Binocular vision measurement system for geometric error of 3D printers at high temperature

The accuracy of material extrusion-based 3D printers is greatly affected by the high temperature in the chamber due to the thermal-induced deformation of components. However, most existing measurement equipment cannot be applied to high-temperature environments, which hinders the corresponding error measurement. To address this issue, a geometric error detection system and identification algorithm based on binocular vision are proposed. Firstly, a corner detection algorithm and a ray-intersection binocular model are used to identify the three-dimensional displacement of the target. Secondly, an error separation and identification algorithm is proposed to identify 21 position-dependent geometric errors. Error measurement experiments are conducted on a 3D printer at room temperature and high temperature, respectively. The experimental results at room temperature are verified using a double-ball bar. Finally, an error compensation experiment is conducted to verify the effectiveness of error identification, which also shows the contribution of error motions of linear axes on the printing accuracy.