A low-cost methodology for quality inspection of metal additive manufactured parts

Additive manufacturing of metallic components has been proved useful in several industrial aspects, such as in on-demand spare parts from a virtual library, resulting in a reduction of inventory costs. Before going into service, components must pass through strict inspection process, attesting its integrity; thus the first stage of this inspection process consists of visual inspection, executed by qualified professionals. However, it does not allow digital data recording for future assessments; it presents limited accuracy due to the traditional gauge kits, resulting in traveling expenses for installation and hindering training of other professionals. The uncertainty relative to the presence of discontinuities in this new manufacturing process discourages future investment. Therefore, the present study proposes a low-cost inspection methodology for Wire Arc Additive Manufacturing components, based on photogrammetry techniques. This measurement process generates a 3D model using a commercial DSLR camera and an open-source processing software. The resulting photogrammetry model is compared to another one obtained using a commercial laser scanning system. The average measurement deviation between the digitized models was 0.104 mm and the accuracy of the photogrammetry system was 0.144 mm (0.104 +/- 0.040 mm), which is adequate for the inspection process of the 115 mm diameter part. Laser scanning is recommended whenever assessment of surface waviness is required, since the geometrical deviation is below 1.440 mm (photogrammetry limit), but above 0.400 mm (laser scanning limit). In spite of that qualitative analysis can be performed with both systems, through 2D profiles, which is useful to understand the surface behavior. Color representation is found only in the photogrammetry model. This is fundamental for visual inspection because even small discontinuities that do not present large geometrical variations can be identified on the surface. Another purpose of this study is to stimulate other researchers and engineers to use remote visual inspection through 3D models and virtual reality. [GRAPHICS] .