Abstract
Prosumer (producing consumer)-based desktop additive manufacturing has been enabled by the recent radical reduction in 3-D printer capital costs created by the open-source release of the self-replicating rapid prototype (RepRap). To continue this success, there have been some efforts to improve reliability, which are either too expensive or lacked automation. A promising method to improve reliability is to use computer vision, although the success rates are still too low for widespread use. To overcome these challenges an open source low-cost reliable real-time optimal monitoring platform for 3-D printing from double cameras is presented here. This error detection system is implemented with low-cost web cameras and covers 360 degrees around the printed object from three different perspectives. The algorithm is developed in Python and run on a Raspberry Pi3 mini-computer to reduce costs. For 3-D printing monitoring in three different perspectives, the systems are tested with four different 3-D object geometries for normal operation and failure modes. This system is tested with two different techniques in the image pre-processing step: SIFT and RANSAC rescale and rectification, and non-rescale and rectification. The error calculations were determined from the horizontal and vertical magnitude methods of 3-D reconstruction images. The non-rescale and rectification technique successfully detects the normal printing and failure state for all models with 100% accuracy, which is better than the single camera set up only. The computation time of the non-rescale and rectification technique is two times faster than the SIFT and RANSAC rescale and rectification technique.
Highlights
Prosumer based additive manufacturing has been enabled by the recent radical reduction in 3-D printer capital costs [1] created by the open-source release of the self-replicating rapid prototyper (RepRap) [2,3,4]
To monitor errors during FFF-based 3-D printing, an open source low-cost reliable real-time optimal monitoring platform for FFF-based 3-D printing from double cameras is presented here. This error detection system is implemented with low-cost web cameras and extended from the basic approaches dis-cussed above for 360 degrees around the printed object from three different perspectives by extending the algorithm using the Scale Invariant Feature Transform (SIFT) [61] and the Random Sample Consensus (RANSAC) [62] models previously described [63]
This paper described an open-source low-cost reliable real-time monitoring platform for FFF-based 3-D printing based on a double cameras system for three perspectives around 360 degrees
Summary
Prosumer (producing consumer) based additive manufacturing has been enabled by the recent radical reduction in 3-D printer capital costs [1] created by the open-source release of the self-replicating rapid prototyper (RepRap) [2,3,4]. Open source desktop 3-D printers have been applied to create high-value items in a wide range of fields including: rapid prototyping [10,11], distributed manufacturing [12,13], education [14,15,16], sustainable technology [17,18,19], scientific tools [20,21,22,23], microfluidics [24,25] Despite this success, these low-cost 3-D printers still suffer from a litany of printing challenges related to building up a part from thermoplastic one layer at time from a flat print bed including warping, elephant foot (thicker part touching the print bed), bed adhesion (prints peeling off of the bed during print), distortion due to shrinking, skewed prints/shifted layers, layer misalignment, clogged nozzles, or snapped filament [10,24,26]. There is an acute need for a low-cost real-time error detection system for prosumer-grade 3-D printers
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.