Autori del lavoro candidato: Stefano Zarini, Paolo Magnoni, Barbara Previtali, Nicola Pedrocchi, Lorenzo Molinari Tosatti
SINTESI CONTENENTE UNA BREVE DESCRIZIONE DEL LAVORO SVOLTO E DEI RISULTATI OTTENUTI: *please refer to the hard copy version for pictures identification Additive Manufacturing processes (AM) allows to make objects from 3D model depositing some material, usually layer upon layer, as opposed to subtractive manufacturing methodologies. This new approach has been widely defined the third industrial revolution because of the new prospective in the realization of complex objects which are revolutionizing the design and realization phases. According to the deposited material and the joining technology, different results in terms of deposition rate, roughness and chemical deposition may be achieved. Considering metals and laser as joining technology, two methods are available: Powder Bed Fusion (PBF), Figure 1a* and Direct Metal Deposition (DMD), Figure 1b*. In the first case, some powder is firstly layered (black cylinder in Figure 1a*) on a moving piston (green T in Figure 1a*) and later melted with laser (red line in Figure 1a*) moved through some rotating mirror (green plate in Figure 1a*). Once this process is finished, the pis ton moves toward the bottom and the process is repeated. In this way, layer after layer, the component is built. This technique gives the unique possibility to obtain small medium part characterized by complex shapes such as the one shown in Figure 1c*. On the contrary, in DMD process, metal powder is sprayed and immediately melted. Differently from the previous case, the deposition head is moved by a Cartesian or anthropomorphic robot (Figure 1b*). This technique, coupled with the advantage of an anthropomorphic robot to reach in different points in the space in different configurations, allows to obtain larger components if compared with PBF systems (Figure 1d*). Moreover, in DMD, it is possible to vary the chemical composition of the powder during the deposition obtaining multi graded functional structures characterized by variable mechanical proprieties within the component. This gives the unique opportunity to optimize material performances according to the specific point of the component and needs of the designer. These techniques of material joining belongs to the category of thermal processes. For this reason, monitoring thermal emission of the molten pool provides crucial information about the physics of the process that once understood allows to improve material deposition and thus quality of the final component. It was proved by scientific studies [1-4] that the local temperature of the molten pool strongly influences the deposition quality. For the realization of large scale components with DMD using anthropomorphic robot (scale dimension of about 1000 mm), deposition rate are in the order of 100 cm3/h. It is thus evident that the time scale for their realization is in the order of tens of hours. Therefore, variation of the deposition quality and thus temperature of the molten pool may lead to the generation of scrap causing drastic increase of resource, energy consumption and thus production costs. For this reason, much emphasis has been payed from research and industrial institutes on the monitoring of the molten pool temperature. In order to obtain reliable data from the thermal emission of the molten pool in DMD with anthropomorphic robots, numerous are the difficulties. One of these is the discrepancy between the dimension of the molten pool (~ 1-3 mm) and the overall component size (up to 1000 mm). Being the position and the geometry of the molten pool so important and so small, little deviation in their measurement leads to drastic drop of quality. For this reason application of static monitoring equipment such cameras in a fixed point are not able to provide reliable information about the variation of the molten pool emission in such a large component (for example, in some point of the space the molten pool can not be observed and thus lack of information is expected). This is why moving monitoring devices are much preferred to static ones in real industrial application. This is accomplished connecting the monitoring device to the robot arm providing two possible point of view: lateral (ancillary lenses in Figure 2*) and/or coaxial (CCD camera in Figure 2*). However, using an anthropomorphic robot, differently from a Cartesian system, allows to reach the same point in the space in different ways. Therefore, possible uncertainties related to the location of the molten pool due to vibration of the robot arm and bad allignment of the instrument could lead to errors in data acquisition. The solution to this problem is found in the architecture of the laser source. Once the laser is generated, it is delivered through some optical path. As the laser light can go in one direction (red line in Figure 3*), the thermal emission emitted by the molten pool (blue line in Figure 3*) is allowed to travel backward exploiting the same optical path. Here, w ith opportune settings, it is possible to place the monitoring device. With this configuration, the monitoring equipment is: • Far away from the process, thus in a safe condition (damages due to external contact between the environment and the equipment are almost zero); • The position of the molten pool is defined without uncertainties: at the center of the cladding head; • The volume of the head is reduced allowing greater motion freedom to the robot and thus allowing the realization of more complex geometries; This research activity focuses on the design and realization of a monitoring device able to detect the thermal radiation emitted by the molten pool exploiting the aforementioned architecture. After the design and realization phases (component selection through modelling and experiments), the device is tested in with real industrial equipment to check its ability to identify defects during the deposition process. Then, the use of the thermal radiation emitted by the molten pool is used to continuously correct some deposition parameters in order to improve the deposition quality. Finally, a user friendly software interface is created allowing external people to exploit advantages of monitoring and controlling thermal emission of the molten pool. At the same time, part of the research activity (developed in ITIA – CNR) focuses on a specific CAD/CAM software for Industrial Robots in Additive Manufacturing processes. The target is to allow the user to simply set all the process parameters and generate complex parts starting from a CAD file. This software can be used for DMD processes but it has also been tested in ITIA – CNR with other processes and materials (large scale Additive Manufacturing with plastic materials using industrial extruders) Moreover, such ad-hoc CAD/CAM software enables other possible features on the system: • Robot motion planning with the addition of External axes and positioners. This can generate visible benefits on the built part because of the increased system dexterity (minor robot wrist displacements are possible, with consequent reduced inaccuracies introduced by robot dynamics and system miscalibration) • Accurate on-line measurement of robot actual position and velocity. This is possible thanks to advanced options recently released by robot manufacturers: “robot open controllers” (i.e. ABB EGM). Using these measures it will be possible to adjust some process parameters (i.e. modulate laser power according to real robot velocity) or to evaluate system accuracy errors between the robot measures and the data collected by external tracking systems. BIBLIOGRAPHY 1. Tang L, Landers RG (2010) Melt Pool Temperature Control for Laser Metal Deposition Processes Part I: Online Temperature Control. Journal of Manufacturing Science and Engineering 132:011010-011010 2. Salehi D, Brandt M (2006) Melt pool temperature control using LabVIEW in Nd:YAG laser blown powder cladding process. The International Journal of Advanced Manufacturing Technology 29:273-278 3. Hua T, Jing C, Xin L, Fengying Z, Weidong H (2008) Research on molten pool temperature in the process of laser rapid forming. J Mater Process Technol 198:454-462 4. Hu D, Kovacevic R (2003) Sensing, modeling and control for laser-based additive manufacturing. Int J Mach Tools Manuf 43:51-60 5. Bi G, Gasser A, Wissenbach K, Drenker A, Poprawe R (2006) Identification and qualification of temperature signal for monitoring and control in laser cladding. Optics and Lasers in Engineering 44:1348-1359 6. B. Siciliano, L. Sciavicco, L. Villani, G. Oriolo (2009) Robotics: Modelling, Planning and Control 7. ABB, Application Manual – ABB Controller Software IRC5
Premio Sapio 