Quality Prediction in Jet Printing Using Neural Networks

Abstract: Surface mount   technology   is   widely   used   in   the   manufacturing   of   commercial  electronics, and  the  demands  on  the  machines  increase  as  the  complexity of  the electronics increases and the size of the components decreases. Mycronic is a company that focuses on addressing those demands with their high-technology jet printing and pick-and-place machines. This master's thesis has been performed at Mycronic and has focused on the MY700 jet printer. Due to unknown factors, the quality of the ejected  solder paste droplets from the machine can vary over time. It was therefore of interest to monitor variables of the MY700 in order to gain more knowledge about the cause of the varying quality, and also to be able to detect substantial changes in deposit quality. In this project, the temperature has been measured at three key locations  on the ej ector as well as the current going through the piezoelectric actuator. This data was fed to a neural network in order to make  quality predictions with respect to the diameter of the solder paste deposits. Different combinations of sensor data were used to evaluate how the different sensors affected the performance of the neural network. Thereby, a better understanding of how big an  impact the different variables had on the quality of the deposits could be achieved.  The results indicate that the current was more significant than the temperature for making quality predictions. Using only the temperature data, the neural network was not able to accurately predict quality deviations, whereas with the piezo current data or both of them  combined,  better predictions could be made. The current data also significantly improved the performance of the neural network when printing jobs with varying diameter were  used. The conclusion is that none of the  three  temperature sensors  significantly  improved  the  performance, and there were no considerable differences between them, while the current did improve it.