Adaptive slicing method and application technology in binder jetting technique

It is one of the most important features of additive manufacturing that it can produce complex parts that cannot be produced by conventional methods and that it lightens parts by methods such as topology optimization. In addition, its use in the industry is increasing, thanks to its ability to relax the supply chain with on-site production and its ability to produce with all kinds of materials. According to many economic reports, the binder jetting technique, one of the additive manufacturing methods investigated within this thesis's scope, is predicted to be the fastest-growing additive manufacturing method in the next decade. Although additive manufacturing methods have many advantages, they can be slower in production speed compared to traditional manufacturing methods. Therefore, this study aims to increase the manufacturing speed of the binder jetting method. Adaptive slicing in binder jetting has been proposed to increase manufacturing speed. The variable binder amount algorithm (VBAA) has been developed to use adaptive slicing efficiently in binder jetting. VBAA was tested with parameter data based on experience and literature in preliminary studies, and successful results were obtained. However, optimization of layer thickness and saturation ratio is required in VBAA. For this purpose, layer thickness and saturation ratio optimization in adaptive slicing have been performed using the Taguchi method. According to the Taguchi experimental design, 27 samples were produced with 9 different designs, each with 3 copies. The dimensions of the raw samples were measured, and then the samples were sintered at 1500°C for 2 hours. After sintering, surface roughness, and density tests were performed. Based on the results of these tests, a layer thickness of 180-250 µm and a saturation ratio of 50% was determined. Then, a separate test sample was designed to apply adaptive slicing. This test sample was produced in 3 different types, which are adaptive (180 µm-250 µm), thin layer (180 µm), and thick layer (250 µm) with the determined parameters. The roughness values of the adaptive and thin-layered samples were similar and better than the thick-layered samples. Similar roughness was obtained by using 12.31% fewer layers in the adaptive sample compared to the thin-layered sample. Therefore, the methods used have been proven to be successful. In addition, image processing was investigated for easy and inexpensive surface roughness measurements. Consistent results were obtained with image processing for the data obtained from the roughness measurement device.