Dependence of the parts dimensions’ accuracy on the internal structure of the layers when 3D printing using Fused Filament Fabrication technology

Abstract

This paper presents the results of the study of the obtained dimensions’ accuracy depending on the internal structure of the layers when manufacturing samples from the composite material ABS pro CCF using the Fused Filament Fabrication technology. The accuracy was estimated as the deviation of the actual obtained dimension from the nominal dimension of the sample in percent depending on the following factors: infill pattern, infill density and nominal dimension value. The experiments were conducted using the full factorial experiment method. Regression models were built using the PRIAM software. Analysis of the obtained model, in particular, the graphs of marginal regression equations and marginal response surfaces, showed that the structure of the infill can differently affect the accuracy of the obtained dimensions: for example, for larger dimension, the best result was obtained when using the “Grid” infill pattern. For a smaller dimension (10 mm), for different infill density options, different patterns showed different results: for a density of 80 %, the “Cubic Subdivision” infill pattern turned out to be the best; for a density of 60 %, the “Grid” infill pattern was the best. When using the “Grid” infill pattern, as the infill density increases from 60 % to 100 %, the response function behaves in the opposite way for small and medium dimensions: for a dimension of 10 mm, the response function moves away from the optimal value, and for a dimension of 150 mm, the response function, on the contrary, approaches the optimal value. The explanation of this phenomenon requires additional research, in particular with a larger number of levels of variation of the measured dimension, as well as taking into account the location of the measured dimension relative to the printer coordinate system during the build.