Rapid prototyping (RP) is a set of technologies that permits building a physical model directly from its design by implementing a single automatic process using a 3D model of the object to be printed. RP systems can be based on different Additive Manufacturing (AM) technologies, such as a Fused Deposition Modeling (FDM) machine that works by extruding and melting together fused plastic filaments, drawing the boundaries and filling the model thin layer by thin layer. Low-cost FDM 3d printers do not work well automatically but require of a calibration phase because the best configuration settings in the slicing software are unknown, and the number of parameters values that needs to be manually defined is very large. The scientific literature reports many interesting articles on this topic, describing how the process can be improved by choosing the correct values of various parameters. Internet websites such RepRap.org discuss 3D printers and ppost detailed FAQ sections where users described improvements in 3D printing with simple methods but with great effort in terms of costs and time. Yet not all questions are answered. This paper would introduces: a) a new method for the analysis of the slicing software parameters that can be done with easy models; b) a second method for improving the effects of the parameters that shows a higher influence in the signal-to-noise ratio analysis.