, which in

turn affects the microstructure at and around the joint

interface [17]. It has also been reported that TIG welded

joints of alloy 6061 T6 of aluminium exhibited the fatigue

strength in the range comparable to that produced by fric-

tion stir welding, thereby proving the capability of the

technique, in spite of being part of fusion welding family

[18].

Mechanical testing and characterization of TIG welded

dissimilar metal joints of Inconel 718 and high strength

steel has revealed that the heat input at the weld interface

has affected the microstructural growth in the heat affected

region, the weldment quality has been signified by the

mechanical properties of the joint [19]. The composition of

the filler wire has been reported to be an influential

parameter in dissimilar metal joints, playing decisive role in

the output quality of the weldment in terms of mechanical

properties [20]. It has been learnt that the use of some

interlayer improves the mechanical properties of the welded

joint, literature has supported the statement reporting the

use of Incoloy 600 during the activated TIG welding of

AISI 316 L steel with P91 steel, it has revealed the

improvement in microstructure and impact toughness of the

joint without compromise of joint strength [21], the role of

flux type in influencing the mechanical properties of A-TIG

welded dissimilar metal joint prepared from P92 and 304H

stainless steel has been explored and the results revealed

that the appropriate selection of the flux positively impact

the joint properties [22].

The contribution of the input process parameters towards

the variation in macrostructure, microstructure and

mechanical properties of the TIG welded joints has been

explored, while using the controlled intermittent wire feed

method, it has been reported that above 40% reduction in

joint strength observed attributed to lower heat input at

weld interface [23]. The effect of welding current and

welding time on mechanical and microstructural properties

during dissimilar arc stud welding of different steel com-

binations viz. AISI 316L, AISI 1020 and AISI 304 has been

reported in literature, it was observed that joint produced at

a welding current of 600 A for 0.25 seconds exhibited the

maximum torque strength of 77 Nm [24]. The significant

influence of the inert gas shielding on the microstructure

and weld quality has been advocated in literature [25].

The use of statistical techniques for the validation of

inferences of the experimental observations, and opti-

mization of the process parameters to obtain the best in

class results of the experiments has been in practice since

long ago. The published literature supports the statement

that the careful selection of the process parameters plays

prominently decisive role as far as the quality characteris-

tics of the welded joints are concerned, the attempts of

optimization of the process variables using statistical

techniques have been reported in the literature [26, 27, and

28].

It is apparent from the brief discussion of the above

literature that various researchers have attempted to

explore different aspects of the TIG welding technique in

preparing different similar and dissimilar metal joints of

numerous materials, very few have discussed the perfor-

mance of TIG welded joint of ASME SA213 Gr. T11 and

BS3059:1987 PT1 ERW320 as discussed by Singh et al[1], it has been observed that the said paper has focused

on only tensile strength of the TIG welded similar and

dissimilar metal joints of the materials under considera-

tion, further the performance of the same material com-

bination under torsional loading has not been reported so

far, therefore an attempt has been made in the present

work to explore and report the behaviour of the similar

and dissimilar metal joints of the same materials under

torsional loading, in order to narrow down the gap in

knowledge about the mechanical behaviour of the said

materials.

2. Materials and methods

The current study has been focused on investigation of

weldments of ASME SA213 Gr. T11 (referred as T11,

hereafter) and BS3059:1987 PT1 ERW320 (referred as

3059, hereafter), the materials of boiler steam tubes being

used in boiler fabrication industry, the outer diameter and

wall thickness of the tubes were 38 mm and 3.7 mm

respectively. As per current industrial practice, the weld-

ments of boiler steam tubes are prepared after pre-heating

the work pieces to 150� to 200�C. Hence it has been

decided to prepare the weldments under three pre-

conditions i.e. welding without preheating (represented as

welding at room temperature, RT), welding after

pre-heating at 150�C (referred as PH150), and welding after

preheating at 200�C (referred as PH200).

TIG welding equipment used for current work supported

manual welding therefore the process variables considered

for the current investigation were, welding current and flow

rate of inert gas. After the pilot investigations, the current

variation has been decided to lie between 90-120 A with

incremental value of 10 A. Similarly, the gas flow rate was

decided to vary from 8 l/min to 12 l/min, incrementing by

2 l/min. Hence the total number of weldments corresponding

231 Page 2 of 12 Sådhanå (2021) 46:231

to each pre-condition piled up to 12. The welding equip-

ment used was of GAIDU make, operating on single phase

power AC power supply of 220 V. Table 1 depicts the

specifications of the welding equipment and figure 1 depicts

the photographs of the one of the dissimilar metal joints

prepared at a welding current of 110 A and under the gas

flow rate of 10 l/min.

Table 2 presents the compositional details other than Fe

in both the parent metals, as revealed from the optical

spectroscopy conducted in the laboratory of Central Insti-

tute of Hand Tools, Jalandhar, Punjab (India) and table 3

presents the different combinations of the process variables

in different experimental conditions. The terms F1, F2 and

F3 represent the flow rate of 8 l/min, 10 l/min and 12 l/min

respectively; whereas C1-C4 represent the welding current

settings varying from 90 A to 120 A, incrementing by 10 A

and L1–L3 represent the pre-condition of the work piece

before welding, herein L1 corresponds to the weldments

prepared at room temperature i.e. weldments prepared

without preheating, L2 represent the weldments prepared

after preheating to 150�C, and L3 corresponds to the

weldments prepared after preheating to 200�C.

3. Experimental results and discussion

Torsion testing of the weldments was conducted on FIE

make torsion testing machine installed in the laboratory of

Anand College of Engineering and Management, Kapur-

thala (Punjab, India). Table 4 contains the specifications of

the machine. ASTM E2207-15 standard has been followed

for specimen preparation and testing of the weldments as

suggested in published literature [29] (figure 2), the welded

portion was kept in the centre of the specimen. Table 5

contains the torsion test results of the parent metals (un-

welded). The torque before fracture has been considered as

the indicative measure of the torsion strength of the spec-

imen, and twist angle indicates the extent of ductile nature

of the specimen. Apparently the specimens of BS3059:1987

PT1 ERW320 have reflected the existence of the more

torsional resistance and ductility than that exhibited by

ASME SA213 Gr. T11, it may be the effect of Cr present in

the later material.

Table 6 depicts the maximum torque and maximum twist

angle observed during the torsion testing of the samples of

similar metal joints prepared from ASME SA213 Gr. T11.

Apparently it can be stated that the torque and angle of

twist are in direct association with the welding current, in

other words the samples prepared at higher welding current

are observed to exhibit higher torque before failure, and the

same is true for angle of twist. It may be attributed to the

reason as stated in literature that higher welding current

leads to higher and quicker heat input at the weld interface

resulting in refined microstructure leading to improved

mechanical properties [18] and [30]. However, the flow rate

of gas has directly influenced the torque and twist angle up

to the value of 10 l/min, and thereafter either the output

parameters tended to stabilise or showed slight decline in

the torque and twist angle. The maximum torque of 21 Nm

and maximum angle of twist of 18� has been observed in

the samples welded without preheating, when the welding

current and flow rate of gas were 110 A and 10 l/min

respectively.

Similarly, the response of similar metal joints prepared

from BS3059:1987 PT1 ERW320, to the torsional loading,

has been presented in table 7, general observation reveals

that the torque and twist angle increased, as the samples

prepared at higher and higher current were encountered.

The variation of the torque and twist angle with the vari-

ation in flow rate of gas has been observed to follow nearly

the similar trends as stated above, the slight decrease in the

torque in the samples welded at increased flow rate of gas

may be attributed to the increased cooling rate at the weld

interface due to the enhanced convective heat transfer rate

owing to the increased flow volume of the inert gas.

The maximum torque of 21 Nm has been observed in the

samples welded at 120 A of current and 12 l/min of gas

flow rate, when welded without preheating. The same value

has also been noticed in the samples welded at 110 A of

current and 10 l/min of gas flow, after preheating to 150�C.The parametric combination resulted in maximum twist

angle of 18�.Table 8 illustrates the data pertaining to the torsional

testing of dissimilar metal joints of ASME SA213 Gr. T11

and BS3059:1987 PT1 ERW320. Herein, the trends

obtained are on similar track as discussed above, the

maximum torque of 23 Nm has been observed in the

samples welded at 110 A of current and 10 l/min of gas

flow rate, when the samples were preheated to 200�C, thecorresponding angle of twist was noticed to be 18�. It hasbeen noticed that the maximum angle of twist of 20�exhibited by the samples welded at 120 A of current and

12 l/min of gas flow rate, when the samples were preheated

to 200�C, but the sample failed at lesser torque i.e. 21 Nm,

it may be attributed to the microstructural refinement and

embrittlement caused by the high heat input at elevated

level of current. The trends of variation of torsion strength

as depicted in tables 6, 7, 8 are in good agreement with the

published literature showing increase in torsion strength

Table 1. Specifications of TIG welding equipment.

Sl.No. Description Specification

1. Make GAIDU

2. Manufactured 2014

3. Power supply 220 V, AC

4. Current range Single phase, 0-200 A

5. Torch cooling Air cooled

Sådhanå (2021) 46:231 Page 3 of 12 231

with increase in heat input at weld interface up to certain

limit and thereafter tends to either stabilise or showing

declining values, however the source of heat input therein

was frictional effect between the mating surface unlike the

welding current in present work [31].

In all the three cases discussed above, it has also been

noticed that the pre heat temperature influences the tor-

sional strength and angle of twist to considerable extent, the

interaction of heat input at the weld interface and the

temperature gradient between weld region and parent metal

may be supposed to be the reason behind the behavioural

trends observed; stating otherwise increase in pre-heat

temperature leads to lowering temperature difference

between the weld interface and the parent metal, thereby

slowing down the cooling to some extent and hence lim-

iting the embrittlement caused by quicker cooling, and

hence improvement in the mechanical properties has been

witnessed, the literature also suggests the microstructural

refinement due to the heat interaction at the weld interface

Figure 1. Photograph demonstrating welded joint between T11 and 3059.

Table 2. Chemical composition of parent metals (in %), other than Fe.

Material C Si Mn P S Cr Mo Ni V

3059 0.371 0.469 1.01 0.0367 0.043 – – – –

T11 0.12 0.646 0.511 0.007 0.0031 1.11 0.469 0.0527 0.0021

Table 3. Experimental design depicting process variable.

Sl.No. Pre-Condition

Flow rate of inert gas (l/min)

F1 F2 F3

1. L1 (RT) C1 C1 C1

2. C2 C2 C2

3. C3 C3 C3

4. C4 C4 C4

5. L2 (PH150) C1 C1 C1

6. C2 C2 C2

7. C3 C3 C3

8. C4 C4 C4

9. L3 (PH200) C1 C1 C1

12. C2 C2 C4

11. C3 C3 C3

12. C4 C4 C4

Table 4. Specifications of torsion testing machine.

Make Model Maximum torque capacity (Nm) Torsion speed (rpm) Clearance between grips (mm)

FIE TT-10 100 1.5 0–420

231 Page 4 of 12 Sådhanå (2021) 46:231

to be the reason behind the mechanical behaviour of the

weldment [32, 33, and 34]

Figures 3, 4 and 5 represent the graphical presentation of

the torsion strength of similar metal joints of ASME SA213

Gr. T11 and BS3059:1987 PT1 ERW320; and dissimilar

metal joints of both the metals respectively. The trends

observed in current investigation are similar to those

specified in literature in terms of the effect of heat input on

the mechanical properties of the welded joint. It has been

reported therein that increase in axial pressure and rota-

tional speed during friction welding resulted in higher heat

input at weld interface leading to the improvement trends in

mechanical properties of the joint bearing direct association

with the heat input [35], likewise the heat input at weld

interface in current study being directly influenced by the

welding current has resulted in the trends presented herein.

4. Statistical validation

The inferences made from unprocessed raw data obtained

from the experimental observation is not considered as

healthy practice, but statistically processed data can reliably

be leading to some authenticated conclusions. The state-

ment is supported by the published literature signifying the

use of numerous mathematical and statistical technique for

the optimization of the process parameters [36].

The results discussed in previous sections indicate

towards the existence of some kind of association between

input process variables and the output joint strength in

terms of torque exhibited before fracture; however the

associations indicated herein by the variance in the results

may be theoretically obvious but the possibility of inclusion

of the variance due to experimental circumstances cannot

be completely ruled out; so the situation of ambiguity can

Figure 2. Torsion test specimen.

Table 5. Torsion test results of parent metals.

Sl.No Strength indicator ASME SA213 Gr. T11 BS3059:1987 PT1 ERW320

1. Torque (Nm) 27 31

2. Angle of twist (Degree) 20 23

Table 6. Torsion test results of similar metal joints of T11.

Flow rate Current

L1 L2 L3

Torque (Nm) Angle of twist (Deg) Torque (Nm) Angle of twist (Deg) Torque (Nm) Angle of twist (Deg)

F1 C1 8 5 9 5 11 7

C2 12 7 12 8 12 9

C3 15 8 16 10 17 10

C4 17 12 17 12 17 12

F2 C1 11 7 11 8 12 9

C2 16 12 14 10 14 11

C3 21 18 16 11 18 12

C4 21 17 17 12 20 12

F3 C1 10 7 11 8 12 10

C2 15 12 15 12 15 14

C3 19 15 18 13 17 15

C4 20 18 18 12 18 15

Sådhanå (2021) 46:231 Page 5 of 12 231

be avoided by testing the variance statistically. As sug-

gested by literature [1, 26] and [27], Analysis of Variance

(ANOVA) has been used to test the variance followed by

regression analysis, in order to develop predictive model of

the input and output relationship. Following are the null

hypothesis tested using two-way ANOVA:

1. H01= No significant influence of welding current exists,

on the torsion strength of the joints

2. H02= No significant influence of the gas flow rate exists,

on the torsion strength of the joints.

3. H03= No significant influence of the temperature of work

piece before welding exists, on the torsion strength of the

joints.

4. H04= No significant interaction exists between the input

parameters.

4.1 Similar metal joints of ASME SA213 Gr. T11

The summary of important statistics of ANOVA test carried

on the test data corresponding to the similar metal joints of

ASME SA213 Gr. T11 has been depicted in Appendix A1.

It is apparent from the tabulation that Fcurrent= 59.005, when

p=0.0005\a=0.05, it implies that H01 cannot be accepted

i.e. there exists a significant association between the

welding current and mechanical property of the joint under

consideration. Fflow=5.933, when p\a compels the rejec-

tion of H02, in other words significant association between

flow rate and the torsion strength has been validated.

Ftemp=0.617, when p[a indicates that the temperature has

no significant impact on the torsion strength of the weld-

ment, hence H03 has been accepted. As far as the interaction

between the input parameters is concerned it is evident

Table 7. Torsion test results of similar metal joints of 3059.

Flow rate Current

L1 L2 L3


F1 C1 10 8 12 8 12 10

C2 14 10 15 12 17 14

C3 16 13 16 14 19 16

C4 19 13 21 15 20 16

F2 C1 13 9 13 10 12 10

C2 17 13 18 13 17 13

C3 18 14 21 18 19 15

C4 19 17 20 18 19 14

F3 C1 12 10 13 9 13 12

C2 17 12 18 12 17 13

C3 18 15 20 12 18 15

C4 21 18 20 13 19 15

Table 8. Torsion test results of dissimilar metal joints of T11 and 3059.

Flow rate Current

L1 L2 L3


F1 C1 12 14 15 13 16 14

C2 16 16 18 15 20 17

C3 19 17 19 16 21 18

C4 20 17 21 18 21 19

F2 C1 14 13 16 14 17 15

C2 18 15 20 16 19 14

C3 20 17 21 19 23 18

C4 21 19 22 19 22 16

F3 C1 14 12 17 13 17 15

C2 19 14 18 14 19 18

C3 21 17 21 17 22 18

C4 21 18 21 18 21 20

231 Page 6 of 12 Sådhanå (2021) 46:231

from the tabulated data in Appendix-A1 that there exists no

significant interaction between current and flow rate,

whereas significant association between current-tempera-

ture and flow rate-temperature has been indicated, so H04

can be rejected for the later combination of parameters.

The model summary of regression analysis has been

presented in Appendix B1, it is evident that ‘R’ theMultiple

Correlation Coefficient has been noticed to bear value

0.914. It implies excellent level of prediction of the torsion

strength by the regression analysis. ‘R2’ the Coefficient ofDetermination signifies the model effectiveness in

explaining the contribution of the input variables in the

variance of output quantity, herein R2=0.835 simply means

that the welding current, flow rate, and temperature

0

5

10

15

20

25

90 A 100A

110A

120A

90 A 100A

110A

120A

90 A 100A

110A

120A

RT PH150 PH2008 l/min 8 12 15 17 9 12 16 17 11 12 17 17

10 l/min 11 16 21 21 11 14 16 17 12 14 18 20

12 l/min 10 15 19 20 11 15 18 18 12 15 17 18

Max

imum

torq

ue(N

m)

SIMILAR METAL JOINTS OF ASME SA213 Gr. T11

Figure 3. Variation of maximum torque (Similar Metal Joints of T11).

0

5

10

15

20

25

90 A 100A

110A

120A

90 A 100A

110A

120A

90 A 100A

110A

120A

RT PH150 PH2008 l/min 10 14 16 19 12 15 16 21 12 17 19 20

10 l/min 13 17 18 19 13 18 21 20 12 17 19 19

12 l/min 12 17 18 21 13 18 20 20 13 17 18 19

Max

imum

torq

ue

(Nm

)

SIMILAR METAL JOINTS OF BS3059:1987 PT1 ERW320

Figure 4. Variation of maximum torque (similar metal joints of BS3059).

Sådhanå (2021) 46:231 Page 7 of 12 231

accounts for 83.5% of variation of torsion strength. The

standard error of estimation has been observed to be 1.47. It

reflects that the average distance of observed data from the

regression line is 1.47 Nm, it further supports the closeness

of the predictive model to the observed experimental

trends. The same is further strengthened by the ANOVA

statistics of the regression model (Appendix C1) , depicting

the fitness of the model to the experimental data, evidently

the f-statistics obtained here i.e. F(3, 32)= 54.023, when

p=0.0005\\ a=0.05, indicates that the regression model

under consideration holds good fit of the experimental

results.

The generalised predictive model proposed by the

regression analysis can be presented as below:

TST11 ¼ �18:114þ 0:269C þ 0:521F � 0:002T

4.2 Similar metal joints of BS3059:1987 PT1ERW320

The summary of important statistics of ANOVA test carried

on the test data corresponding to the similar metal joints of

BS3059:1987 PT1 ERW320 has been presented in

Appendix-A2. It is apparent from the tabulation that Fcur-

rent= 221.851, when p=0.0005\a=0.05, it implies that H01

is rejected i.e. there exists a significant association between

the welding current and mechanical property of the joint

under consideration. Fflow=2.679, when p\a compels the

rejection of H02, in other words significant association

between flow rate and the torsion strength has been

validated. Ftemp=1.767, when p[a indicates that the tem-

perature has no significant impact on the torsion strength of

the weldment, hence H03 has been accepted. As far as the

interaction between the input parameters is concerned, it is

evident from the Appendix-A2 that there exists no signifi-

cant interaction between current and flow rate, whereas

significant association between current-temperature and

flow rate-temperature has been indicated, so H04 can be

rejected for the later combination of parameters.

The model summary of regression analysis depicts the

fitness of regression model (Appendix-B2), the MultipleCorrelation Coefficient, R = 0.910, means excellent level of

prediction of the torsion strength by the regression model.

The Coefficient of Determination R2=0.827 signifies that

the regression model is effective in explaining the contri-

bution of the input variables in the variance of output

quantity to the tune of 82.7%. The standard error of esti-

mate of 1.35 leads to inference that on an average the

observed values are 1.35 Nm far from the predictive model

represented by the regression equation, which strongly

supports the inference driven by the value of R2=0.827 and

further the closeness of fitment of the observed experi-

mental results and the predictive regression model has been

confirmed by the ANOVA statistics of the regression

model, (Appendix C2) depicting the fitness of the model to

the experimental data, evidently the f-statistics obtained

here i.e. F(3, 32)= 51.060, when p=0.0005\\a=0.05, indi-cates that the regression model under consideration holds

good fit of the experimental results.

The generalised predictive model proposed by the

regression analysis can be presented as below:

0

5

10

15

20

25

90 A 100A

110A

120A

90 A 100A

110A

120A

90 A 100A

110A

120A

RT PH150 PH2008 l/min 12 16 19 20 15 18 19 21 16 20 21 21

10 l/min 14 18 20 21 16 20 21 22 17 19 23 22

12 l/min 14 19 21 21 17 18 21 21 17 19 22 21

Max

imum

torq

ue(N

m)

ASME SA213 Gr. T11 & BS3059:1987 PT1 ERW320 DISSIMILAR METAL JOINT

Figure 5. Variation of maximum torque (dissimilar metal joints of T11 and 3059).

231 Page 8 of 12 Sådhanå (2021) 46:231

TS3059 ¼ �12:518þ 0:243C þ 0:313F þ 0:005T

4.3 Dissimilar metal joints of ASME SA213 Gr.T11 and BS3059:1987 PT1 ERW320

Observing the summary of important statistics of ANOVA

test carried on the test data corresponding to the dissimilar

metal joints of ASME SA213 Gr. T11 and BS3059:1987

PT1 ERW320 (Appendix A3), it is evident that Fcurrent=

46.201, when p=0.0005\a=0.05, it is an indication of the

fact that there exists a significant association between the

welding current and mechanical property of the joint under

consideration, hence H01 has be rejected. Fflow=6.419, when

p\ a compels the rejection of H02, in other words signif-

icant association between flow rate and the torsion strength

has been validated. Ftemp=6.200, when p\a indicates that

the temperature has significant impact on the torsion

strength of the weldment, hence H03 needs not to be

accepted i.e. temperature of the work pieces before welding

has significant impact on the torsion strength for dissimilar

metal combination. As far as the interaction between the

input parameters is concerned, it is evident from Appendix

A3 that there exists no significant interaction between

current-flow rate and temperature-flow rate, whereas sig-

nificant association between current-temperature has been

indicated. So H04 can be rejected for the later combination

of parameters.

The model summary of regression analysis (Appendix

B3) reflects the authenticity of the predictive regression

model, the Multiple Correlation Coefficient, R = 0.908,

signifies the existence of excellent level of prediction of the

torsion strength by the regression model. The Coefficient ofDetermination R2=0.825 implies that the 82.5% of the

variance of torsion test results due to variance of input

process variables has been explained by the regression

model. The standard error of estimate appears to be 1.15, it

simply implies that the explanation of the variance pre-

sented by coefficient of determination of 0.825 has been

strongly supported by the calculated standard error of

estimate, in other words the average distance between the

predictive model presented by the regression line and the

observed experimental results is 1.15 Nm only, and hence

the high degree of fitness of the predictive model can safely

be claimed, which is again supported by the ANOVA

statistics of the regression model (Appendix C3), depicting

the quality of fit of the regression model to the experimental

data, apparently F(3, 32)=50.289, when p\\a, indicates theexcellent fit of the model to the test results.

The generalised predictive model for the dissimilar metal

joints of the metals under investigation has been presented

as follows:

TSDis ¼ �5:627þ 0:196C þ 0:271F þ 0:011T

5. Conclusions

The present investigation has successfully explored the

performance to TIG welded similar and dissimilar metal

joints of T11 and 3059 materials of boiler steam tubes

under the action of torsional loading. The torque experi-

enced by the welded sample before fracture has been taken

as an indicative measure of the torsion strength. The

observations have revealed that there exists significant

individual impact of input process variables on the output

mechanical property of the welded joints at 95% level of

confidence. The trends depicted here are in good agreement

to those depicted in literature [37, 38, and 39]. The joint

efficiency in case of dissimilar metal joints has been

observed to be 85% and 74% as compared to the parent

metals T11 and 3059, respectively, whereas in case of

similar metal joints the joint efficiency has been found to be

78% in case of T11 and 68% in case of 3059 materials,

respectively. However, the results depicted here are far

better than those reported in the literature [23].

The volume of a particular research work can never be so

vast that the researcher could claim the perfectness, there

are certain unavoidable constraints of time and finances that

limit the extent of work envelop, certainly such limitations

serve as the scope for future work by fellow researchers,

following are such scope for future work.

1. The investigation can be focused on detailed exploration

of more mechanical properties. Especially, the impact of

process variables on fracture behaviour can be explored.

2. Effect of filler metal composition can be explored.

Appendix

Appendix A1

See Table 9.

Table 9. ANOVA Statistics for similar metal joints of ASME SA

213 GR. T11

Source of variation SS df MS F P

Current 342.556 3 114.185 59.005 0.0005

Flow rate 39.389 2 19.694 5.933 0.046

Temperature 5.722 2 2.861 0.617 0.569

Current*flow rate 1.944 6 0.324 0.526 0.778

Current*temperature 11.611 6 1.935 3.143 0.043

Flow rate*temperature 13.278 4 3.319 5.391 0.010

Sådhanå (2021) 46:231 Page 9 of 12 231

Appendix A2

See Table 10.

Appendix A3

See Table 11.

Appendix B1

See Table 12.

Appendix B2

See Table 13.

Appendix B3

See Table 14.

Appendix C1

See Table 15.

Table 10. ANOVA Statistics for similar metal joints of

BS3059:1987 PT1 ERW320


Current 289.639 3 96.546 221.851 0.0005

Flow rate 12.500 2 6.250 2.679 0.0183

Temperature 7.167 2 3.583 1.767 0.317




Table 11. ANOVA Statistics for dissimilar metal joints of

ASME SA213 Gr. T11 and BS3059:1987 PT1 ERW320


Current 191.222 3 63.741 46.201 0.0005

Flow rate 11.056 2 5.528 6.419 0.046

Temperature 22.389 2 11.194 6.200 0.033




Table 12. Model summary of regression analysis for similar

metal joints of ASME SA 213 Gr. T11

Model R R2Adjusted R

square

Std. error of the

estimate

1 0.914a 0.835 0.820 1.47442

a Predictors: (Constant), Current (C), Flow Rate (F), Temperature (T)

Table 13. Model summary of regression analysis for similar

metal joints of BS3059:1987 PT1 ERW320


square

Std. error of the

estimate

1 0.910a 0.827 0.811 1.35251

a Predictors: (Constant), Current (C), Flow rate (F), Temperature (T)

Table 14. Model summary of regression analysis for dissimilar

metal joints of ASME SA 213 Gr. T11 and BS3059:1987 PT1

ERW320


square

Std. Error of the

estimate

1 0.908a 0.825 0.809 1.15486

a Predictors: (Constant), Current (C), Flow rate (F), Temperature (T)

Table 15. ANOVA Statistics of regression model for similar

metal joints of ASME SA 213 Gr. T11

Model Sum of squares df Mean square F P

1Regression 352.324 3 117.441 54.023 .0005b

Residual 69.565 32 2.174

Total 421.889 35

a Dependent Variable: Torsion Strength (TS)b Predictors: (Constant), Current (C), Flow rate (F), Temperature (T)

231 Page 10 of 12 Sådhanå (2021) 46:231

Appendix C2

See Table 16.

Appendix C3

See Table 17.

Acknowledgements

The author is grateful to I. K. Gujral Punjab Technical

University, Kapurthala, Punjab (India) for providing oppor-

tunity to work on the project and providing the able

professional support through my supervisors Dr. Vikas

Chawla and Dr. G S Brar.

References

[1] Singh S, Chawla V and Brar G S 2019 The performance of

TIG welded similar and dissimilar weldments under tensile

loading: An experimental investigation. Int. J. Mech. Prod.Eng. Res. Dev. 9(4): 1015–1026

[2] Arivazhagan N, Devendranath R K, Karthikeyan S,

Manikandan N and Surendra S 2012 Hot Corrosion Studies

on Gas Tungsten Arc Welded AISI 304 and AISI 4140

Dissimilar Joint. Trends Intell. Robot. Autom. Manuf. 330:436–441

[3] Wahab M A, Dewan M W and Gonzalez G 2015 Effects of

rotating-bending and torsional fatigue loads on Gas Tungsten

Arc (GTA) welded AISI 1018 low carbon steel joints. In:

Proceedings of ASME 2015 International ManufacturingScience and Engineering Conference, June 8–12, Charlotte,

North Carolina, USA

[4] Lah N A C, Ali A and Ismail N 2019 Characterization of

Fusion Welded Joint: A Review. Pertanika J. Sci. Technol.17(2): 201–212

[5] Kilikevicius S, Cesnavicius R and Krasauskas P 2017 Low-

cycle fatigue life of aw 1050 aluminium alloy FSW and TIG

welded joints. In: Proceeding of the 7th InternationalConference on Mechanics and Materials in Design, June11–15, Albufeira (Portugal)

[6] Shaikh I A and Rao M V 2015 A Review on Optimizing

Process Parameters for TIG Welding using Taguchi Method

& Grey Relational Analysis. Int. J. Sci. Res. 4(6): 2449–2452[7] Vyas A H and Patel R M 2017 A Review Paper on TIG

Welding Process Parameters. IJSRD 5(2): 1301–1304

[8] Anand K R and Mittal V 2018 Review on parametric

optimization of TIG welding. Int. Res. J. Engg. Technol.4(1):1266–1268

[9] Fox G, Hahnlen R and Dapino M J 2012 Fusion welding of

nickel–titanium and 304 stainless steel tubes: Part II:

Tungsten inert gas welding. J. Intell. Mater. Syst. Struct.24(8): 962–972

[10] Raj J, Agrawal N, Thakur M and Baghel A 2017 A review on

TIG/MIG welded joints. Int. J. Sci. Tech. Engg. 4: 65–71[11] Eltai E, Mahdi E and Alfantazi A 2013 The Effects of Gas

Tungsten Arc Welding on the Corrosion and Mechanical

Properties of AA 6061 T6. Int. J. Electrochem. Sci. 8:

7004–7015

[12] Eltai E and Mahdi E 2014 Tensile, Hardness and Torsion

Behavior of Welded AA. Appl. Mech. Mater. 575: 400–404[13] Fox G, Hahnlen R and Dapino M 2011 TIG Welding of

Nickel-Titanium to 304 Stainless Steel. In: Proceedings ofASME 2011 Conference on Smart Materials, AdaptiveStructures and Intelligent Systems, September 18–21, Scotts-

dale, Arizona, USA, 1: 625–632

[14] Soni H and Dwivedi A 2019 Study of Effect of TIG Welding

Process Parameters for Welding of Aluminium plates.

IAETSD J. Adv. Res. Appl. Sci 6(10): 230–237[15] Sathish R, Naveen B, Nijanthan P, Geethan K A V and Rao

V S 2012 Weldability and Process Parameter Optimization

of Dissimilar Pipe Joints using GTAW. Int. J. Eng. Res.Appl. 2(3): 2525–2530

[16] Hussain A K, Parmesh T, Javed M and Lateef A 2012

Influence of Welding Speed on Tensile Strength of Welded

Joint in TIG Welding Process. Int. J. Appl. Eng. Res.518–527

[17] Shi Y, Cui S, Zhu T, Gu S and Shen X 2018 Microstructure

and intergranular corrosion behavior of HAZ in DP-TIG

welded DSS joints. J. Mater. Process. Tech. 256: 254–261[18] Mishra A and Nidigonda G 2018 Comparative Mechanical

and Microstructure Properties Analysis of Friction Stir

Welded and TIG Welded AA6061-T6 Similar Joints. J.Adv. Res. Manuf. Mater. Sci. Metall. Eng. 5(1 & 2): 1–8

[19] Cheepu M, Anuradha M, Das V C and Venkateswarlu D

2019 Microstructure Characterization in Dissimilar TIG

Welds of Inconel Alloy 718 and High Strength Tensile Steel.

Mater. Sci. Forum 969: 496–501

[20] Karthi S, Babu S P K, Shanmugham S and Balaji V P 2020

Study on the Dissimilar Joining of Martensitic Stainless Steel

Table 16. ANOVA Statistics of regression model for similar

metal joints of BS3059:1987 PT1 ERW320


1Regression 280.213 3 93.404 51.060 .0005b

Residual 58.537 32 1.829

Total 338.750 35

a Dependent Variable: Torsion Strength (TS)b Predictors: (Constant), Current (C), Flow Rate (F), Temperature (T)

Table 17. ANOVA Statistics of regression model for dissimilar

metal joints of ASME SA213 Gr. T11 and BS3059:1987 PT1

ERW320


1Regression 201.211 3 67.070 50.289 .0005b

Residual 42.678 32 1.334

Total 243.889 35

a Dependent Variable: Torsion Strength (TS)b Predictors: (Constant), Current (C), Flow rate (F), Temperature (T)

Sådhanå (2021) 46:231 Page 11 of 12 231

and Carbon Steel Using TIG Welding. Adv. Additi. Manuf.Joining 545–554

[21] Kulkarni A, Dwivedi D K and Vasudevan M 2019 Dissimilar

metal welding of P91 steel-AISI 316L SS with Incoloy 800

and Inconel 600 interlayers by using activated TIG welding

process and its effect on the microstructure and mechanical

properties. J. Mater. Process. Technol. 274:116280[22] Sharma P and Dwivedi D K 2019 A-TIG welding of

dissimilar P92 steel and 304H austenitic stainless steel:

Mechanisms, microstructure and mechanical properties. J.Manuf. Process. 44: 166–178

[23] Baskoro A S, Amat M A, Pratama A I, Kiswanto G and

Winarto W 2019 Effects of tungsten inert gas (TIG) welding

parameters on macrostructure, microstructure, and mechan-

ical properties of AA6063-T5 using the controlled intermit-

tent wire feeding method. Int. J. Adv. Manuf. Technol. 105:2237–2251. https://doi.org/10.1007/s00170-019-04400-y

[24] Abass M H, Abood A N, Alali M, Hussein S K and Nawi S A

2021 Mechanical Properties and Microstructure Evolution in

Arc Stud Welding Joints of AISI 1020 with AISI 304.

Metallogr. Microsturct. Anal. https://doi.org/10.1007/

s13632-021-00744-8

[25] Kah P and Martikainen J 2013 Influence of shielding gases in

the welding of metals. Int. J. Adv. Manuf. Technol. 64:

1411–1421. https://doi.org/10.1007/s00170-012-4111-6

[26] Gopinath V, Kumar M T, Sirajudeen I, Yogeshwaran S and

Chandran V 2015 Optimization of Process Parameters In

TIG Welding of AISI 202 Stainless Steel Using Response

Surface Methodology. Int. J. Appl. Eng. Res. 10(13):

11053–11055

[27] Balaram A, Naik A and Reddy C 2018 Optimization of

tensile strength in TIG welding using the Taguchi method

and analysis of variance (ANOVA). Thermal Sci. Eng.Progress 8: 327–339

[28] Chaudhary V, Bodkhe V, Deokate S, Mali B and Mahale R

2019 Parametric Optimization of TIG Welding on SS 304

And MS using Taguchi Approach. Int. Res. J. Eng. Technol.6(5): 880–885

[29] Carrion P, Shamsaei N, Simsiriwong J and Imandoust A

2018 Effects of Layer Orientation on the Multiaxial Fatigue

Behavior of Additively Manufactured Ti-6Al-4V. In: Pro-ceedings of the 29th Annual International SFF Symposium-

An Additive Manufacturing Conference, November 2018,

Austin, TX. United States

[30] Deng C, Li X, Gong B, Liu X and Li Y 2019 Effects of

Hydrogen and Microstructure on Tensile Properties and

Failure Mechanism of 304L K-TIG Welded Joint. Mater.Sci. Eng. A 735: 208–217

[31] Handa A 2012 Studies of Mechanical Properties and HotCorrosion Behaviour of Friction Welded Dissimilar Steels.Ph.D. Thesis submitted to IKG Punjab Technical University,

Kapurthala, Punjab (India)[32] Gong B, Li X, Deng C and Li Y 2019 Effect of Pre-Strain on

Microstructure and Hydrogen Embrittlement of K-TIG

Welded Austenitic Stainless Steel. Corros. Sci. 149: 1–17[33] Raj J, Agrawal N, Thakur M and Baghel A 2017 A Review

on TIG/MIG Welded Joints. Int. J. Eng. Sci. Technol. 4:65–71

[34] Dey A, Singh A K and Rai R N 2017 Techniques to Improve

Weld Penetration in TIG Welding (A Review). Materialsto-day Proc. 4(2):1252–1259

[35] Handa A and Chawla V 2016 Experimental Evaluation of

Mechanical Properties of Friction Welded Dissimilar Steels

under Varying Axial Pressures. J. Mech. Eng. 66(1): 27–36[36] Balestrassi P P, De Oliveira L G, De Paiva A P, Ferreira J R,

De Costa S C and Campos P H D S 2019 Response surface

methodology for advanced manufacturing technology opti-

mization: theoretical fundamentals, practical guidelines, and

survey literature review. Int. J. Adv. Manuf. Technol. 104:1785–1837. https://doi.org/10.1007/s00170-019-03809-9

[37] Wu A, Zhang D, Wang G, Zhao Y, Li Q, Liu X, Meng D,

Song J and Zhang Z 2019 Study on The Inconsistency in

Mechanical Properties of 2219 Aluminium Alloy TIG-

Welded Joints. J. Alloys Compd. 777: 1044–1053[38] Huang B, Zhang J, Cui K, Mao X and Zheng M 2020

Influence of Heat Input on The Microstructure and Mechan-

ical Properties of CLAM Steel Multilayer Butt-Welded

Joints. Fusion Eng. Des. 152. https://doi.org/10.1016/j.

fusengdes.2019.111413

[39] Dewangan S, Mohapatra S K and Sharma A 2020 An

Assessment in to Mechanical Properties and Microstructural

Behavior of TIG Welded Ti-6Al-4V Titanium Alloy. GreySyst. Theory Appl. https://doi.org/10.1108/GS-11-2019-

0052

231 Page 12 of 12 Sådhanå (2021) 46:231

Torsion strength of TIG welded similar and dissimilar ...

Documents