Journal Information
Vol. 5. Issue 3.
Pages 275-281 (July - September 2016)
Share
Share
Download PDF
More article options
Visits
4456
Vol. 5. Issue 3.
Pages 275-281 (July - September 2016)
Original Article
DOI: 10.1016/j.jmrt.2016.02.002
Open Access
Non-contact sheet forming using lasers applied to a high strength aluminum alloy
Visits
4456
Rafael Humberto Mota Siqueiraa, Sheila Medeiros Carvalhob, Ivan Kwei Liu Kamc, Rudimar Rivaa,d, Milton Sergio Fernandes Limaa,d,
Corresponding author
msflima@gmail.com

Corresponding author.
a Space Sciences and Technology Program, Instituto Tecnológico de Aeronáutica, São José dos Campos, SP, Brazil
b Space Propulsion Division, Instituto de Aeronáutica e Espaço, São José dos Campos, SP, Brazil
c Instituto de Ciência e Tecnologia, Universidade Federal de São Paulo, São José dos Campos, SP, Brazil
d Photonics Division, Instituto de Estudos Avançados, São José dos Campos, SP, Brazil
This item has received
4456
Visits

Under a Creative Commons license
Article information
Abstract
Full Text
Bibliography
Download PDF
Statistics
Figures (11)
Show moreShow less
Tables (2)
Table 1. Thermal and mechanical properties of the aluminum sheet.
Table 2. Process parameters.
Show moreShow less
Abstract

Laser beam forming (LBF) is a contactless mechanical process accomplished by the introduction of thermal stresses on the surface of a material using a laser in order to induce plastic deformation. In this work, LBF was performed on 1.6mm thick sheets of a high strength aluminum alloy, AA6013-T4 class by using a defocused continuous Yb-fiber laser beam of 0.6mm in diameter on the sheet top surface. The laser power and process speed were varied from 200W to 2000W and from 3 to 30mm/s, respectively. For these experimental conditions, the bending angle of the sheet ranged from 0.1° to 2.5° per run. In the highest bending angle condition, 1000W and 30mm/s, the depth of remelted pool was 0.6mm and the microstructure near the plate bottom surface remained unaltered. For the whole set of experimental conditions, the hardness remained constant at approximately 100 HV, which is similar to the base material. In order to verify the applicability of the method, some previously T-welded sheets were straightened. The method was efficient in correcting the distortion of the sheets with a bending angle up to 5°.

Keywords:
Lasers
Aluminum alloys
Sheet forming
Laser beam welding
Full Text
1Introduction

Due to the high level of automation and the process speed, laser beam welding (LBW) has been used by the general industry. Another advantage of LBW is the low heat input, which results in a small heat-affected zone and reduced distortion of the welded assemblies. Typically, the distortion of the metal sheet in keyhole welding is much smaller than in arc welding, even with high thickness sheets as those observed in the shipbuilding industry [1].

It has been observed that conduction welds carried out using low laser beam intensities produces an effect of permanent deformation of the sheets. The phenomenon involved is usually called temperature gradient mechanism (TGM) [2], which is shown schematically in Fig. 1[3]. Moving a defocused laser beam over the metal sheet surface generates a steep thermal gradient between the surface exposed to the laser and the lower surface. At the beginning of the process, when the plate is being heated, the central part of the sheet near the laser beam moving line is “raised”, leading to a bending in the direction opposite to the surface normal. This folding is limited by the appearance of the liquid, which does not offer more tensile force on the sheet. When the beam has passed through the solidification point on the sheet surface, two phenomena occur: (a) cooling contraction and (b) compressive shrinkage forces. The end result is a V-shaped bend with a thin line mark on the surface of the sheets [4].

Fig. 1.

The bending mechanisms of a metal sheet according to MGT [3].

(0.17MB).

The process is relevant for ductile materials like steels [5], aluminum [6] and titanium [6] and is proven to be feasible on a laboratory scale for the forming of composite laminates such as glass-fiber aluminum [7].

A particularly interesting feature of the laser forming process is the possibility to combine it with conventional laser welding. For example, an aircraft panel can be formed at a given angle by the small residual heat of the welding process, directly producing the necessary curvature [8]. Alternatively, one can use a defocused laser beam to produce the required curvature, after the welding process [9].

Watkins et al. [6] found that the bending angle of an AA2024 aluminum sheet with 0.8mm thickness was between 0.3° and 6° per scan using a CO2 laser. The authors used a laser power ranging from 250 to 1300W and a speed between 10 and 140mm/s for a beam diameter of 10mm, and applied graphite to the surface of the aluminum to increase the absorptivity. The authors also noted the possibility of increasing the bending angle by increasing the number of the laser runs over the same track. Edwardson et al. [10] showed that increasing the number of runs to increase the angle reaches limit efficiency due to the loss of the absorber layer, the local deformation hardening, and loss of the initial geometry of the part, blocking the action of the laser.

The modeling of laser forming process is quite complex, having been initially performed by the finite element method by Ji and Wu [11]. Vollertsen and Shen [12] conducted a review of the laser beam forming from the point of view of analytical and numerical methods and empirical observations. The use of modern computational tools, such as the finite element analysis software Sysweld®[13], allows estimating the temperatures, stresses and deformations that the material undergoes during the LBF process. This software was used in this work.

The purpose of this work is to study the laser beam forming process using a laser fiber without the need for an absorbing layer. This laser allows joint welding operations and forming of the welded panels such as those used in a commercial aircraft. Thus, the step of forming the panels before welding, which causes problems in setup and delay the work of welding robots, could be eliminated. The choice of an Al–Mg–Mn–Si–Cu–Fe alloy, AA6013 under T4 condition, reflects the future demands of aerospace and general transportation sectors. Finally, the usefulness of the technique was tested by using LBF to straighten a T-joint laser welded panel.

2Materials and methods

The material used was a grade AA6013 alloy (Al–0.94Mg–0.27Mn–0.62Si–0.82Cu–0.20Fe, wt.%) aged at ambient temperature (T4) in the form of sheets with 1.6mm thickness. The coupons dimensions were 100mm (L)×50mm (W). Firstly, the samples were used as flat sheets and then T-joint laser welds were performed to test the efficiency of the process. Some selected thermal and mechanical properties of the aluminum sheet are summarized in Table 1[13]. In the table, K is the thermal conductivity, Cp is the specific heat, α is the thermal expansion coefficient, E is the Young's modulus and σy is the yield strength. The number between parentheses refers to the T4 aluminum condition (1) and the as-welded (2) condition. LS means liquid state, where mechanical strength does not apply – N/A.

Table 1.

Thermal and mechanical properties of the aluminum sheet.

T (°C)  K(1)
(W/mmK) 
K(2)
(W/mmK) 
Cp(1)
(J/kgK) 
Cp(2)
(J/kgK) 
α(1)
(m/K) 
α(2)
(m/K) 
E(1)
(MPa) 
E(2)
(MPa) 
σy(1)
(MPa) 
σy(2)
(MPa) 
20  0.175  0.180  780  898  2.75·10−6  2.75·10−6  73,000  73,000  324  130 
100  0.170  0.190  –  951  2.73·10−6  2.73·10−6  70,000  70,000  310  100 
200  0.170  0.190  –  –  2.71·10−6  2.71·10−6  64,000  64,000  131  54 
250  0.175  0.175  –  1003  –  –  58,000  58,000  62  40 
350  0.180  0.175  –  1055  2.69·10−6  2.69·10−6  –  –  41  32 
400  0.185  0.200  –  1108  2.66·10−6  2.66·10−6  51,000  51,000  28  25 
500  0.170  0.215  900  1195  2.63·10−6  2.63·10−6  41,300  41,300  10  10 
LS  0.200  0.235  1000  1300  2.23·10−6  2.23·10−6  N/A  N/A  N/A  N/A 

The laser was an Yb-fiber laser, model YLR-2000 of IPG Photonics, with maximum power of 2kW. The minimum beam diameter was 0.1mm, which corresponds to the diameter of the optical fiber itself. The material displacement was performed by a three-axis CNC table, and a vertical Z axis focused the laser beam on the material surface. For the experiments, the laser head was positioned 5mm above the focal plane, resulting in an effective focal diameter of 0.6mm on the sheet surface. The laser power was varied from 200 to 1000W, and the scan speed varied between 3 and 30mm/s. The laser made a track through the full width of the sheet (50mm) and the middle of the length. In the current setup, the laser tracks were always perpendicular to the sheet rolling direction. One side of the sample was firmly attached to a bench and the other was free to bend. The schematic experimental setup is shown in Fig. 2[14].

Fig. 2.

A schematic LBF process applied in the present study [14].

(0.07MB).

The microstructural analyses were performed by optical microscopy and the Vickers hardness test. Analyses by optical microscopy were performed on the cross section of the sheets, which were polished and etched with Keller's solution (2% HF, 10% HNO3 and 88% H2O) for 5s. The Vickers hardness test was performed with a load of 50gf applied for 9s.

The bending angle was experimentally obtained by measuring the height at the end of the sample with the use of a dedicated dial gauge. The dial was zeroed at the untreated sample surface and it was tightly fixed to a granite table.

The experimental data were compared with those obtained by the simulation software Sysweld®. The physical data were obtained from the AA6013 aluminum alloy Sysweld® database. The simulated sheet dimensions were the same as those of the real sheet, with a track of the laser treatment in the middle. The radiation losses followed the Stefan–Boltzmann Law for gray bodies, where the emissivity was 0.8. The heat transfer conditions were open air free convection and the radiation heat losses by all of six open surfaces at 25Wm−2. The laser beam heat source is considered Gaussian-type in agreement to our experimental measurement and the analytical model for energy input followed Goldak et al. theory [15]. The finite element analysis software calculated thermo-metallurgical data firstly (temperatures and phases) and then calculated the mechanical data (strain, stresses and deformations). All physical properties were temperature dependent, as shown in Table 1. Interpolation applies in the case of a missed temperature-dependent parameter was needed. The simulated speed and laser power are also the same as those used in the experiments. The laser absorptivity was set to 30%, since this value approximates the actual measurements of the laser marks. For each condition, the maximum temperature both at the upper and lower surface of the sheet, the residual stress on the lower surface of the sheet and the final bending angle were estimated.

After the flat sheet studies, T-joint welds were performed. The welding process was carried out without filler metal (autogenous), with a single pass from one side. Aluminum stringers of 100mm length by 20mm width were welded in a T-joint geometry to the skin sheets measuring 100mm by 100mm. The weld was made in the same laser system, but on the CNC table a special tool was used to ensure the perpendicularity of the parts, as shown in Fig. 3a. The welding parameters were: laser power 1.5kW, welding angle 12.5°, welding speed 6m/min and helium as process gas with a flow rate of 20 l/min. Fig. 3b shows an image of a T-joint sample just after welding [16]. LBF was performed on the bottom surface of the skin (Fig. 3b), exactly over the joint line.

Fig. 3.

Schematic of the special tool used in the welding of the T-joint (a) and an image of a T-joint sample (b).

(0.18MB).
3Results and discussion3.1Sheet forming

Table 2 presents the experimental conditions of the laser beam forming process used in this work. Conditions A to F produced a remelted track as shown in Fig. 4. The remelted track is more visible at conditions A, B, C and F and is harder to see in conditions D and E. In the case of conditions G to I, there was no visible change in the surface exposed to the beam, and therefore they are not illustrated in Fig. 4.

Table 2.

Process parameters.

Condition  V (mm/s)  P (W)  Fixed parameters 
1000  He gas shielding
Flow 15L/min
Defocusing the beam Δz=5mm
Beam diameter on the surface of the sheet=0.8mm
Laser head inclination=
10  1000 
30  1000 
500 
10  500 
30  500 
200 
10  200 
30  200 
Fig. 4.

Optical microscopy of the laser tracks cross section. The letters refer to the conditions presented in Table 1.

(0.74MB).

Fig. 5 shows the obtained bending angles as a function of the speed and laser power. As seen, the bending angle can vary between 0° and 2.5° for one laser run.

Fig. 5.

The bending angle measured in terms of speed and laser power.

(0.22MB).

The melting of the sheet surface during the laser processing increased the folding angle, when compared to solid-state processing. The conditions A to D produced the angles between 2.11° and 2.52° after being processed with the laser. Experiments under conditions E and F show angles of 0.74° and 0.62° respectively. Finally, heating in the solid-state (conditions G, H and I) shows bending angles between 0.07° and 0.32°. This indicates that the solidification shrinkage has a more important role than the effect of thermal expansion.

Regardless of the conditions, there is no obvious difference in the hardness levels in the laser processed volumes. The hardness were between 100 and 160HV in all analyzed regions: base material, heat affected zones, fusion zones and plastically deformed zones. Fig. 6 shows the measured Vickers hardness for condition A, as a function of the distance from the center of the fusion zone. The upper surface corresponds to where the laser was focused. In the range where fusion was observed, up to about 0.7mm with condition A, as seen in Fig. 6, a small variation in hardness can be noticed, which lies slightly above the base material (HV 100).

Fig. 6.

Measurements of Vickers hardness as a function of distance, where x=0 represents the middle of the track. Melt range and upper and lower surfaces points were indicated. Condition A.

(0.08MB).

The depth of interaction must be left sufficiently low not to affect the metallurgical structure of the sheet exposed to the external environment part (skin). As can be seen in Fig. 4, the depths of the remelted layer were 0.3, 0.5, 0.6 and 0.4mm for conditions A, B, C and D, respectively. This equals to up to 38% of the sheet thickness. This value is low enough to be removed by the machining during crenellation of the panel. However, in any case, the C condition must be discarded since it had porosities in the bead.

The fact that there were no changes in hardness before and after laser forming implies that one can expect no significant changes in the static mechanical behavior of the panel after the process. However, this result is not sufficient to prove the usefulness of this technology, since the panel is subject to several factors, such as structural fatigue and corrosion.

Since the actual dimensions were very small, it was not possible to measure the process temperatures during LBF. The results obtained with the software Sysweld were: Ttop – the maximum temperature at the top in the center of the sheet, Tver – the maximum temperature in the verso of the sheet, TR – the residual stress in the back of the sheet and the αb – the final bending angle. Fig. 7 shows the evolution of the temperatures for different speeds and laser powers. The horizontal dotted line is the liquidus temperature of the alloy in question, showing that fusion occurs at the surface of the samples processed at 1000W, and those processed at 500W and 3–10mm/s. In the back of the sheet, the temperatures were always below the melting point.

Fig. 7.

Estimated sheet temperatures at the top surface (Ttop) and in verso (Tver).

(0.11MB).

Fig. 8 shows the calculated residual stress obtained on the back of the sheet. The value of residual stress varies between 0 and 100MPa, depending on the conditions, and tended to increase with increasing laser power. The higher values of residual stress are the conditions in which the material was remelted (Fig. 4), indicating the strong effect of shrinkage during solidification. Because of the thin laser track and the bending of the sheet, it was not possible to evaluate the residual tension by analytical methods, such as X-ray diffraction.

Fig. 8.

Calculated results of residual stress as a function of process parameters V and P.

(0.16MB).

It could be noted that Figs. 8 and 5 are connected, since the force for bending comes from the residual strain. Both curves presented marked increases in the bending angle and the residual stresses when the laser power increased. Also, both curves presents a limit for the bending angles (600W; 30mm/s) and for the residual stresses (1000W; 20mm/s) indicating some related phenomenon. When liquid appears, the tensile residual stresses were relaxed and the bending attains its limiting value. Since the tensile residual stresses were measured on the back side of the sample (positive bending), the surface was not remelted and the maximum stress attained 110MPa (Fig. 8).

The bending angles calculated for each condition can be seen in Fig. 9. The simulated bending angle, as a function of V and P, points to a similar trend as the one observed in the experiments. However, the value of the theoretical angles is approximately half of those obtained experimentally. This may reflect the clamping conditions of the sheets during the process. In the simulation, the sheets are free to move, whereas in the experiments the sheets need to be fixed on one side to maintain the alignment of the laser treatment.

Fig. 9.

Bending angles calculated for different powers and speeds.

(0.18MB).
3.2T-joints

Laser welded T-joints were carried out using the AA6013-T4 sheets as presented in Fig. 10. Fig. 10a shows the coupon after being sectioned from a larger panel and. Fig. 10b presents a micrograph of the welded region together with the LBF application region.

Fig. 10.

(a) Laser beam welded coupon and (b) sample cross-section indicating the joint and the LBF interaction region.

(0.18MB).

Following the same methodology of the flat sheet experiments, Condition A was applied to the verso of the welded coupons (Fig. 10). Differently from the former experiments, one, two or three runs were applied at the same location. Fig. 11 presents the as-weld bend angle and the resulting angle after a number of LBF runs. As can be seen, the efficiency of LBF was quite variable depending on the real as-weld conditions and the number of runs. As a general trend, one could estimate the overall efficiency around 80%, although, for some conditions, the initial bending was reduced to zero. Additionally, the efficiency of LBF seems to be independent of the number of runs.

Fig. 11.

Measured angles before and after LBF for laser welded stringer-skin sets. The line indicates a general trend.

(0.1MB).

The potential use of the technique for the welding engineering seems to be very large, since aluminum welded structures are found anywhere from ordinary windows to airplanes. As proposed here, the laser beam forming is a step after laser welding and thus only applies to the high production rate, medium cost manufacturing. The limitations of the technique belong mainly to the access at the rear part of the T-joined panels and due the laser mark. The Condition A (Fig. 4) produces a small remelted zone, the laser mark, which could eventually change the mechanical and corrosion behavior of the panel. This fact must be taken into account for some application such as aerospace structures. The use of low power conditions, avoiding fusion (Conditions E to I, Table 2), could be a possible solution to undesired metallurgical problems. However, on the other hand, the unbending efficiency (Fig. 9) will be greatly reduced.

4Conclusions

Non-contact sheet forming using a laser for a 1.6mm thick AA6013-T4 was proposed. A high power fiber laser can provide sufficient intensities both for welding and forming operations without the need of additional equipment. Within the range of experimental parameters used, specimens with and without melting were obtained. The coupons that presented the highest bending angles were those where there was melting, which never exceeded 38% of the sheet thickness. A folding angle of 2.5° was obtained from a power of 1000W and a speed of 30mm/s. The simulation results allowed to estimate the temperatures, residual stresses and bending angles fairly accurately, but the model must be refined to take into account the actual experimental conditions. Considering the application of technology in shaping aircraft panels of the skin-to-stringer type, it was shown that the process is fully applicable to a real case. The efficiencies in unbending were around 80%, for one to three laser runs. In some cases, it is possible to straighten the entire welded panel.

Conflict of interest

The authors declare no conflicts of interest.

Acknowledgement

The authors thank the Empresa Brasileira de Aeronáutica (EMBRAER) for supplying of the aluminum sheets.

References
[1]
G. Dearen, S.P. Edwardson.
Laser assisted forming for ship building.
Proceedings of SAIL 2003,
[2]
Y. Shi, Z. Yao, H. Shen, J. Hu.
Research on the mechanisms of laser forming for the metal sheet.
Int J Mach Tools Manuf, 46 (2006), pp. 1689-1697
[3]
Economic advanced shaping processes for integral structures (ECOSHAPE). Avaliable from: http://cordis.europa.eu/project/rcn/72813_en.html [accessed 10.10.2015].
[4]
J. Zhong, S. Wu.
FEM simulation of the temperature field laser during the forming of sheet metal.
J Mater Process Technol, 74 (1998), pp. 89-95
[5]
L.J. Yang, J. Tang, M.L. Wang, Y. Wang, Y.B. Chen.
Surface characteristic of stainless steel sheet after forming pulsed laser.
Appl Surf Sci, 256 (2010), pp. 7018-7026
[6]
K.G. Watkins, S.P. Edwardson, J. Magee, G. Dearden, P. French, R.L. Cooke, et al.
Laser forming of aerospace alloys.
Proceedings of AMTC,
[7]
C. Carey, W.J. Cantwell, G. Dearden, K.R. Edwards, S.P. Edwardson, K.G. Watkins.
Towards a rapid, non-contact method for shaping metal laminates using fiber laser source.
Int J Adv Manuf Technol, 47 (2010), pp. 557-565
[8]
J.F. Jullien, A. Combescure, M. Zain-ul-abdein.
Investigation of laser-beam welding induced distortions in thin sheets of Al 6056T4.
Structures and Granular Solids, pp. 145-151
[9]
T. Hornfeck, A. Trautmann, F. von Oefele.
Laserstrahlumformen Aluminiumlegierungen fuer Luft und Raumfahrt dir, IWB, Report.
Institut fuer und Werkzeugmaschinen Betriebswissenchaften, (2009), pp. 1
[10]
S.P. Edwardson, E. Abed, C. Carey, K.R. Edwards, K. Bartkowiak, G. Dearden, et al.
Key factors Influencing the bend per pass in laser forming.
Proceedings of ICALEO 2007 congress proceedings, laser materials processing conference, pp. 508-514
[11]
Z. Ji, S. Wu.
FEM simulation of the temperature field laser during the forming of sheet metal.
J Mater Process Technol, 74 (1998), pp. 89-95
[12]
H. Shen, F. Vollertsen.
Modelling of laser forming – an review.
Computat Mater Sci, 46 (2009), pp. 834-840
[13]
ESI Group. Welding & assembly. Avaliable from: https://www.esi-group.com/software-services/virtual-manufacturing/welding-assembly [accessed 10.10.2015].
[14]
J. Magee, L.J. De Vin.
Process planning for laser-assisted forming.
J Mater Process Technol, 120 (2002), pp. 322-326
[15]
J.A. Goldak, A.P. Chakravarti, M. Bibby.
A new finite element model for welding heat sources.
Metall Trans, 15B (1984), pp. 299-305
[16]
R.H.M. Siqueira, A.C. Oliveira, R. Riva, C.A.R.P. Baptita, M.S.F. Lima.
Mechanical and microstructural characterization of laser-welded joints of 6013-T4 aluminum alloy.
J Braz Soc Mech Sci Eng, 37 (2015), pp. 133-140
Copyright © 2016. Brazilian Metallurgical, Materials and Mining Association
Journal of Materials Research and Technology

Subscribe to our newsletter

Article options
Tools
Cookies policy
To improve our services and products, we use cookies (own or third parties authorized) to show advertising related to client preferences through the analyses of navigation customer behavior. Continuing navigation will be considered as acceptance of this use. You can change the settings or obtain more information by clicking here.