FBE- Isı Akışkan Lisansüstü Programı - Yüksek Lisans

Bu koleksiyon için kalıcı URI

http://hdl.handle.net/11527/52

Gözat

Konu "Additive manufacturing" ile FBE- Isı Akışkan Lisansüstü Programı - Yüksek Lisans'a göz atma
  • Öge
    Numerical investigation of natural convection in dmlm process
    (Institute of Science And Technology, 2020-06-15) Özada, Berk ; Çadırcı, Sertaç ; 503171108 ; Head and Fluids ; Isı Akışkan
    The importance of additive manufacturing is increasing day by day and competing with conventional manufacturing methods. Unlike the material removal or casting process, additive manufacturing is premature technology and struggling with some issues. Especially after became a focus of the industry and implicitly the academic studies, additive manufacturing is growing exponentially and physics behind the process is being revealed. Even though aviation, automotive and healthcare looks like the pioneers of the technology, usage of the technology becomes popular among almost all industries. Additive manufacturing or can be named as 3-dimensional (3D) manufacturing consists of numerous sub-methods that are based on the same principles even there might be some differences. In short, CAD file converted into an STL file which becomes readable by the additive manufacturing machine, and part is generated layer by layer until the final geometry is achieved. Additive manufacturing mainly divided into two subcategories which are using materials and power supply which melts the powder. Although fused deposition modeling (FDM) is common currently, metal and ceramic based additive manufacturing importance is increasing significantly. Enables to complex designs and high strength/weight benefits, interests of aviation and automotive industry increases in particular. Metal-based additive manufacturing can be segregated based on the power supply. Electron discharge and laser are the main technologies that are competing in the industry. DMLM obtains a high amount of energy from the laser whereas EDM gets from electron discharge. Both methods have their advantages and drawbacks which need to be considered while selecting the process. The part is printed under either vacuum or inert gas environment which prevents the oxidation at the high-temperature levels. Generally, the oxygen level inside the chamber should be decreased below 0.1% to begin the process. Besides preventing the oxidation, an inert gas is responsible to sweep spatters which can be generated with the interaction of the laser and powder. As the spatters may stick on the melted joint or diverts the laser light, process efficiency can drop significantly in the absence of inert gas for DMLM process. For that purpose, Argon or Nitrogen is preferred which is not reacting with the metals. The thin powder layer is melted with the existence of the laser and weld to the previously solidified layers. In each step, the build plate moves down and allows the recoater to spread a new powder layer. The part is removed from the powder pool once the process finishes. One of the drawbacks of the DMLM process is to have high costs. Even though additive manufacturing is promising to use for the highly complex and xx rapid prototyping, because of the high machine cost which can reach up to 75% of the total cost, additive manufacturing stays behind the conventional methods at the mass production. It revels with the experimental studies that mechanical properties of DMLM parts are below the forged parts whereas higher than the cast parts. Porosity, lack of fusion, residual stress, and rough surfaces are the parameters that have a significant impact on the mechanical properties. Because of the aforementioned defects, numerous studies in the literature investigate the process parameters. Laser power, scan speed, and scan strategies are the main focus areas to predict imperfections on the part. Energy density is the parameter that combines those parameters. High energy density brings about porosity because of keyhole formation and high residual stresses along the part since the thermal gradients increases parallel to laser energy. The other way around, low energy densities lead to obtaining balling and lack of fusion. In order to predict and avoid defects, process parameters of DMLM process are intensely being studied. Melt pool sizes and temperature within the parts are should be predicted during the process to control the overall part properties. As the physics behind the process is highly complex, each research makes some reasonable assumptions to simplifies the analyzes. Two methods which come forward are the analyze the domain as solid or fluid once the temperatures exceed the liquidus temperature. Even though melt pool dimensions are on the macro scale and return to solid-state within milliseconds, as the thermal gradients are extremely high, natural convection plays significant roles. This study investigates the effect of Marangoni and buoyancy forces during the DMLM process. For that purpose, two separate models is generated which one of that takes flow equations into account whereas others does not. Nine process parameters are selected based on the selected material which is Ti-6Al-4V. Ti-6Al-4V or Ti6-4 is one of the common materials among the aviation and healthcare industries which the advantage of high strength/weight ration. Powder material thermomechanical properties significantly vary from the bulk material properties. However, as the conventional manufacturing data are used in case of DMLM powder material curve is missing, Ti6-4 is one of the preferred materials for the literature. Marangoni and buoyancy are the forces acting on to the molten particles. Marangoni is a surface force that exerts on the free surface between inert gas and melts pool whereas buoyancy force is body force. Under laminar, incompressible assumptions, the impact of Marangoni and buoyant forces are investigated on melt pool dimensions and temperatures. Since the surface temperatures reached to 3500K, those two forces played a significant role in the high temperatures. Analytical results are compared and validated with the results which are obtained from the literature. Although there are several experimental studies for each parameter, maximum process parameter similarity between the CFD and experimental studies are considered during the experimental data selection. It is revealed that the importance of the flow equations becomes significant with the increase of energy densities. Even though melt pool dimensions are on the microscale, maximum velocity among the melt pool reached 1.84 m/s locally. Low energy density applications bring about reasonable results with solid modeling where the accuracy of the methods scatter with an increase of energy density. Especially error at the high energy density applications significantly increases with the solid modeling approach. However, results indicate that melt pool depth sizes captured accurately with both xxi approaches whereas error percentage of melt pool temperature and width predictions are significantly increasing at the high energy densities without taken flow equations into account. As the temperature gradients increases, natural convection within the melt pool plays a significant role in the surface temperature and width measurements. With the help of the Marangoni force, flow is directed along the perpendicular direction of the laser movement which distributes the heat along with the plate.