Purpose
Additive manufacturing (AM) is a promising alternative to the conventional production methods (i.e., machining), providing the developers with great geometrical and topological freedom during the design and immediate prototyping customizability. However, frictional characteristics of the AM surfaces are yet to be fully explored, making the control and manufacturing of precise assembly manufactured mechanisms (i.e., robots) challenging. The purpose of this paper is to understand the tribological behavior of fused deposition modeling (FDM) manufactured surfaces and test the accuracy of existing mathematical models such as Amontons–Coulomb, Tabor–Bowden, and variations of Hertz Contact model against empirical data.
Design/methodology/approach
Conventional frictional models Amontons–Coulomb and Tabor–Bowden are developed for the parabolic surface topography of FDM surfaces using variations of Hertz contact models. Experiments are implemented to measure the friction between two flat FDM surfaces at different speeds, normal forces, and surface configuration, including the relative direction of printing stripes and sliding direction and the surface area. The global maximum measured force is considered as static friction, and the average of the local maxima during the stick-slip phase is assumed as kinematic friction. Spectral analysis has been used to inspect the relationship between the chaos of vertical wobbling versus sliding speed.
Findings
It is observed that the friction between the two FDM planes is linearly proportional to the normal force. However, in contrast to the viscous frictional model (i.e., Stribeck), the friction reduces asymptotically at higher speeds, which can be attributed to the transition from harmonic to normal chaotic vibrations. The phase shift is investigated through spectral analysis; dominant frequencies are presented at different pulling speeds, normal forces, and surface areas. It is hypothesized that higher speeds lead to smaller dwell-time, reducing creep and adhesive friction consequently. Furthermore, no monotonic relationship between surface area and friction force is observed.
Research limitations/implications
Due to the high number of experimental parameters, the research is implemented for a limited range of surface areas, which should be expanded in future research. Furthermore, the pulling position of the jaws is different from the sliding distance of the surfaces due to the compliance involved in the contact and the pulling cable. This issue could be alleviated using a non-contact position measurement method such as LASER or image processing. Another major issue of the experiments is the planar orientation of the pulling object with respect to the sliding direction and occasional swinging in the tangential plane.
Practical implications
Given the results of this study, one can predict the frictional behavior of FDM manufactured surfaces at different normal forces, sliding speeds, and surface configurations. This will help to have better predictive and model-based control algorithms for fully AM manufactured mechanisms and optimization of the assembly manufactured systems. By adjusting the clearances and printing direction, one can reduce or moderate the frictional forces to minimize stick-slip or optimize energy efficiency in FDM manufactured joints. Knowing the harmonic to chaotic phase shift at higher sliding speeds, one can apply certain speed control algorithms to sustain optimal mechanical performance.
Originality/value
In this study, theoretical tribological models are developed for the specific topography of the FDM manufactured surfaces. Experiments have been implemented for an extensive range of boundary conditions, including normal force, sliding speed, and contact configuration. Frictional behavior between flat square FDM surfaces is studied and measured using a Zwick tensile machine. Spectral analysis, auto-correlation, and other methods have been developed to study the oscillations during the stick-slip phase, finding local maxima (kinematic friction) and dominant periodicity of the friction force versus sliding distance. Precise static and kinematic frictional coefficients are provided for different contact configurations and sliding directions.