Electromagnetic Joining of Hybrid Tubes for Hydroforming

View metadata, citation and similar papers at core.ac.uk brought to you by CORE provided by Elsevier - Publisher Connector Available online at www.sciencedirect.com ScienceDirect Procedia CIRP 23 (2014) 1 – 6 5th CATSConference 2014 - CIRP on Assembly Systems and Technologies onConference Assembly Technologies and Systems Electromagnetic joining of hybrid tubes for hydroforming Verena Psyk*, Thomas Lieber, Petr Kurka, Welf-Guntram Drossel Fraunhofer Institute for Machine Tools and Forming Technology, Reichenhainer Strasse 88, 09126 Chemnitz, Germany * Corresponding author. Tel.: +46-371-5397-1731; fax: +49-371-5397-61731. E-mail address: [email protected] Abstract Joining of functionally adapted and weight-optimized semi-finished parts by electromagnetic compression is investigated taking into account that the resulting component is further processed by hydroforming. For this purpose both ends of an aluminum tube are connected to steel tubes. During the subsequent hydroforming of the hybrid tube the joints are exposed to extreme loading. Finite-Element-Simulation is applied for characterizing and quantifying these loads. The electromagnetic joining process is analyzed, the joints are characterized by microstructural analysis and the resulting properties of the joint are evaluated by model experiments. © by Published Elsevier B.V. This is an open access article under the CC BY-NC-ND license © 2014 Published The Authors. by Elsevier B.V. Selection and peer-review under responsibility of the International Scientific Committee of 5th CATS 2014 in the person of the Conference (http://creativecommons.org/licenses/by-nc-nd/3.0/). Chair Prof. Dr.peer-review Matthias Putz [email protected]. Selection and under responsibility of the International Scientific Committee of 5th CATS 2014 in the person of the Conference Chair Prof. Dr. Matthias Putz [email protected] Keywords: Electromagnetic forming, joining, hydroforming 1. Introduction and motivation Against the background of climate change and global warming, social pressure as well as legal restraints force industry not only to consider technological and economic aspects but also to attach great importance to the ecological tenability of new developments [1]. This concerns especially – although not exclusively – the automotive industry. According to WALTL, the reduction of CO2-emissions is one of the most important challenges in automotive manufacturing and the consequent implementation of lightweight construction concepts is a key strategy to cope with this demand [2]. Maximum weight reduction can only be achieved if all aspects of lightweight strategies including material as well as construction-related aspects are ideally exploited. As a result, the material distributions as well as material properties have to be ideally adapted to the according load profile and function of the component [3]. Frequently, this can only be achieved by combining different materials – as it is e.g. done in the multimaterial body of the Audi A3 [4] – and increasing the geometric complexity of the components. This leads to high demands on forming and joining technologies. 1.1. Hydroforming as a lightweight construction technology Hydroforming is a technology with high potential considering forming of complex shapes from tube and sheet metal material as it is impressively shown by the application examples presented in [5-7]. According to [8], the overall success of hydroforming products significantly depends on the applied semi-finished parts. In principle, tubes and hollow-profiles made of different materials including typical lightweight materials as e.g. aluminum alloys, high-strength steels, and under certain conditions even magnesium alloys can be used, but maximum benefit with regard to lightweight design can be achieved by applying tailored semi-finished parts. This means that optimized distribution of the material and the material properties in the finished component are achieved by applying blanks or tubes with locally different thickness (tailored blanks), locally different grades (tailored heat treated blanks), or even locally different materials (hybrid blanks), so that in contrast to the application of conventional material with homogenous properties local oversizing is avoided [3]. Tailored tubes for tube hydroforming can be round seam welded or longitudinally seam welded [8]. 2212-8271 © 2014 Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/). Selection and peer-review under responsibility of the International Scientific Committee of 5th CATS 2014 in the person of the Conference Chair Prof. Dr. Matthias Putz [email protected] doi:10.1016/j.procir.2014.10.063 2 Verena Psyk et al. / Procedia CIRP 23 (2014) 1 – 6 1.2. Electromagnetic forming and magnetic pulse welding Welding of different materials is frequently related to severe problems or might even be completely impossible. In many of these cases, impact welding processes as magnetic pulse welding offer an interesting alternative. Magnetic pulse welding is a process variant of electromagnetic forming frequently using the compression setup and resulting in a metallically bonded joint [9]. For electromagnetic compression a tubular workpiece is positioned in a cylindrical inductor. Due to a sudden discharge of a connected capacitor, damped sinusoidal current flows through the inductor, inducing an according magnetic field and a current in the tube– provided that the latter is made of highly electrically conductive material. Due to the interactions of the currents and the magnetic field, Lorentz forces result and if the resulting stresses in the material reach the yield stress, workpiece compression is caused. During this deformation the workpiece is accelerated within several microseconds and can achieve velocities in the magnitude of 102 m/s and strain rates in the magnitude of 103 to 104 s-1 [10]. For welding tubes by electromagnetic compression, the ends of the two joining partners are positioned coaxially oriented, overlapping each other, and with a small gap inbetween them in the cylindrical inductor. The above mentioned procedure causes deformation of the outer tube, so that it collides with the inner one. If the collision parameters (i.e. especially the impacting angle and velocity) are suitable, a jet effect occurs sweeping away oxides and impurities. Thus, the workpiece surfaces are cleaned and accordingly reactive. Due to the close approximation and supported by the contact pressure, atomic bonding of the surfaces results, which is often – but not necessarily – related to a wavy shape of the contact zone. [11] The advantages of magnetic pulse welding as high strength of the joint and none or significantly less formation of intermetallic phases compared to thermal joining have frequently been described in literature [11-15]. of this special process combination. However, in contrast to [17], where form-fit joints are preferred, the focus is set on magnetic pulse welding, here, because it is expected that the loads acting on the joint during the hydroforming process are extremely high and the according demands on the connection cannot be fulfilled by interference-fit or form-fit joints. 2. Description of the manufacturing task For the analysis, an aluminum tube ((EN AW-6060-O; ø76 mm x 2.5 mm; length: 340 mm) shall be connected at both sides to steel tubes (St34 mod; ø76 mm x 2.5 mm; length: 290 mm). The resulting hybrid tube part shall be mechanically preformed (crash formed) to allow insertion into the hydroforming die for the final forming with pressurized fluid. The welding zone has to withstand significant cross section deformations during crash forming and for hydroforming leak tightness of the connection has to be guaranteed. Considering lightweight design aspects, the overlap of the joining partners should be as short as possible. Here, a length of 10-30 mm is a realistic value. 3. Quantification of the weld loading In order to estimate the loads acting in the welding zone a simplified finite-element-analysis (FEA) of the crash forming and the hydroforming was carried out. Thereby, only a section of the work piece with an axial length of 90 mm including one joint in the middle of the regarded area was considered. The tubes were modeled in a simplified way as cylindrical components with an overlap of 10 mm and also a simplified tool geometry was considered. The latter one was derived from the geometry of the real part by extruding the approximately rectangular cross section in the joint area featuring a width of 98 mm, a height of 42.5 mm and a corner radius of 15 mm. 1.3. Combined processes In several cases, electromagnetic forming technologies have successfully been combined with other, more conventional forming technologies as extrusion, bending, or deep drawing in order to increase forming limits [9]. In [16] electromagnetic compression was already used for realizing optimized semi-finished parts for hydroforming. The technology was applied for locally adapting the circumference of straight or bended tubes. Only by applying such precontoured components in a subsequent hydroforming operation the desired complex part geometry could be produced in aluminum. In [17] combined electromagnetic joining and subsequent hydroforming is suggested, but information about the feasibility of the process combination is missing here and has also not been published elsewhere. To fill this gap, one aim of the joint research project „Skalierbare Module aus Antrieb und Achse für die Elektromobilität – ESKAM“ is using joining by electromagnetic forming for producing hybrid tubes for hydroforming and thus proving the feasibility and the potential Fig. 1. 3D-FEA model for simplified numerical analysis The 3D-FEA-model based on volume elements is shown in Fig. 1 as well as the considered process combination (crash forming and hydroforming). The aluminum tube (EN-AW 6060-O) is geometrically considered as former mentioned. The steel tube is assumed as a St34 mod material with an outer Verena Psyk et al. / Procedia CIRP 23 (2014) 1 – 6 diameter of 71 mm and a thickness of 2.5 mm. This implies a former reduction of the steel tube diameter as necessary for the magnetic pulse welding process. Fig. 2. Shear force to be transferred by the joint The tube ends are considered as clamped in axial direction. In the joined area the tubes are assumed as contact zones with sticking properties, so no slip and no contact release may occur and a direct force transition is possible. After crash forming the hydroforming process starts with a linear ramp up of pressure up to 800 bar. The resulting shear force within the tube interacting zone is displayed in Fig. 2. During crash forming compressive forces are acting within the joined area. Within the hydroforming process those change to tensile forces due to the increasing diameter of the tubes and respectively the material flow out of the joined area. The maximum shear force of 42 kN is achieved at a level of 500 bar. At this certain point the contact zone is completely formed into the die, so that no further increase of the shear force is possible. Fig. 3. Setup for magnetic pulse welding of hybrid tubes. In the experiments a capacitor charging voltage of 16 kV corresponding to a capacitor charging energy of 42 kJ was applied. Tests were carried out with and without a supporting mandrel. In principle for both cases metallically bonded joints could be realized. This was shown on the basis of specimens that were cut from the joining area. The influence of the supporting mandrel on the geometry and the microstructure can be seen in Fig. 4. 4. Magnetic pulse welding of hybrid tubes To enable magnetic pulse welding of the two tubes featuring the same diameter, the steel tubes were subject to a preceding mechanical forming aiming at reducing the diameter at the ends. This preforming operation was performed by Salzgitter Hydroforming. As a consequence the steel tubes feature significant marks, which have been smoothed by grinding. This preparation of the joining zone simultaneously removes oxide layers, so that the surfaces become more reactive. The magnetic pulse welding experiments were carried out at the Fraunhofer Institute for Machine Tools and Forming Technology using a pulsed power generator PS103-25 Blue Wave by PSTproducts with a capacitance of 330 μF and an inductance of 85 nH (including the collector and inductor connection). The according setup for the magnetic pulse welding includes an inductor system that was especially designed for this specific joining task (compare [18]). The inductor system was built by Poynting GmbH. The principle setup is shown in Fig.3. Fig. 4. Contour and micrographs of joints a) without and b) with a mandrel Comparing the contour, joining without supporting mandrel leads to more diameter reduction even deforming the inner 3 4 Verena Psyk et al. / Procedia CIRP 23 (2014) 1 – 6 steel tube remarkably. The joining zone shows significant local curvature. Consequently, these joints can be expected to feature stronger strain hardening effects and higher geometric stiffness compared to the straighter ones achieved with a mandrel. Both effects necessitate higher forces during hydroforming in order to reverse the bulging. As a consequence, more expensive equipment will be required for hydroforming. However, what is even more important concerning the feasibility of the combined magnetic pulse welding and hydroforming is the consideration of the accumulated strain in the hybrid tube over the complete production chain. The more deformation is introduced in the mechanical preforming operations and during magnetic pulse forming the less formability is available for hydroforming (see also [16]). Moreover, the diameter reduction introduced during mechanical preforming and magnetic pulse welding has to be reversed by hydroforming. Consequently, necessary deformation during hydroforming increases if the compression is more significant. This means that the compression should be as slight as possible. Therefore, joining with a supporting mandrel seems to be more appropriate considering the influence of the different geometries on the parameters of the subsequent hydroforming process as well as on the formability of the hybrid tube. Additionally, the microstructure of the steel tube clearly shows an influence related to the application of a supporting mandrel. In case of joining without mandrel the grain structure close to the contact surface is significantly finer compared to that farer away from the contact surface. In case of joining with a supporting mandrel this effect was not observed. Due to the extreme and local occurrence of the grain refinement, it is not likely that higher deformations due to the more severe compression and the curvature of tube surface in the joining area are the reason for the phenomenon. However, the microstructure reminds of a surface that has been hardened by hammering. The difference in the microstructures of the specimens joined with and without a mandrel suggests that also the impacting process develops differently. The unsupported steel tube gives way to the moving aluminum tube by being compressed itself. Thereby, the deceleration of the aluminum tube is related to the deformation – i.e. at first the acceleration and later the deceleration of the steel tube. This might include multiple small impacting processes similar to hammering the surface. In contrast, the steel tube that is supported by the mandrel cannot be deformed significantly. In this case the aluminum tube impacts with the steel tube and is stopped abruptly. Consequently, the microstructure is similar to cases, in which tubes are magnetic pulse welded to massive inner parts [19, 20]. 5. Estimation of the feasibility of the process chain on the basis of model experiments If and how these differences influence the joint quality was investigated by suitable model experiments. These tests also serve as a basis for a first estimation of the feasibility of hydroforming of magnetically pulse welded hybrid tubes. 5.1. Tensile tests As a first model experiment, tensile tests of the joint were carried out. The according specimens were milled from the hybrid tubes. The specimen geometry is shown in Fig. 5 together with the most important results of these tests – i.e. relevant strength values and photos of the torn specimens indicating the failure mode. Fig. 5. Tensile tests of joints a) without and b) with a mandrel The most important conclusion from these tests is that, with the exception of very few specimens, failure typically occurred in the aluminum base material. This shows that the endurable stress in the joint area is higher than the tensile stress of the aluminum tube, which is slightly below 100 MPa, here. In case of the tubes welded with supporting mandrel, visual inspection of the joint area shows no damage or deterioration in this zone. For the specimens joined without supporting mandrel, a slight detachment of the aluminum and the steel tube can be observed at the end of the joint. This phenomenon occurs already early in the process at a stress in the range of 15-35 MPa and is related to a temporary drop of the transferable tensile force and the according stress. However, after overcoming this critical load, the joint strength is again higher than the tensile stress of the aluminum material. Nevertheless, the joints produced with supporting mandrel turn out to be better compared to the specimens joined without support. Apart from this general conclusion, the results of the tensile tests can also be exploited for a first feasibility estimation of the combined process. An important output value from the numerical simulation, presented in section 3, was that during crash forming and subsequent hydroforming the joint has to withstand a maximum total force of Verena Psyk et al. / Procedia CIRP 23 (2014) 1 – 6 approximately 42 kN. Assuming that this force is homogeneously distributed along the circumference of tube and joint, respectively, a circumferential section of 10 mm length, corresponding to the width of the tensile specimens, has to transfer a maximum force of 1.9 kN. This value corresponds more or less to the yield stress of the aluminum and the joined specimens, respectively. The maximum transferable force is in the range of 2.5 kN. This suggests that under the precondition that the hydroforming geometry can be achieved with the regarded aluminum material, the joint can be expected to withstand the load. detachment of the steel and the aluminum tube might have occurred (compare Fig. 6a, last forming step). 5.2. Cross section deformation In order to evaluate the performance of the joint zone during crash forming, a model experiment representing this load case was developed. Similar to the numerical simulations, a simplified tool geometry was considered. However, the corner radii of the rectangular cavity were disregarded here to further simplify the model experiment. Since corner filling is achieved during pressurization in the hydroforming process and not during crash forming this simplification is reasonable. The tool, consisting of a flat bottom, straight rectangular walls with an adjustable distance in-between them, and an open upper side, is mounted to a press. Wall distance was set to 98 mm. During the press stroke, a flat punch dips into and thus closes the cavity. The height of the resulting rectangle depends on the press stroke. During the experiments it was successively decreased in order to show the course of the deformation and its influence on the joining area. The minimum regarded height of 44 mm is in good agreement with the real part’s cross section height in the joining zone. Considering the initial outer diameter of the hybrid tube of 76 mm, this corresponds to a press stroke of 32 mm. In the experiments, tube sections with a length of 195 mm were tested. The joining zone was located in the middle of these specimens. Fig. 6 shows the axial view into the hybrid tube at different forming stages as well as the joint after the last forming step. Analogues to the investigations shown above, hybrid tubes joined with and without supporting mandrel are compared. As it was already predicted by the simulation (compare Fig. 1 Crash forming) the tube buckles at the long sides of the rectangle during the cross section deformation. With increasing press stroke and accordingly increasing deformation this effect becomes more pronounced. Moreover, it can be seen, that the effect is not absolutely symmetric. Considering the aluminum tube, the buckling is more significant at the lower side while buckling of the steel tube is more pronounced at the upper side. Higher wall-thickness and accordingly higher stiffness of the joining zone leads to less buckling in this area compared to the individual tubes. However, it cannot be completely avoided here, either. Comparing the hybrid tube joined with supporting mandrel to that joined without any support, it is remarkable that in case of joining without mandrel the aluminum tube buckles less, while buckling of the steel tube and of the joining zone is stronger. The axial view into the tube suggests that a local Fig. 6. Cross section deformation tests of specimens a) joined without and b) joined with a mandrel For a first estimation of the suitability of such deformed hybrid tubes for being further processed by hydroforming, the watertightness of the specimens was tested. Therefore, one end of the hybrid tube was sealed before filling it with water (no pressurization). The tube joined with supporting mandrel turned out to be watertight even after maximum deformation of its cross section, the tube joined without support showed significant leakage in the region, where detachment of aluminum and steel tube was already assumed. This supports the better evaluation of the tubes joined with mandrel already made on the basis of the micrographic analysis and the tensile tests. It clearly proofs that for magnetic pulse welding of hybrid tubes for hydroforming such a support is absolutely recommended. In the regarded case the hybrid tubes joined without support have to be assessed as unsuitable for further processing by hydroforming. 5 6 Verena Psyk et al. / Procedia CIRP 23 (2014) 1 – 6 6. Summary In order to provide optimized semi-finished parts for hydroforming, magnetic pulse welding of hybrid tubes consisting of an aluminum middle section and steel end sections was suggested. To estimate the feasibility of the subsequent forming operations, a combined numerical and experimental approach was applied. Numerical simulations were carried out, assuming an ideal joint. The forces to be transferred during further processing of the semi-finished parts hydroforming were quantified via the simulation. Complementing magnetic pulse welding experiments were carried out. The weld properties and especially the influence of applying an inner supporting mandrel were characterized by micrographic investigations, tensile tests and cross section deformation tests. It was shown that applying such a mandrel avoids unnecessarily strong deformations of the tubular aluminum steel composite as well as local grain refinement in the surface of the steel tube. The tensile tests have shown that the magnetically pulse welded region is stronger compared to the aluminum, leading to failure in the aluminum base material in nearly all cases. Comparing the forces determined in the numerical simulation to those measured during tensile tests suggests that the joint will withstand the load if the hydroforming geometry can be achieved with the regarded aluminum material. However, joining with supporting mandrel again shows a benefit. The cross section deformation tests and especially a watertightness test performed on the deformed specimens finally proofed that for magnetic pulse welding of hybrid tubes for hydroforming, applying a supporting mandrel is absolutely recommended. The tubes joined without support turned out to be untight and consequently not suitable for further forming with a fluid medium, here. For the final feasibility proof of the suggested process combination of magnetic pulse welding and subsequent hydroforming, technological tests will be carried out in close cooperation of Fraunhofer Institute for Machine Tools and Forming Technology and Salzgitter Hydroforming GmbH. Acknowledgements The presented study has been performed by the authors within the joint research project „Skalierbare Module aus Antrieb und Achse für die Elektromobilität – ESKAM“. The project has been funded by the Federal Ministry of Education and Research and the consortium has the following partners: Hirschvogel Automotive Group, Groschopp AG, PTSPrüftechnik GmbH, REFU Elektronik GmbH, Wilhelm Vogel GmbH, Salzgitter Hydroforming GmbH & Co. KG, Wilhelm Funke GmbH & Co. KG, Fachhochschule Düsseldorf, Fraunhofer Institute for Machine Tools and Forming Technology, Aalen University, and University of Stuttgart. References [1] Meleghy, G.; Bayer, S.: Die Automobilzulieferindustrie – Perspektiven innerhalb einer globalen Branche im Strukturwandel am Beispiel Tower Automotive. In: Manufacture of Car Bodies – Sustainable and Efficient; Verlag Wissenschaftlicher Scripten, 2011, ISBN: 978-3-642267-32-8, pp. 25-38. [2] Waltl, H.: Produktionstechnische Herausforderungen an den Karosseriebau durch aktuelle Trends. In: Manufacture of Car Bodies – Sustainable and Efficient; Verlag Wissenschaftlicher Scripten, 2011, ISBN: 978-3-642267-32-8, pp. 13-23. [3] Kleiner, M.; Chatti, S.; Klaus, A.: Metal forming techniques for lightweight construction. Journal of Materials Processing Technology 177 (2006) 2-7. [4] AUDI AG (editor): Workshop Audi A3. Product and Technology Communications, May 2012, www.audi-mediaservices.com (shown on October 11th, 2013). [5] Hartl, C.: Research and advances in fundamentals and industrial applications of hydroforming. Journal of Materials Processing Technology 167 (2005) 383-392. [6] Lang, L. H.; Wang. Z. R.; Knag. D. C.; Yuan, S. J. Zhang, S. H.; Danckert, J.; Nielsen: Hydroforming highlights: sheet hydroforming and tube hydroforming. Journal of Materials Processing Technology 151 (2004) 165-177. [7] Vollertsen, F.: Accuracy in process chains using hydroforming. Journal of Materials Processing Technology 103(2000) 424-433. [8] Koç, M.; Altan, T.: An overall review of the tube hydroforming (THF) technology. Journal of Materials Processing Technology 108 (2001) 384393. [9] Psyk, V.; Risch, D.; Kinsey, B. L.; Tekkaya, A. E.; Kleiner, M.: Electromagnetic Forming – A Review. Journal of Materials Processing Technology, 211 (2011) 787–829. [10] Beerwald, C.: Grundlagen der Prozessauslegung und -gestaltung bei der elektromagnetischen Umformung. Dr.-Ing.-Dissertation, TU Dortmund, 2005, ISBN 3-8322-4421-2. [11] Shribman, V.; Tomer, Y.: Magnetic puse technology for improved tube joining and forming. In: Tube & Pipe Technology, pp. 91-95, 2006. [12] Kore, S. D.; Imbert, J.; Worswick, M. J.; Zhou, Y.: Electromagnetic impact welding of Mg to Al sheets. Science and Technology of Welding and Joining 14(6), 549-553, 2009. [13] Göbl, G.; Kaspar, J.; Herrmannsdörfer, T.; Brenner, B.; Beyer, E.: Insights into imtermetallic phases on pulse welded dissimilar metal joints. In: Proceedings of the 4th International Conference on High Speed Forming – ICHSF 2010, Columbus, pp. 127-136. [14] Psyk, V.; Gershteyn, G.; Barlage, B.; Weddeling, C.; Albuja, B.; Brosius, A.; Tekkaya, A. E.; Bach, F.-W.: Process Design for the Manufacturing of Magnetic Pulse Welded Joints. Key Engineering Materials Vol. 473, S. 243-250, 2011. [15] Aizawa, T.; Kashani, M.; Okagawa, K.: Application of magnetic pulse welding for aluminium alloys and SPCC steel sheet joints. Welding in the world, Vol. 49, pp. 212-222, 2005. [16] Psyk, V.: Prozesskette Krümmen – Elektromagnetisch Komprimieren – Innenhochdruckumformen für Rohre und profilförmige Bauteile. Dr.-Ing. Dissertation, TU-Dortmund, 2010, ISBN 978-3-8322-9026-9. [17] Dudziak, K.-U.; Falke, J.: Gesamtbauteil aus wenigstens zwei Einzelbauteilen und Verfahren zu dessen Herstellung. Patent application DE 10 2008 048 216 A1, 2008. [18] Psyk, V.; Scheffler, C.; Drossel, W.-G.: Werkzeugauslegung für das elektromagnetische Fügen. Zeitschrift für wirtschaftlichen Fabrikbetrieb – ZWF 11/2013. [19] Faes, K.; Baaten, T.; De Waele, W.; Debroux, N.: Joining of Copper to Brass Using Magnetic Pulse Welding. In: Proceedings of the 4th International Conference on High Speed Forming– ICHSF 2010, Columbus, pp. 84-96. [20] Elsen, A.; Ludwig, M.; Schaefer, R.; Groche, P.: Fundamentals of EMPT-Welding. In: Proceedings of the 4th International Conference on High Speed Forming– ICHSF 2010, Columbus, pp. 117-126.