Single print optimisation of fused filament fabrication parameters

Single print optimisation (SPO) of process parameters, for the improvement of dimensional accuracy of an additively manufactured (AM) printed part with fused filament fabrication (FFF), is presented. A test object is sliced with different slicer parameters, as determined by a design of experiments (DOE) run table. The resulting files are merged into a single G-code file, which is used to perform a single print run. This prints the objects sequentially in super layers, where a super layer consists of several normal layers of sequentially printed parts limited by the printer Z-axis clearance. The test object is designed to be measured with accessible, affordable and easy-to-use dimensional measurement instruments. The goal of the optimisation is to improve the deposition of a single track, since all FFF objects are built up from several extrusions called tracks in the printer XY plane. The track height and width are key quantifiable measurands. Retraction effectiveness is also considered, with a qualitative response variable called stringing length. Ordinary least squares (OLS) regression is used to fit linear models to the three response variables. The bed level, as a function of X and Y coordinates, is used to improve part height accuracy. Track width is modelled with liquefier temperature and feed speed, whilst stringing length is modelled as a function of temperature. The models are verified by selecting an operation point which maximises print speed and minimises a cost function for the track width.