FEASIBILITY OF BUILDING AN OVERHANG STRUCTURE USING DIRECT METAL DEPOSITION

Proceedings of the 5th Annual ISC Research Symposium ISCRS 2011 April 7, 2011, Rolla, Missouri FEASIBILITY OF BUILDING AN OVERHANG STRUCTURE USING DIRECT METAL DEPOSITION Sriram Prabhu Department Of Manufacturing Engineering Missouri University of Science and Technology Rolla, Missouri, USA Todd E. Sparks Department of Mechanical and Aerospace Engineering Missouri University of Science and Technology Rolla, Missouri, USA Jianzhong Ruan Production Innovation and Engineering LLC Rolla, Missouri, USA Frank W. Liou Department of Mechanical and Aerospace Engineering Missouri University of Science and Technology Rolla, Missouri, USA ABSTRACT This paper summarizes a theoretical and an experimental approach to employ the Direct Metal Deposition (DMD) process to adjudge the feasibility of depositing overhang features. The DMD process is a process capable of directly manufacturing a fully dense metal part without the use of any intermediate steps. This process, comprised of a laser deposition setup coupled with a 5-axis CNC milling system, provides a capability to produce a finished part with superior surface qualities.. This paper discusses the issues and related tools in the feasibility of depositing an overhang structure using the hybrid deposition-machining process, including laser deposition process, system design and integration & process planning. A successful experimental work is also been carried out using one of the tools to justify the feasibility of the overhang structure. The preliminary model is constructed which resulted in a combination of 178.9 mm/min feed-rate and an angle of 34.9 degrees of laser incidence angle were determined to be the optimum process parameters in extension mode (4 axis) for maximizing the overhang distance. 1. INTRODUCTION A five-axis deposition process coupled with five-axis machining capability can simplify the obstacles for overhang deposition [1]. This paper summarizes the issues and related approaches in the research and development of the hybrid deposition-machining process for building overhanging structures. The use of layered manufacturing techniques to produce fully functional parts is extensively being researched for application [2]. Production of fully dense near net shaped parts has been made possible by using these technologies. The production of parts from difficult-to-machine materials like Titanium has also been made possible, without the need to machine excess materials by using additive manufacturing technologies like DMD. The advantage of using additive manufacturing technique for part production is that the usage of materials is limited to its occupied volume. The box volume is of no concern for these technologies. Hence, material as well as excessive post-processing waste can be saved. The scope of DMD is still limited as the production of overhang feature in part geometry for all cases is still unachievable. Thus it is quite difficult to build complex geometries having internal features, or parts having extreme overhanging features. Although it is possible to build overhangs, there are no methods that are feasible for all possible orientations and geometric structures. The present work comprises the tools of constructing overhanging features that will aid in depositing an overhang under no special circumstances. 2. DIRECT METAL DEPOSITION PROCESS In the Direct metal deposition (DMD) process, metal powder is directed on a substrate through a nozzle into a melt pool created by a laser. Thus, a melt pool is formed and material is deposited onto the substrate. The rate of at which material is being deposited and the thickness of the deposited layer can be adjusted according to the need and application. However, a functional speed limit has been established through the various research activities for a suitable deposition. It is quite difficult to accurately control the 1

material properties of the part and its physical dimension simultaneously. objects without which it would be impossible to generate the feature. The DMD process can be used for building parts directly from the CAD model (Additive Manufacturing), to clad a part to incorporate corrosion resistance, wear resistance and other surface enhancements, to repair a damaged part. 2.1 Support Structures Layered Manufacturing comprises of many processes but the scope of the current work is to concentrate only DMD processes which might require the need for external support structures in certain specific cases for the part building. By the term external support, it is generally meant about the additional columns that are required in the vertical downward direction from the current orientation to be built along with the required part in order to prevent some of its features from collapsing to the next available surface. Whenever a specific CAD design needs to be deposited, there arises the need for external support structures in three possible cases without which a particular feature of the part cannot be build. Technically, these three cases are [3] : Case 1: When the material on the previously deposited layer is too short to form as a base for the next layer to be deposited making the layer to be deposited to overhang by some amount. As shown in Fig. 1, conventionally speaking in DMD process, the layer 1 will be the first deposited layer and layer 2 will be deposited next. It can be clearly seen that layer 2 overhangs by some amount with respect to the layer 1. Fig. 2 Case 2 Case 3: Some structures, as shown in fig. 4, due to lack of proper support structures, face instability and result into incorrect / over deposition thus increasing post processing operations and time. Fig. 1 Case 1 Case 2: While depositing the layers, there sometimes arises situation when the feature of the part to be built floats, having no base/bonding neither sideways nor beneath the feature. Suppose the required part geometry has to be deposited is as shown in the Fig 2, during the deposition of this part, at some point of time, there will arise a need for supporting the floating Fig. 3 Case 3 The third case is somewhat trivial, since it can be detected readily in the process planning phase. The current research work is carried out taken into account the first 2 cases. This research work deals with the part geometries that can be built by DMD. Also, there is a need to emphasize the fact that this work is being carried out to eliminate the need for support structures and broaden the scope by increasing the complexity level of the structures that can be built using Hybrid DMD process. 3. POSSIBLE OVERHANG SOLUTIONS: After a thorough understanding of the process and the combinations of the process parameters to have a suitable deposit, the following solutions seem to make the deposition of overhang feature possible. 2

3.1 2-Axis Mode 3.1.1. Extension Mode: In this scenario the layers are deposited on the edge of the substrate. The first layer is deposited on the substrate to obtain some amount of overhang that would be visible as a bulge on the interface between the substrate and the deposited layer at the edge of the substrate. This bulge would then be as a support for the next deposited layer for the part of the layer that would be extending from the substrate. Thus the next layer is deposited exactly above the previously deposited layer, but this time the layer needs to be deposited a bit longer that the previous layer. This would again create a bulge at the interface between the 2 nd layer and the 1 st layer. Subsequent deposition in this fashion at an optimum set of process parameters would result in a possibility of forming an overhang that would be extended from the substrate. Figure 4 clearly indicates the above concept. 3.2 4-Axis Mode Fig. 5 Transverse Mode (2 Axis) 3.2.1 Extension mode (4 axis): Technically the deposition is done in the same manner as done in the Extension Mode -2 axis, except that the angle of incidence between the laser beam and the substrate never remains same. As shown in the fig. 6, once the first layer is deposited the substrate is then tilted at an angle where a suitable melt pool can be formed on the substrate/ previously deposited layer. Fig. 4 Extension Mode (2-Axis) 3.1.2. Transverse Mode: Here, the arrangement is done like the extension mode but the 1 st layer need not be deposited on the edge of the substrate. The process parameters are to be selected in such a manner that the deposited bead gives an extended welding angle w.r.t. to the substrate. The second layer is then deposited, without changing the angle of the substrate, on the previous layer but after some transverse shift. The subsequent layers are deposited with a certain amount of transverse shift w.r.t. previous layer. As shown in the fig. 5, an overhang can thus be built in the transverse mode. Fig. 6 Extension Mode (4 axis) 3

3.2.2 Transverse Mode (4-axis): The deposition is done in the same fashion as done in the 2-axis mode, the only difference being that the angle of the substrate never remains same with respect to the horizontal. As shown in the fig. 7, the first layer is deposited to form a weld bead angle with respect to the substrate, the substrate is then tilted at an angle where the normal of 1 st layer is in line with the center of the nozzle. In this scenario for the same number of layers deposited, the amount of overhang is expected to be larger than the one obtained in 2 axis mode. The main consideration in this situation is to deposit a layer with a low weld bead angle. material. This hybrid deposition system employs 5-axis positioning system which includes of 3 linear axis (X,Y,Z) and 2 rotary axis (A and B). The substrate and the powder used for this experimental investigation is 316L Stainless Steel. Figure 8 shows the LAMP (Laser Aided Manufacturing Process) system in process. Fig. 8 DMD system at LAMP lab. 4.2 Experimental Plan: 3.3 5- Axis Mode A genuine approach while using 5- axis is to deposit by changing the orientation of the deposited structure[4]. In this method, possible change in orientations is considered to deposit the overhang feature. But this method is largely limited by the built structure and part geometry as well as the possible orientations. 4. EXPERIMENTAL WORK: 4.1 Experimental set up: Fig. 7 Transverse Mode (4 Axis) A 5-axis CNC machining center, a 1.0 KW Diode laser, a powder feeder system, and a real time control system comprises the laser deposition system which has been used for this experimental investigation. For a deposition, the substrate is clamped on the fixture of the 5-axis CNC. The nozzle through which the laser beam is passed and metal powder flows is vertically mounted on a column fixed to the z-axis of the CNC machine. The laser focuses on a small area of the substrate thus creating a melt pool. The metal powder is delivered into the melt pool creating a weld bead of deposited After a thorough research through all possible tools, the Extension Mode - (5 Axis) was assumed to be the most qualified of all the tools. The objective of the experiment was to minimize the overhang angle to find a suitable angle for overhang deposition. So an experiment was designed were the following process parameters were selected: Process parameter Table Speed Laser incident angle Values 125, 237.5, 350 mm/min 0, 10, 22.5, 35 Degrees A 2-factor- 2 level factorial experiment with a center point and an additional point at 0 Degree of angle of incidence was carried out with 3 replications. The runs on 0 degrees were carried out to simultaneously test the eligibility of the Extension Mode (2-Axis). The laser power and the powder flow rate were kept constant at 1 kw and 16 g/min respectively. A 316L stainless steel plate with size 4 x 2 x.25 inches was procured. The runs were then randomized to obtain an unbiased result. 5. Results and Discussion Table 1 Variable Parameters Figure 9, shows the results of the height of the overhang obtained from the deposition in comparison to the feed rate. The plot clearly shows that the magnitude of the overhang height reduces with increasing feed-rate. It was also noted that these results remained constant at all the laser incidence angles. 4

to obtain a suitable weld bead angle which can then be effectively used for transverse Mode (2 and 5 Axis).This also throws light on the fact that a bead with an appropriate weld bead angle can be formed at lower feed rates which would be useful for constructing an overhang feature in Transverse Mode (2 and 5 axis) Fig 9: The plot of Height of the overhang and the feed rate Figure 10, compares the height of the overhang obtained with the laser incidence angle. Thus it can be said that the Height increases with the laser incidence angle Figure 12: A plot of Width of the deposited layer and the laser Incidence Angle From these figures a regression analysis was carried out using the GNU/R software. A Preliminary Model was then constructed in the form of Regression equations representing the bead width, bead height, and overhang thickness with respect to the travel speed and deposition angle system parameters. These three regression equations were then used in a linear programming optimization to find an optimal setting for producing overhang structures. Figure 10 A plot of height of the overhang and the laser incidence angle. Max:.309 0.00123S+0.00769A Sub. To. : 2.25<3.28-0.004S+0.000668A<2.75 0.75<1.729-0.00457S+0.00965A<1.25 0<A<35 125<S<350 The result obtained after maximizing the above equation usinf the MATLAB gave the result that following optimum set of process parameters: S (Feed Rate) = 178.8 g/min A (Laser incidence angle) =34.9 Z (Overhang thickness) =.358 mm Figure 11 A plot of the width of deposited layer and the Feed rate. Figure 11 & 12, shows that a wider bead resulted at lower table speed whereas the laser incidence angle does not make any significant change in the width of the deposited bead. A wider bead at the lower table speed is obtained because at lower table speed an effective melt pool is formed and at appropriate powder flow rate a wide bead can be formed. Thus there arises a need to check the possibility of obtaining process parameter 5

6. CONCLUSION A Preliminary Model is constructed through this research work which would guide the future research work. A total of 12 tracks were deposited, including the replicates, out of which the 7 th track with 22.5 Laser incident angle and table speed of 237.5 mm/min gave an exceptional overhang. The results of the multivariate regression analysis were used to maximize the equation of overhang height in terms of process variables. A combination of 178.9 mm/min feed-rate and an angle of 34.9 of laser incidence angle were determined to be the optimum process parameters for maximizing the overhang distance for the 5-axis extension mode case addressed in this project. The tracks deposited during the experiment were successful in producing an overhang of up to 0.621 mm that would form as a suitable base for the successive layers. [3] Allen, S., Dutta, D., 1995, Determination and evaluation of support structures in Layered Manufacturing, J. of Design and Manufacturing, 5,pp. 153-162. [4] Yang, Y., Fuh, J.Y.H., Loh, H.T., and Wong, Y.S., 2003, Multi-Orientational Deposition to Minimize Support in the Layered Manufacturing Process, Journal of Manufacturing Systems, 22 (2), pp.116-129. 7. FUTURE WORK Further experimental work is planned to optimize the process parameters further in this direction. The Preliminary Model thus generated will be aimed on developing further and verified. The capability to build extreme overhanging structures without any special considerations or limitations can be achieved with a thorough research on the deposition strategies. An effort is being made to develop en empirical model based on the relation between surface tension of the weld bead and the Temperature. The weld bead angle is also an important factor that needs to be considered. Further research work is being planned to develop more robust tools build overhanging structures without the use of support and more importantly unconditionally. ACKNOWLEDGMENTS This research was supported by the national science foundation grants iip-0822739 and iip-1046492. The support from Boeing phantom works, Product Innovation and Engineering, LLC, Missouri S&T Intelligent Systems Center, and the Missouri S&T Manufacturing Engineering Program, is also greatly appreciated. REFERENCES [1] Liou, F.W., Choi, J., Landers, R.G., Janardhan, V., Balakrishnan, S.N., Agarwal, S.,2001, Research and Development of Hybrid Rapid Manufacturing Process Twelfth Annual Solid Freeform Fabrication Symposium, Austin, Texas, August 6 8, pp. 138 145. [2] Liou, F.W., Slattery, K., Kinsella, M., Newkirk, J., Chou, H., Landers, R.G., 2006, Applications of Hybrid Manufacturing Process for Fabrication and Repair of Metallic Structures, SFF Symposium-Aug 2006. 6