The Study Of Processing Condition Optimazation In High Accuracy Gear Project

During the development of new polymer material for high accuracy gear, we usually mold out the gear to validate the accuracy level of new material. For high accuracy gear, the material property and the processing condition are two most important criteria for the accuracy level. But how to find out the ?optimized? processing condition is the key for the val idation. During the long-term study, we found there are several challenges. The first, since the gear is for high accuracy application, even minor variation of the processing condition will affect the accuracy level of the gear. The second, each processing condition has it?s own processing window, It?s had to found out the best level and meanwhile, since every processing condition may have affect to each other, so it?s also very difficult to get the best combination for every processing condition. The third, currently people try to find out best accuracy level by personal expertise and continually trial and error, which takes long time and hard to make sure can find the best level. Sometime even gave the wrong result which misleading the material development direction.In order to solve the above issues, we try to use DOE methodology. DOE is the abbreviation of Design of Experience, which is a methodology to develop the relationship of multi-factors and response. DOE plays an important role in the quality control process nowadays and is widely used to improve product quality and optimize the processing condition. The DOE methodology includes several different methods, like full factorial design, fractional factorial design, screening DOE, RSM etc. We compared the difference of each method and decided to use RSM as the tool for this study. By using the professional software named Design Expert, we designed the study of 25 runs by choosing 7 different processing conditions with 3 different levels for each. By measured the gear accuracy level and analysis the result, we find out the significant processing conditions. Meanwhile, we also fit the model and got the formula, which show the relationship of significant factors and gear accuracy level. We also had the validation trial and the result showed improvement of gear accuracy. The study showed the new material is good for accuracy gear application and gave the right direction for further material development. Also, with the study of all DOE methods, we developed different working procedures to solve similar problems in the future. These methods approved to be a good method to solve problem of multi-factors with multi-levels.