With the rapid increase in utilization of the cheap and user friendly Fused Filament Fabrication, FFF 3D printer, a deeper knowledge about the technique is needful. The frame restricting the 3D printers for prototyping purposes is fading and a new phase of endless application possibilities is emerging. To bridge the gap in possible applications from prototypes to real products it is key to know and improve the factors affecting durability. With over a hundred settings and parameters to tweak the FFF 3D printing process there are a lot of opportunities, opportunities to optimize for durability.The tensile properties of some of the most used FFF 3D printing materials together with a few nylon based materials are examined, which are popular in engineering applications. The materials tested are ranging from rigid to flexible, rubber like materials. The most common failure scenario of a FFF 3D printed product is layer bonding failure. The factors affecting layer bonding performance are studied.The measurements are carried out using tensile testing equipment at Blekinge Institute of Technology. All tested specimens are manufactured at Creative Tools AB Halmstad with the FFF 3D printers Flashforge Dreamer and Makerbot Replicator 2X.The tensile strength of 3D printed PLA is found to be 51 MPa. PET has a tensile strength of 40 MPa and ABS 34 MPa. Stress-strain behavior of the materials shows that ABS is slightly softer than PLA and PET are slightly softer than ABS. PLA being the hardest material in the test. ISO 527-2 tensile testing standard is used but the tests diverge from the standard in several ways. The measurement data presented in this study can be very useful to guide the design engineer to choose the most durable plastic for the unique application.Five basic 3D printing settings are evaluated for layer bonding performance, by measuring the load capacity of a PLA specimen loaded transversally relative to the layers. Four of the settings show to possibly affect the layer bond’s load capacity by 50 % or more individually.The results of this study are presented in graphs, diagrams and pictures. These may help the 3D printer user to tweak basic settings to increase layer bonding performance and ultimately the durability of the product significantly.

An obstacle to the wide spread use of additive manufacturing (AM) is the difficulty in estimating the effects of process parameters on the mechanical properties of the manufactured part. The complex relationship between the geometry, parameters and mechanical properties makes it impractical to derive an analytical relationship and calls for the use of a numerical model. An approach to formulate a numerical model in developed in this thesis. The AM technique focused in this thesis is fused filament fabrication (FFF). A numerical model is developed by recreating FFF build process in a simulation environment. Machine instructions generated by a slicer to build a part is used to create a numerical model. The model acts as a basis to determine the effects of process parameters on the stiffness and the strength of a part. Determining the stiffness of the part is done by calculating the response of the model to a uniformly distributed load. The strength of the part depends on it's thermal history. The developed numerical model serves as a basis to implement models describing the relation between thermal history and strength. The developed model is suited to optimize FFF parameters as it encompass effects of all FFF parameters. A genetic algorithm is used to optimize the FFF parameters for minimum weight with a minimum stiffness constraint. / Ett hinder för att additiv tillverkning (AT), eller ”3D-printing”, ska få ett bredare genomslag är svårigheten att uppskatta effekterna av processparametrar på den tillverkade produktens mekaniska prestanda. Det komplexa förhållandet mellan geometri och processparametrar gör det opraktiskt och komplicerat att härleda analytiska uttryck för att förutsäga de mekaniska egenskaperna. Alternativet är att istället använda numeriska modeller. Huvudsyftet med denna avhandling har därför varit att utveckla en numerisk modell som kan användas för att förutsäga de mekaniska egenskaperna för detaljer tillverkade genom AT. AT-tekniken som avses är inriktad på Fused Filament Fabrication (FFF). En numerisk modell har utvecklats genom att återskapa FFF-byggprocessen i en simuleringsmiljö. Instruktioner (skriven i GCode) som används för att bygga en detalj genom FFF har här översatts till en numerisk FE-modell. Modellen används sen för att bestämma effekterna av processparametrar på styvheten och styrkan hos den tillverkade detaljen. I detta arbete har strukturstyvheten hos olika detaljer beräknats genom att utvärdera modellens svar för jämnt fördelade belastningsfall. Styrkan, vilket är starkt beroende på den tillverkade detaljens termiska historia, har inte utvärderats. Den utvecklade numeriska modellen kan dock fungera som underlag för implementering av modeller som beskriver relationen mellan termisk historia och styrka. Den utvecklade modellen är anpassad för optimering av FFF-parametrar då den omfattar effekterna av alla FFF-parametrar. En genetisk algoritm har använts i detta arbete för att optimera parametrarna med avseende på vikt för en given strukturstyvhet.