Mapping of deposition-stable and defect-free additive manufacturing via material extrusion from minimal experiments
- 120 Downloads
Despite significant advances and established advantages of material extrusion additive manufacturing, limitations still exist regarding ability to increase productivity and still maintain high shape fidelity and defect-free parts within the range of process parameter values that a 3D printer may support. In the present study, a predictive process-planning tool to balance productivity and quality of fabricated parts according to user’s requirements is developed. Taguchi design of experiments (DoE) method is employed in order to determine the process parameters that impact shape fidelity in terms of material deposition stability that results in surfaces of acceptable flatness and defect-free parts. The development of four well-documented fabrication defects is quantified by CAD-to-part 3D comparison. Then, analysis of variance (ANOVA) was performed to reveal, characteristically, that, even though reduced speed generally results in improved shape fidelity, defects arise when small layer thickness is also set. On the other hand, increased layer thickness may be coupled with moderate material deposition speed to improve process productivity without sacrificing part shape. A highly reliable artificial neural network (ANN) is constructed in an optimum way and trained on the previous experiments with the aim to predict shape fidelity across the entire factor level range. Hence, material flow rate which is a function of investigated factors is provided with direct feedback by the ANN regarding possible defects or deteriorated part shape. The method presented lays the basis for robust process optimization on any 3D printer.
KeywordsMaterial extrusion Fused filament fabrication Shape fidelity Taguchi method Artificial neural network Process optimization
Unable to display preview. Download preview PDF.
Charlotte Lassime from École Nationale d’Ingénieurs de Saint-Étienne (ENISE) is gratefully acknowledged for helping with measurements.
This study is financially supported by the National Technical University of Athens PhD stipends.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
- 1.ISO/ASTM 52900 (2015) Additive manufacturing—general principles—terminology. International Standard Organization (ISO), ASTM InternationalGoogle Scholar
- 10.Michaeli W (2003) Extrusion dies for plastics and rubber. doi: https://doi.org/10.3139/9783446401815
- 12.Venkataraman N, Rangarajan S, Matthewson MJ, Harper B, Safari A, Danforth SC, Wu G, Langrana N, Guceri S, Yardimci A (2000) Feedstock material property—process relationships in fused deposition of ceramics (FDC). Rapid Prototyp J 6:244–253. https://doi.org/10.1108/13552540010373344 CrossRefGoogle Scholar
- 14.Ajinjeru C, Kishore V, Liu P, Lindahl J, Hassen AA, Kunc V, Post B, Love L, Duty C (2018) Determination of melt processing conditions for high performance amorphous thermoplastics for large format additive manufacturing. Addit Manuf 21:125–132. https://doi.org/10.1016/j.addma.2018.03.004 CrossRefGoogle Scholar
- 16.Spoerk M, Gonzalez-Gutierrez J, Lichal C, Cajner H, Berger G, Schuschnigg S, Cardon L, Holzer C (2018) Optimisation of the adhesion of polypropylene-based materials during extrusion-based additive manufacturing. Polymers (Basel) doi: https://doi.org/10.3390/polym10050490
- 23.Brooks HL, Rennie AEW, Abram TN, et al (2012) Variable fused deposition modelling—analysis of benefits, concept design and tool path generation. Innov Dev Virtual Phys Prototyp - Proc 5th Int Conf Adv Res Rapid Prototyp 511–517. doi: https://doi.org/10.13140/2.1.2280.2887
- 28.Gu Q, Herakleous K, Poullis C (2014) 3DUNDERWORLD-SLS: an open-source structured-light scanning system for rapid geometry acquisition. 1–30Google Scholar
- 36.Dias FM, Antunes A, Vieira J, Mota AM (2004) Implenting the Levenberg-Marquardt algorithm on-line: a sliding window approach with early stopping. IFAC Proc Vol 37:49-54. https://doi.org/10.1016/S1474-6670(17)30849-2
- 38.Madić M, Madić MJ, Radovanović MR (2011) Optimal selection of ANN training and architectural parameters using Taguchi method: a case study. FME Trans 39:79–86Google Scholar
- 40.Hagan MT, Demuth HB, Beale MH, De Jesús O (2014) Neural network design. Martin HaganGoogle Scholar