Research And Application Of High Precision Vision Alignment Technology On Full-automatic Solder Paste Printer

As solder paste printing is the first step of SMT technology, its printing quality plays an essential role on the quality of the whole products. The surface mount electronic components inevitably develop toward miniaturization, high integration and light-thin direction, as electronic products develop toward the direction of portable, compact and high reliability. The size of component is becoming smaller and smaller, which makes the printing accuracy of solder paste printer have to become increasingly demanding. In this study, we take on the solder paste printer as the research object and combine with the actual production. In order to improve the accuracy of the solder paste printer, we developed a set of vision system for the solder paste printer alignment platform, then proposed a new matching positioning algorithm for the alignment markers and constructed correction model for solder paste printer's alignment platform(1)We developed a set of vision system for the solder paste printer. We focued on researching and developing a FPGA industrial smart camera which can achieve the functions of single-frame / continuous acquisition, software / hardware trigger,image processing and can output 30-megapixel clear image by the velocity of 60 frames per second, and also can meet the application requirements of solder paste printer. According to the various actual performance technical indexes requirement, we had customed processing the visual processor, and had written the program frame of the visual processing software in the visual system.(2)We proposed a new generalized hough transform(GHT) matching algorithm, which was based on local gray sequence feature, for the mark-hole positioning problem of the circuit board and the SMT stencil in the process of Solder Paste Printer alignment. By using the relative size of the gray value as the feature of the edge point, the lookup table is constructed and then vote to locate the object. We elaborated the algorithm development process detailed.(3)Correction model of solder paste printer's alignment platform was constructed. We constructed the structural model for platform and established four coordinate systems which used for representing the positional relationship among the coordinate systems. In order to clarify the relationships among the coordinate systems, we transformed the coordinate systems. Then established the UVW platform's rotate model, carried out kinematic analysis of the platform, derived kinematic formulas for the platform, and eventually got the rectification calculation formula for the platform. At the same time, we analyzed the alignment system for the platform.(4)The accuracy of the the local feature matching algorithm based on GHT is tested, and the rubustness of the proposed matching algorithm is verified by comparing rthe test results. We completed the accuracy tests on the Solder Paste Printer produced by the Dongguan GKG Company. The experimental results show that using the visual system, matching algorithm and correction model in the paper improved the printing accuracy of the whole Solder Paste Printer from ± 0.025 mm to ± 0.02 mm and has an important application value for the development of solder paste printer industry.