Back-contact Cell Assemble

A novel photovoltaic-module assembly system for back contact solar cells using laser soldering technique Marc Köntges1*, Maren Gast1, Rolf Brendel1, Rüdiger Meyer2, Alexander Giegerich3, and Paul Merz3 1 Institut für Solarenergieforschung Hameln (ISFH), Am Ohrberg 1, D-31860 Emmerthal, Germany 2 Stiebel Eltron GmbH & Co.KG, Dr.-Stiebel-Straße, D - 37603 Holzminden, Germany 3 Reis GmbH & Co Maschinenfabrik, Walter-Reis-Straße 1, D - 63785 Obernburg, Germany, www.reisrobotics.de, phone: ++49 6022 503-0 ABSTRACT: We present a prototype of a new module assembly machine, which addresses five aspects of current fabrication issues: (i) Increase of module production speed for (ii) cells with both contacts on the rear side, (iii) use of lead free solders, (iv) reduction of cell handling and (v) applicability to thin solar cells. We name the prototype ATLAS. The ATLAS system lays up back contact cells directly onto the module lamination foil and solders the interconnectors to the cells using a laser. Our newly developed prototype system handles each cell only once. The problem of accumulation of particles in the soldering station is avoided as each new lamination sheet serves as a clean substrate. The system is capable of assembling back-contacted cells. Cell and cross connection is done in one machine. Our ATLAS prototype system solders a complete connector bone between two solar cells in 5 s with one 200 W diode laser and one handling system. For a production system the speed may easily be increased to 2 s per cell by using three diode lasers. Compared to a standard stringer, the ATLAS system concept will double the productivity. Keywords: PV module, laser soldering, module production 1. Introduction New back contact solar cells are under development by various solar cell manufactures. Four concepts of cell design are in development or already in production: Metallization Wrap Through MWT [1,2,4], Emitter Wrap Through (EWT) [3,4], Metallization Wrap Around (MWA) [1-4] and Interdigitated Back-contacted (IBC) [5-7]. Some of the designs require an additional insulation on the interconnector (e.g. Pin up Module) [8] or an insulation on the cell itself (busbarless emitter warp through) [9]. Other cell designs require a conventional ribbon or must have a special bone shaped interconnector. Thus the number of interconnection patterns is high. Most of the current module production lines are not able to interconnect such cells. Beside this great variety of interconnection designs for back contact cells, there are two main categories of module assembling. 1. Today’s production lines assemble the solar cells one by one into a string by soldering. Afterwards the soldered strings are transported to a lay up station. The strings are arranged onto a lamination foil and then laminated. This technique is hard to adept to new contact patterns, since the stringer are specialised for one kind of interconnector. 2. Some new techniques arrange the cells direct onto the lamination foil. The conventional interconnectors are replaced by a conductive pattern on the rear side foil. This technique is flexible to interconnect various back contact cell designs, because only the back sheet has to be changed. The interconnection between cells and the pattern on the back sheet foil is done by conductive adhesives or low melting solders [11]. The lamination and interconnection is done in one process step. These concepts are called Monolithic Module Assembly (MMA). The MMA is a very smooth module processes for thin and fragile solar cells. Compared to a conventional module assembling process no high temperature for soldering and no string transport is necessary. We think that the conventional soldering process is still more reliable compared to conductive adhesives or low melting solder alloys. The aim of this work is to combine the reliability of the conventional module assembling and the smoothness of the MMA processes. In previous work we showed, that a * e-mail:[email protected]

reliable interconnection of solar cells by laser soldering on the lamination foil is feasible [12]. We named this concept On Laminate Laser Soldering. Based on this concept we build up a prototype machine for interconnecting back contact solar cells. We focus on rear side contact solar cells connected by a bone shaped interconnector. Machines for interconnecting such solar cells are already available on the market, but these machines do not yet take advantage of the back contact solar cell concept in module production. 2. The ATLAS prototype machine We built a robot based prototype solar module assembling machine that we named ATLAS (On Laminate Laser Soldering). The ATLAS system was specified by ISFH, designed and built by Reis Robotics, and finally tested and optimised by ISFH and Stiebel Eltron. The machine is capable of interconnecting any back contact solar cells by bone shaped interconnectors, i.e. the above mentioned IBC solar cells. Figure 1 shows a 3D-overview of the ATLAS prototype system. The system is build up with two standard robots of Reis Robotics. To save costs the ATLAS prototype system is equipped with only one 6-axis handling robot for cross connector, solar cell and interconnector handling. The handling robot automatically changes the tools for the cross connectors, the cells, and the interconnectors. The two robots were chosen by costs aspects and not for high handling speed. The diode laser head used for the On Laminate Laser Soldering process is attached to a linear robot. The details of the infrared diode laser and the process-head are published in Ref. [12]. A vision system identifies the parts at the handling system, checks the quality and measures their position. The robot uses a lay up position that is calculated from the position measured by the vision system. The ATLAS prototype assembles modules configured in matrixes of 6 x 10 or 4 x 9 solar cells. For optimisation purposes a string of 1 x 10 solar cells was also programmed. The current size of the solar cells is 12.5 x 12.5 cm2, in square or semi square format. The final solar cell size of 15.6 x 15.6 cm² is intended but not jet implemented.

With a special teaching technique for the handling robot we achieve a high precision for placing the centre of the solar cells and the connector bones over the whole module size.

Figure 1: 3D view of the ATLAS prototype. 1 Handling robot, 2 Linear robot for movement of the laser head, 3 lay up table, 4 part magazine, 5 vision system, 6 handling tool magazine, 7 flux dispenser. 3. Module assembly In Figure 2 the concept of the module assembly is shown. It goes through the following stages: a)

The front glass and the first lamination sheet are laid down manually onto a workbench. In a production machine, this step will be done using a standard handling system.

b)

The handling robot picks the cross connectors from a magazine and puts it down at both ends of the module. Accurate positioning of the cross connectors is guaranteed by the handling and the vision system.

c)

While depositing the cross connectors onto the lamination sheet, the robot carrying a diode laser processing head heats up some points on the cross connector by the laser beam. This process fixes the cross connectors to the lamination sheet by slightly melting the sheet underneath the cross connector.

d)

A solar cell matrix is placed onto of the lamination sheet by the handling robot. The thin and fragile solar cells are taken from a stack. Accurate positioning of the solar cells on the lamination sheet is again realized by the handling and the vision system.

e)

The handling system picks the interconnector from the magazine.

f)

The soldering lugs of the interconnector are fluxed with a dispenser unit. The vision system enables accurate positioning of the connector bones between the solar cells.

g)

The laser robot solders the soldering lugs of the connector bones to the solar cells and at the end of the strings to the cross connector.

h)

Having soldered all connector bones to the solar cells, the second lamination sheet and the back sheet are placed manually onto the module. In production, this will be done by a foil handling system. Now, the module is ready for a standard lamination process.

In our experiments we use connector bones that we cut by a laser from a copper ribbon that is coated with the solder Sn96.5Ag3.5. This alloy has a melting point of around 217°C. The cross connectors are from standard production coated with the same solder alloy. The tin plating is approximately 20-25 µm thick. For testing the module assembling we used back contact solar cells that were bought from Sun Power™. As lamination foil we use foils from various materials such as normal cure ethylen venyl acetate (nc-EVA), fast cure-EVA (fc-EVA), ultra fast cure-EVA (ufc-EVA), thermoplastic polyurethan (TPU), and Polyvinylbutyral (PVB). .

Figure 2: Assembling concept of the ATLAS system. All module components are placed onto the lamination foil and then soldered by a diode laser.

4. Results The melting point at 217°C of the Sn96.5Ag3.5 is no problem for the On Laminate Laser Soldering process. The solar cells are slightly surface-fused onto the lamination foil by the laser soldering process, but the lamination sheet is not damaged. Figure 3 depicts typical on-laminate solder joints before and after lamination for ufc-EVA. The fusing protects the solar cells from displacement on the lamination sheet during the further production. After the lamination process, the surfacefused parts of the lamination sheet are not visible any more. All lamination foils show this result. If necessary, the solar cells fused to the lamination sheet are still removable without damaging the cell or the foil. This may be necessary for the repair of a not-yet laminated string.

We perform pull tests on these solder joints according to Norm DIN EN 50461:2006. The forces required for pulling an interconnector from the cell is always so high (>10 N) that it is the cell that breaks and not the interconnection. It is thus impossible to pull the interconnector from the cell. Solder joints between interconnector and cross connector, which looks like the joints shown in Figure 4, always have pull forces in excess of 20 N, which is the larges force that we can apply with our measurement system and which is sufficient for stability.

a)

b)

c) Figure 4: High quality solder joints: a) Top view of interconnector lug on cross connector, b) bottom view of interconnector lug on cross connector, c) top view of interconnector lug on a solar cell. The slightly molten area around the connection indicates a reliable joint.

Figure 3: Photograph through the module glass after laser soldering and before lamination (top) and after lamination (bottom). Even if the laminate (ufc-EVA) underneath the solder joints melts no damage of the foil is visible after lamination. The slight surface-fusion that occurs during soldering of the strings is probably advantageous in a production line, because after the soldering process the strings must not move on the laminate. Pre-fixation of the strings onto the laminate in a separate process is not necessary The ATLAS prototype solders one soldering lug of the interconnector in 0.3 s to a solar cell contact pad with a laser power of about 200 W. For one connector bone with 6 soldering lugs, the prototype machine requires 1.8 s for soldering and about 3.5 s for the movement of the laser processing head. Hence it takes 5.3 s to fix one interconnector between two cells. The heat capacity and the heat conductance of a cross connector is higher compared to a solar cell. Therefore the soldering of an interconnector lug to a cross connector takes 0.4 s. In summary it is the movements of the laser and the handling robots that limit the assembling speed of our prototype system. Figure 4 shows typical solder joints after an on laminate laser soldering process. Up to now the reproducibility of such solder joints is not good enough for implementation into a module production. The main remaining challenge is to assure a well defined contact pressure during the soldering process for all soldering lugs of an interconnector.

We built a complete solar module with the ATLAS system. It shows a good electric performance when compared to a commercial module made from similar cells by standard technique. Both modules contain solar cells from SunPower. The ATLAS-configured module consists of four strings with nine solar cells and the commercial module consists of four strings with eight solar cells. Hence we have to give the electrical parameters of the module per cell for comparison. Table I shows that both solar modules have nearly identical quality. Table I: Comparison of an ATLAS-configured and a commercial solar module.

Module parameter

ATLAS 4x9 cells

Commercial 4x8 cells

Isc [A]

5.6

5.5

FF [%]

77.5

77.0

η per cell area [%]

19.4

19.0

Uoc per cell [V]

0.661

0.659

Power per cell [W]

2.86

2.80

5. Conclusion We built the ATLAS-prototype for automatic assembling of photovoltaic modules that contain back-contacted solar cells. The ATLAS system applies laser soldering of solar cells that are lying on the lamination foil. For a production line we estimate that an assembling speed of two seconds per cell is possible when two handling

systems for the solar cell lay-up, two handling systems for the interconnector lay-up and three lasers are being used. The ATLAS-system fully eliminates conventional string handling. Further optimisation of the ATLAS prototype is under way. 6. Acknowledgements We thank the company Dilas Diodenlaser GmbH for their support concerning the diode laser system and the company Vitronic GmbH for equipping the ATLAS system with the vision system. 7. References [1] W. Jooss. K. Blaschek. R. Toelle. T. M. Bruton, P. Fath, i. Bucher, 17% back contact buried contact solar cells, Proc. 16th European Photovoltaic Solar Energy Conference, Glasgow,1-5 May, 2000, p. 1124-1127 [2] A. M. Gabor, D. L. Hutton, J. I. Hanoka, Monolithic modules incorporating ribbon silicon solar cells with wrap around contacts, Proc. 16th European Photovoltaic Solar Energy Conference, Glasgow, 1-5 May, 2000, p. 2083-2086 [3] B. T. Cavicchi, N. Mardesich, S. M. Bunyan, Large area wraparound cell development, Proc. 17th IEEE Photovoltaic Specialists Conference, 1984, pp. 128-133 [4] H. Haverkamp, H. Knauss, E. Rueland, P. Fath, W. Jooss, M. Klenk, C. Marckmann, L. Weber, H. Nussbaumer, H. Burkhardt, Advancements in the development of back contact cell manufacturing processes, Proc. 19th European Photovoltaic Solar Energy Conference, 7-11 June 2004, Paris, France, p. 967-969 [5] D. D. Ceuster, P. Cousins, D. Rose, D. Vicente, P. Tipones and W. Mulligan, Low cost, high volume production of >22% efficiency silicon solar cells, Proc. 22nd European Photovoltaic Solar Energy Conference, 3.-7. Sept. 2007, Milano, Italy, p. 816-819 [6] D. M. Huljić, T. Zerres, A. Mohr, K. v. Maydell, K. Petter, J. W. Müller, H. Feist, N.-P. Harder, P. Engelhart, T. Brendemühl, R. Grischke, R. Meyer, R. Brendel, F. Granek, A. Grohe, M. Hermle, O. Schultz, S. W. Glunz, Development of a 21% back-contact monocrystalline silicon solar cell for large-scale production, Proc. 21st European Photovoltaic Solar Energy Conference, 4-8 Sept. 2006, Dresden, Germany, p. 765-768 [7] P. Engelhart, A. Teppe, A. Merkle, R. Grischke, R. Meyer, N.-P. Harder, R. Brendel, The rise-ewt solar cell – a new approach towards simple high efficiency silicon solar cells, Proc. 15th International Photovoltaic Science & Engineering Conference, 2005, Shanghai China, p. 802-803 [8] J. H. Bultman, A. W. Weeber, M. W. Brieko, J. Hoonstra, J. A. Dijkstra, A. C. Tip, F. M. Schuurmans, Pin up module: a design for higher efficiency, easy module manufacturing and attractive appearance, Proc. 16th European Photovoltaic Solar Energy Conference, Glasgow, 1-5 May, 2000, p. 1210-1213 [9] P. Hacke, B. Murphy,D. Meakin, J. Dominguez, J. Jaramillo, M. Yamasaki, J. Jee, Busbarless emitter wrap-

through solar cells and moduls, Proc. 33rd IEEE Photovoltaic Specialists Conference, 11-16 May, 2008, California, USA, in press [10] J. M. Gee, S. E. Garrett and W. Simplifyed module assembly using crystalline-silicon solar cells, Proc. Photovoltaic Specialists Conference, Sept. Anaheim, CA, 1997, pp. 1085-1088

P. Morgan, back-contact 26th IEEE 30- Oct. 3.,

[11] M. Späth, P.C. de Jong, I. J. Bennett, T. P. Visser, J. Bakker, A novel module assembly line using back contact solar cells, Proc. 33rd IEEE Photovoltaic Specialists Conference, San Diego, USA, May 11-15, 2008, in press [12] M. Gast, M. Köntges, and R. Brendel, Lead free laser soldering for a new module manufacturing concept, Progress in Photovoltaics: Research and Applications 16, p. 151-157 (2008)