Research On Simulation Technology Of Mold Flow For High-density Module Packaging

With the rapid development of integrated circuit technology,the electronic market has put forward higher requirements for packaged devices in terms of density,size and integrated function,which makes the manufacturing process of packaging complicated,and some quality defects such as bubbles,sutures,warping,volume shrinkage and other problems often appear in the process of plastic encapsulation.The existence of these defects greatly reduces the reliability of the packaged devices.In addition,there are many process parameters affecting the quality of products in the process of injection molding.At the same time,due to the long production cycle and high cost of products,the traditional"trial and error"method of setting injection molding process parameters based on previous experience has been unable to meet the current demand of injection molding production.In this context,how to improve the production efficiency and the reliability of packaged devices need to be solved urgently.At the same time,there is insufficient research on the reliability of plastic-sealed devices in the humid environment in China.Therefore,this paper increases the research on the moisture diffusion and wet stress distribution of packaged devices in the humid environment.The main research achievements of this paper are as follows:1.According to the basic principle of injection molding of packaging devices and CAE technology,this paper uses Moldex3D software to establish the injection molding model of high-density module packaging and carries out simulation and analysis on it,and makes filling comparison with four different epoxy plastic sealing materials(EMC).It is found that the average viscosity of the material is controlled in the range of 10~2～10~4[g/(cm·s)],and the ratio of the thickness of the plastic seal body to the gap at the top of the device is controlled below 32.8,which can reduce the probability of filling holes.In addition,the optimal injection gate location can make up for the imbalance of melt flow in the mold cavity and improve the filling quality of the melt.These studies are very important to improve the reliability of injection molding.2.In order to study the reliability of the module package after injection molding in the humidity environment,this paper uses ANSYS finite element software to simulate the moisture diffusion distribution cloud map of the module package in different time periods in the humidity environment,and obtains the hygroscopic change curves of five reference points inside the package.After 20 hours,the growth rate of wet stress inside the package gradually decreases and tends to be saturated and stable,and the lamination probability inside the package can be reduced by the plastic sealing materials with low saturated humidity and large elastic modulus.At the same time,the reliability of the Si chip inside the package is evaluated by the stress-strength interference theory,and the reliability reaches 97.26%,so the overall reliability of the module package is predicted to be good.3.This paper uses Taguchi orthogonal experimental design method to design 27 groups of tests with six factors and three levels.According to the experiment,the mean of SNR is calculated,and the best combination of injection process parameters is found,which reduces the warping deformation by 14.5%and the volume shrinkage by 20.7%.In order to improve the accuracy of the prediction model,a neural network prediction model between six injection molding process parameters and two quality indexes was established,and the ant colony algorithm was used to optimize the BP neural network parameters to improve the convergence performance of the neural network,and the extreme value and corresponding process parameters were found through continuous iterative optimization.The simulation results show that the warp deformation is reduced by 19.74%and the volume shrinkage is reduced by 22.75%compared with the actual injection molding parameters.Therefore,compared with the orthogonal experimental design described above,the injection molding process parameters have been further optimized.