Publication | Closed Access
A parametric study of part distortions in fused deposition modelling using three-dimensional finite element analysis
302
Citations
15
References
2008
Year
EngineeringMechanical EngineeringFdm Part DistortionsParametric StudyResidual StressComputer-aided DesignFused DepositionStructural OptimizationComputational MechanicsAdvanced ManufacturingPart DistortionsResidual StressesLayer ThicknessNumerical SimulationMaterials ScienceSolid MechanicsManufacturing EngineeringMetal Forming3D PrintingCivil EngineeringDirected Energy DepositionStructural MechanicsSolid ModelingMechanics Of Materials
A finite element analysis (FEA) model was previously developed by the current authors to simulate the fused deposition modelling (FDM) process. The model considered coupled thermal and mechanical phenomena and incorporated an element activation function to mimic the additive nature of FDM. Due to repetitive heating and cooling in the FDM process, residual stresses accumulate in a part during deposition. In this study, an FEA model is used to evaluate the distortions of a part. A parametric study, with three factors and three levels, is performed to evaluate the effects of the deposition parameters on residual stresses and part distortions. Prototype models with larger sizes are fabricated, measured, and compared with the simulations. The simulation results are summarized as follows. First, the scan speed is the most significant factor affecting part distortions, followed by the layer thickness. Second, the road width alone is insignificant. However, the interaction between the road width and the layer thickness is significant as well. Third, there are other two-way and three-way interactions that are of secondary significance. In general, residual stresses in FDM parts increase with the layer thickness. Residual stresses also increase with the road width, but to a lesser extent, although largely affected by the layer thickness. The FDM part distortions from the experiment show a similar trend as concluded in the simulations, but without quantitative correlation.
| Year | Citations | |
|---|---|---|
Page 1
Page 1