Metall. Res. Technol.
Volume 117, Number 4, 2020
|Number of page(s)||9|
|Published online||07 July 2020|
Tensile and wear behaviour of friction stir welded AA5052 and AA6101-T6 aluminium alloys: effect of welding parameters
Research Scholar, Department of Mechanical Engineering, St. Joseph’s College of Engineering,
2 Department of Mechanical Engineering, St. Joseph’s College of Engineering, Chennai, India
3 Department of Mechanical Engineering, K.L.N. College of Engineering, Madurai, India
* e-mail: email@example.com
Accepted: 11 June 2020
To improve the performance and effectiveness of cost, constructing lightweight structure is the important factor for automobile, naval and aerospace industries. AA5052 and AA6101-T6 aluminium alloys are widely applied in transport industries, due to their lightweight and high strength and hence, joining of these two are unavoidable. Friction stir welding is an unconventional welding method, which is developed for constructing lightweight structures. This work describes the detailed study of friction stir welded dissimilar AA5052 and AA6101-T6 alloys. AA5052 and AA6101-T6 plates are welded with rotation rates of 765–1400 rpm and offset distances at advancing side of 0–2 mm. For this purpose, four levels of welding parameters based on Taguchi L16 orthogonal array are chosen. To determine the optimum combinational levels and identify the effect of above-mentioned parameters on tensile and wear properties, Signal to Noise ratio and ANOVA respectively are used. From the results, it is observed that the combination of 1 mm offset distance at advancing side and 1400 rpm rotating speed produces better tensile and wear properties, which is due to high heat generation, sufficient flow of materials and balanced precipitation and strain hardening effects. On the other hand, the combination of 2 mm tool offset at advancing side and 765 rpm rotational rate exhibits poor properties, which is associated with low heat input, defects formation, precipitate coarsening and lesser strain hardening effects.
Key words: FSW / Taguchi / rotating speed / offset at advancing side / material flow
© EDP Sciences, 2020
Friction Stir Welding (FSW) is an environmental friendly welding process, which is used to join aluminium and its alloys, which are difficult to weld, by conventional fusion welding techniques. This process also avoids the solidification related problems such as hot cracking, embrittlement and porosity. It does not require any filler material and shielding gas for welding [1,2]. In FSW, specially designed, hard and non-consumable rotating tool is plunged into the edges of the plates to be joined and the rotating tool is travelled along the joint line. The combination of rotation and translation of the tool generates frictional heat, deforms the base material plastically and promotes the material flow across the joint . It refines the grains at weld nugget. Hence, FSW produces better mechanical properties, when compared to fusion welds [4,5]. In which quality of weld is decided by welding variables such as tool design, rotational speed, translation speed, tool tilt angle etc. [6,7].
Rodriguez et al.  examined the joining properties of 6061 to 7050 aluminium alloys and found that intermixing of material and joint strength were improved with increasing rotating speed. Jariyaboon et al.  determined that grain size in the weld nugget of friction stir welded AA2024-T351 was much influenced by rotational speed. Sundaram et al.  stated that lower rotational speed causes poor plastic deformation of materials, while joining 2024-T6 and 5083-H321 alloys. Shazly et al.  concluded that low traversing speed causes more voids on the bottom surface of weldments, which affects the quality of weld. On the other hand, high traversing speed produces defect free joints on AA5083 aluminium alloys. Sato et al.  found that increment in rotational speed and decrement in traversing speed result the higher process temperature as well as grain growth in the weld nugget of AA1050 alloy. Kumar et al.  and Heidarzadeh et al.  found that lower axial force produces defects in the weld nugget and at the same time higher axial force exhibits defect free weld while joining aluminium alloys. Nejad et al.  examined the effects of threaded and unthreaded tools on AA2024-T4 joints. They found that unthreaded tool exhibits uniform structure and hardness. Elangovan et al.  reported that while using polygon profiled tools, pulsating stirring action as well as tensile properties can be increased with increment in number of polygon sides in AA6061 alloy joints. Ashok kumar et al.  also reported that hexagonal pin profiled tool improves the flow behavior of materials during joining, which gives better quality weld on AA6061-T6 and A356 joints, when compared to square and triangular pin profiled tools.
In 1990, a Japanese engineer Genichi Taguchi introduced the Taguchi technique. This technique is a simple and powerful technique, which can optimize the performance characteristics within the combination of process parameters. In Taguchi method, specially developed tables called as orthogonal arrays are used to design the experiments, which diminish the number of experiments. This reduces time and cost . Ashok kumar et al.  adopted Taguchi technique with L9 orthogonal array to determine the effects of welding parameters on AA6101-T6 and AA1350 aluminium alloys. They used ANOVA and found that rotational speed of the tool is the highest influencing variable on tensile strength, which is associated with heat generation and transportation of materials.
Most of the previous investigations on the friction stir welding of aluminium alloys were focused on optimizing the basic welding parameters such as traversing speed, rotating speed, axial load and tool geometry with respect to properties and defect formations of joints. But, no systematic investigation has been carried out so far to study the effects of offset distances on dynamic recrystallization, material flow, and defect formation and no attempt has been made to investigate the feasibility of joining stronger non-heat treatable AA5052 alloy with heat treatable AA6101-T6 alloy, which is difficult by fusion welding techniques. Hence, this work is carried out to analyze the weldability of AA5052 and AA6101-T6 alloys with varying offset distance at advancing side of 0–2 mm and tool rotational speed of 765–1400 rpm. To evaluate the quality of weld, tensile strength, wear rate and hardness values were measured and microstructural studies were carried out. For analyzing the impacts of each variable on properties of weld, Taguchi L16 orthogonal array with Signal to Noise ratio and ANOVA was employed.
AA5052 and AA6101-T6 aluminium plates with 100 mm length x 50 mm width × 6 mm thickness were used for fabricating weldments. Table 1 shows the chemical compositions of AA5052 and AA6101-T6 alloys which are given in Refs. [19,20] and Table 2 shows its mechanical properties. During joining, different levels of tool rotation rates (765 rpm, 980 rpm, 1190 rpm and 1400 rpm) and different positions of tool (0 mm, 0.5 mm, 1 mm and 2 mm) at advancing side with respect to the weld line were used (Tab. 3). Tensile, hardness, wear and microstructural tests were performed on each fabricated joints. Tensile strengths were evaluated using five ton computerized universal testing machine. Tensile specimens were cut as per ASTM E8M-04 standard using Wire EDM. Microstructural analyses were carried out with the help of Dewinter metallurgical microscope. Hardness values were measured using Wolpert microvickers hardness tester. Rotary drum abrasion resistance tester was used for measuring the weight loss. In which, wear rate was measured in terms of weight loss.
Chemical compositions (wt.%).
Levels of welding parameters.
Two important tools in Taguchi technique are orthogonal array and S/N ratio. In this full factorial design, tool rotational rate and offset distances at advancing side were considered as controlled variables meanwhile average tensile strength and weight loss were considered as output factors. For all 16 trials, S/N ratios were calculated (Tab. 4). The maximum S/N ratio value for tensile strength and weight loss are observed at rotation rate of 1400 rpm and 1 mm offset distance at advancing side.
Three stages in S/N ratio are larger is best, nominal is best and smaller is best. Since maximum tensile strength and minimum weight loss are desirable, larger the best and smaller the best approaches were used respectively. The expression for larger the best approach is: (1) The expression for smaller the best approach is: (2)
where, n is the number of replications and yi is observed response value. S/N response table for tensile strength and weight loss are presented in Tables 5 and 6. These tables show that offset distance at advancing side is the most influencing parameter on tensile strength and weight loss. While joining AA6101-T6 and AA1350 alloys, rotational rate is the more influencing parameter on tensile strength and weight loss, when compared to tool traversing speed and tool tilt angle [18,21] and tool offset is the more influencing factor, when compared to number of passes .
The main purpose of analysis of variance (ANOVA) is to investigate the most significant welding parameter on output factors. This analysis is carried out with 95% confidence level. In addition to that, ANOVA is also used to identify the percentage of influence of each parameter on output factors. The percentage contribution (PCR) is calculated by the following expression : (3) (4) where, ηi is the mean S/N ratio for the ith experiment, ηn is total mean S/N ratio and ni is number of experiments.
Tables 7 and 8 show the results of ANOVA. All the R2 values are above 0.9. The higher values of R2 indicate that the designed models are adequate and significant. Lower P values of term B (< 0.05) reveal that offset distance at advancing side is more significant factor on output factors. For both output factors, F values are A < B, which denotes that the term B is the more significant factor, when compared to the term A. From these tables, offset distance at advancing side influences more with 52.86% and 57.36% on tensile strength and weight loss respectively, whereas rotational rate influences lesser with 43.73% and 40.09% on tensile strength and weight loss respectively. At the same time, rotational rate influences more with 64.45% and 59.31% on tensile strength and weight loss respectively on AA6101-T6 and AA1350 joints [18,21]. According to Ref. , tool offset influences more with 96.72% and 99.45% on tensile strength and weight loss respectively, when compared to number of passes.
Taguchi L16 orthogonal array.
Response table for S/N ratios (larger is best) for tensile strength.
Response table for S/N ratios (smaller is best) for weight loss.
ANOVA for tensile strength.
ANOVA for weight loss.
Tensile properties of the weldments are linearly related with rotational rate of tool i.e. increment in rotational speed improves the tensile properties of the joints (Fig. 1). Rotational speed of the tool decides the amount of heat generation and rate of mixing . When the rotational rate is very low, heat generated is very much insufficient and the tool gets stuck between the base metals and it is impossible to conduct the joining . On the other hand, at the very higher rotational rate, heat generated is very much higher and larger defects appeared due to the turbulence in material flow . Hence, rotation rate of the tool is chosen from 765 rpm to 1400 rpm. When the rotational rate is at 1400 rpm, frictional heat generation is high and sufficient, which causes sufficient plasticization. This improves the flow properties of the materials (Fig. 2b) and interdiffusion during stirring. Thus, enhanced frictional heat and strong stirring action produces sufficient reaction, strong bonding and no defects at weld nugget (Fig. 3a). Kwon et al.  found that grain recrystallization increases with increase in process temperature. Thus, the higher degree of recrystallization produces finer grains at weld nugget (Fig. 2a). Hence, superior tensile strengths are observed at the rotational rate of 1400 rpm (Tab. 4). When the rotational speeds are at 1190 rpm and 980 rpm, amount of heat generation is insufficient which causes insufficient plasticization and lack of stirring. This produces poor flow properties of materials (Figs. 4b and 5b), insufficient reaction and poor bonding at weld nugget (Figs. 4a and 5a), which lowers the tensile strength of joints (Tab. 4). But, when the rotational rate is at 765 rpm, frictional heat generation and rate of stirring are very low. This causes poor plasticization and flow of materials (Fig. 6b). As a result of which, lack of intermixing of materials and continuous cracks are obtained and the lower degree of recrystallization produces larger grains at weld nugget (Fig. 6a). Hence, lowest tensile strengths are observed at the rotational rate of 765 rpm (Tab. 4).
Previous investigation revealed that tool probe offset at advancing side gives better mechanical properties. While it is located at retreating side, heat input to harder material at advancing side is not sufficient to plasticize it and hence, tool scratches the larger pieces of harder material, which are difficult to mix. Meanwhile, extensive plastic deformation of softer material at retreating side results in excessive flow materials. Hence, very poor bonding is observed . At the same time, tool probe offset at advancing side is larger, retreating side is not connected with weld nugget during welding . Based on this, various levels of tool probe offset at advancing side are chosen (Fig. 7). In which, tensile properties are improved upto 1 mm probe offset after that poor tensile properties are observed (Fig. 1). When the welding tool is plunged into joint line (zero offset), contact area of tool shoulder with advancing and retreating sides are equal and 50% of the tool probe is in AA5052 plate and remaining 50% is in AA6101-T6 plate (Fig. 7a). This generates same amount of frictional heat on both aluminium plates. But, AA5052 alloy is the hardest of both. Hence, plastic deformation of AA5052 and AA6101-T6 alloys are different i.e. plastic deformation of AA5052 alloy is lowest. This insufficient plastic deformation makes hard to flow of material from advancing side to retreating side and softer AA6101-T6 alloy is extruded out during stirring which results in poor interaction. This produces volume defects (Fig. 8a and 8b) and weak bonding at weld nugget, which causes poor tensile strengths, when the probe offset is zero (Tab. 4) and all the fractures are observed at weld nugget during tensile testing (Fig. 8c).
When the probe offset is 0.5 mm and 1 mm at advancing side, 60 % and 70 % of the probe is in AA5052 plate respectively (Fig. 7b and 7c). In both the cases, tool shoulder as well as probe contact area with advancing side are larger than retreating side. This increases frictional heat generation and distribution to AA5052 alloy. The different levels of heat input to AA5052 and AA6101-T6 alloys produce almost uniform plastic deformation on both sides especially in probe offset of 1 mm. This improves the flow properties of materials during stirring, which results in sufficient reaction and strong metallurgical bonding at weld nugget (Fig. 3a). Hence, high tensile strengths are observed especially in probe offset of 1 mm (Tab. 4) and the fractures are observed at Heat Affected Zone (HAZ) of retreating side during tensile testing (Fig. 3b). This is due to annealing effect , which coarsens the grains (Fig. 3c) and softens the Heat Affected Zone (HAZ). Thus, the combination of 1 mm tool offset at advancing side and 1400 rpm rotational rate produce better tensile properties.
While the probe is moving with 2 mm offset at advancing side, almost 90% of the probe is in AA5052 plate and tool shoulder is shifted 2 mm towards advancing side (Fig. 7d). Therefore, more amount of frictional heat is generated and distributed to AA5052 alloy, which results in very low heat input to AA6101-T6 alloy. This affects breaking of material at retreating side and transportation of material at weld nugget. Therefore, insufficient mixture, tunnel defects and discontinuities are produced at weld nugget (Fig. 9a). Hence, the propagation of discontinuities result fractures at weld nugget of the joints during tensile testing (Fig. 9b) and lower tensile strengths are observed, when the probe offset is 2 mm (Tab. 4). Hence, the combination of 2 mm tool offset at advancing side and 765 rpm rotational rate result in poor mechanical properties.
Main effects plot for tensile strength.
Microstructural images of Exp. No. 15.
Images of weldment of Exp. No. 15.
Microstructural images of Exp. No. 10.
Microstructural images of Exp. No. 6.
Microstructural images of Exp. No. 4.
Schematic illustration of (a) zero offset, (b) 0.5 mm offset at advancing side, (c) 1 mm offset at advancing side and (d) 2 mm offset at advancing side.
Images of weldment of Exp. No. 13.
Images of weldment of Exp. No. 16.
AA6101 is a heat treatable aluminium alloy and AA5052 is a non-heat treatable aluminium alloy. Hence, hardness of nugget zone relies on the combinational effects of precipitation hardening and strain hardening. In precipitation hardening process, hardness values are based on coarsening, dissolution and reprecipitation of precipitates. Previous researches suggested that whenever 6xxx series alloys involved, Mg2Si precipitates are formed [26,27]. Coarsening, dissolution and reappearance of Mg2Si precipitates depend upon the heat input during welding. In strain hardening process, severity of plastic deformation decides the hardness profile . Wear resistance of weld nugget is directly proportional to hardness of weld nugget i.e. higher hardness results lower weight loss [28,29]. Higher wear resistance is observed at 1 mm offset in advancing side and rotational rate of 1400 rpm (Fig. 10). This is due to rotational rate of 1400 rpm generates maximum frictional heat and offset of 1 mm provides sufficient heat to AA6101 alloy. This heat input enhances the amount of reappearance of dissolved Mg2Si precipitates. Probe offset of 1 mm in advancing side with 1400 rpm produces enhanced strain hardening of AA5052 alloy. Thus, the balanced effect of precipitation and strain hardening increase the hardness at nugget zone (Fig. 11), which resists the weight loss during wear test. Thus, higher wear resistance is observed.
Lowest wear resistance is observed at 2 mm offset in advancing side and rotational rate of 765 rpm (Fig. 10). This is due to rotational rate of 765 rpm generates minimum frictional heat and offset of 2 mm provides very low heat input to AA6101 alloy. This heat input is just enough to coarsen the Mg2Si precipitates in AA6101 (Fig. 6a). Lower rotational rate also reduces the strain hardening of AA5052 alloy. As a result of this, lowest hardness as well as highest weight loss at nugget zone is observed. While the tool is travelling with zero and 0.5 mm offset at advancing side with 1400 rpm and 1190 rpm, heat input to AA6101 alloy is high which causes over-ageing [30,31] and hardness is fully based on strain hardening of AA5052 alloy. Hence, reductions in wear resistance are observed. At the same time, 765 rpm and 980 rpm of rotational rate on zero and 0.5 mm offset are just enough to dissolute the precipitates on AA6101 side and reduce the strain hardening effect on AA5052 side. As a result, lower wear resistances are observed. On the other hand, rotational rate of 1400 rpm and 1190 rpm at 2 mm offset provide high heat to AA5052 side, which increases the cooling time. This increases the grain growth rate, which produces coarsened grains (Fig. 6a). At the same time, heat input to AA6101 alloy is low which is just enough to coarsen the grains. As a result of these effects lower hardness and lower wear resistance are observed. For the same offset, rotational rate of 765 rpm and 980 rpm produce strain hardening effect on AA5052 side and very low heat input to AA6101 side coarsens the precipitates, which minimizes the hardness (Fig. 11) as well as wear resistance.
Main effects plot for wear properties.
Hardness profile across various zones.
Friction stir welding of dissimilar AA5052 and AA6101-T6 aluminium alloys were studied. The following points were concluded, based on the obtained results:
the Taguchi L16 orthogonal designed experiments of friction stir welding of dissimilar AA5052 and AA6101-T6 aluminium plates were conducted successfully by varying rotation rates and offset distances at advancing side;
most influencing welding parameter was determined using signal to noise ratio and ANOVA. It was found that the tool offset distance at advancing side played a vital role with a contribution of 52.86% and 57.36% on tensile strength and weight loss respectively;
the maximum tensile strength was observed at the rotational rate of 1400 rpm and tool offset distance at advancing side of 1 mm. This was due to both the plates had experienced high heat input, sufficient plasticization and dynamic recrystallization;
the minimum tensile strength was observed at the rotational rate of 765 rpm and tool offset distance at advancing side of 2 mm. This was due to both the plates had experienced lowest heat input, poor flow properties and dynamic recrystallization;
the balanced effect of precipitation and strain hardening exhibited maximum hardness at nugget zone, which improved the wear resistance;
in future, this work can be extended with changing the amount of frictional heat generation using various dimensions of tool shoulder diameter.
- B.T. Gibson, D.H. Lammlein, T.J. Prater, W.R. Longhurst, C.D. Cox, M.C. Ballun, K.J. Dharmaraj, G.E. Cook, A.M. Strauss, J Manuf. Process. 16(1), 56–73 (2014) [Google Scholar]
- D.C. Douglas, K.S. Vecchio, Mater. Sci. Eng. A 465(1–2), 165–175 (2010) [Google Scholar]
- R. Ashok kumar, M.R. Thansekhar, Adv. Mater. Res. 984–985, 586–591 (2014) [Google Scholar]
- M.W. Mahoney, C.G. Rhodes, J.G. Flintoff, R.A. Spurling, W.H. Bingel, Metall. Mater. Trans. A 29(7), 1955–1964 (1998) [Google Scholar]
- P.B. Berbon, W.H. Bingel, R.S. Mishra, C.C. Bampton, M.W. Mahoney, Scr. Mater. 44(1), 61–66 (2001) [Google Scholar]
- T.S. Mahmoud, A.M. Gaafer, T.A. Khalifa, Mater. Sci. Technol. 24(5), 553–559 (2008) [Google Scholar]
- R. Nandan, T. Deb Roy, H.K.D.H. Bhadeshia, Prog. Mater. Sci. 53(6), 980–1023 (2008) [Google Scholar]
- R.I. Rodriguez, J.B. Jordon, P.G. Allison, T. Rushing, L. Garcia, Mater. Des. 83, 60–65 (2015) [Google Scholar]
- M. Jariyaboon, A.J. Davenport, R. Ambat, B.J. Connolly, S.W. Williams, D.A. Price, Corros. Sci. 49(2), 877–909 (2007) [Google Scholar]
- N. Shanmuga Sundaram, N. Murugan, Mater. Des. 31, 4184–4193 (2010) [Google Scholar]
- M. Shazly, S. Sorour, A.R. Alian, J. Mech. Sci. Technol. 30(1), 267–278 (2016) [Google Scholar]
- Y.S. Sato, M. Urata, H. Kokawa, K. Ikeda, Mater. Sci. Eng. A 354(1–2), 298–305 (2003) [Google Scholar]
- K. Kumar, S.V. Kailas, Mater. Des. 29, 791–797 (2008) [Google Scholar]
- A. Heidarzadeh, H. Khodaverdizadeh, A. Mahmoudi, E. Nazari, Mater. Des. 37, 166–173 (2012) [Google Scholar]
- S. Gholam Nejad, M. Yektapour, A. Akbarifard, J. Mech. Sci. Technol. 31(11), 5435–5445 (2017) [Google Scholar]
- K. Elangovan, V. Balasubramanian, S. Babu, Mater. Des. 30, 188–193 (2009) [Google Scholar]
- G. Taguchi, Introduction to quality engineering, McGraw-Hill, New York, 1990 [Google Scholar]
- R. Ashok kumar, M.R. Thansekhar, Mater. Sci. 23(1), 78–83 (2017) [Google Scholar]
- R. Ashok kumar, M.R. Thansekhar, J. Mech. Sci. Technol. 32(7), 3299–3307 (2018) [Google Scholar]
- N.T. Kumbhar, K. Bhanumurthy, J. Metall. 2012, 1–7 (2012) [Google Scholar]
- M. Ilangovan, S. Rajendra Boopathy, V. Balasubramanian, Def. Technol. 11, 174–184 (2015) [Google Scholar]
- H. Oktem, T. Erzurumlu, I. Uzman, Mater. Des. 28, 1271–1278 (2007) [Google Scholar]
- M. Jayaraman, R. Sivasubramanian, V. Balasubramanian, A.K. Lakshminarayanan, J. Mater. Sci. Technol. 25, 655–664 (2009) [Google Scholar]
- K. Ramanjaneyulu, R.G. Madhusudhan, G. Hina, Def. Technol. 11, 209–219 (2015) [Google Scholar]
- Y.J. Kwon, N. Saito, I. Shigematsu, J. Mater. Sci. Lett. 21(19), 1473–1476 (2002) [Google Scholar]
- U. Donatus, G.E. Thompson, X. Zhou, J. Wang, K. Beamish, Mater. Des. 83, 203–213 (2015) [Google Scholar]
- M. Cabibbo, H.J. McQueen, E. Evangelista, S. Spigarelli, M. Di Paola, A. Falchero, Mater. Sci. and Eng. A 460–461, 86–94 (2007) [Google Scholar]
- R. Ashok kumar, M.R. Thansekhar, Metallofiz. Noveishie Tekhnol. 41(2), 203–211 (2019) [Google Scholar]
- R. Ashok kumar, G.R. Raghav, K.J. Nagarajan, S. Rengarajan, P. Suganthi, V. Vignesh, Metall. Res. Technol. 116(6), 631 (2019) [EDP Sciences] [Google Scholar]
- F.O. Valerio, C. Kennedy, L.E. Murr, D. Brown, S. Pappu, B.M. Nowak, J.C. McClure, Scr. Mater. 38(5), 703–708 (1998) [Google Scholar]
- A. Shafiei-Zarghani, S.F. Kashani-Bozorg, A. Zarei-Hanzaki, Mater. Sci. Eng. A 500(1–2), 84–91 (2009) [Google Scholar]
Cite this article as: G. Kasirajan, Sathish Rengarajan, R. Ashok kumar, G.R. Raghav, V.S. Rao, K.J. Nagarajan, Tensile and wear behaviour of friction stir welded AA5052 and AA6101-T6 aluminium alloys: effect of welding parameters, Metall. Res. Technol. 117, 405 (2020)
Schematic illustration of (a) zero offset, (b) 0.5 mm offset at advancing side, (c) 1 mm offset at advancing side and (d) 2 mm offset at advancing side.
|In the text|
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.