Polymer injection molding parameters optimization into computational analysis using design of experiments methodologies
2018-36-0093
09/03/2018
- Features
- Event
- Content
- In the current scenario of the global automotive industry, it is chased an increasingly better fit and finish between the parts which leads to a good quality perception from the customer point of view, thus, once the warpage is one of the most common dimensional failures usually seen in injection molded parts, it becomes needful to take a closer look on this failure mode still during early vehicle design phases seeking for a feasible assembly, with desirable visual quality and meeting their structural roles. However, the set of injection process variables is still used to be taken as an empirical study and it is commonly adjusted after many injection runs (trial and error) creating needless trials and reworks costs. At this theme, this subjected work gets use of tools from Design of Experiments (DOE) methodologies in order to reach enhancement on polymers injection molding parameters in the interface of numerical computing simulations seeking to mitigate warpage failures. Rheological data are gathered from the computational analysis, then, it is performed statistical inferences aiming to estimate the effects of each process parameter and it is defined which ones are indeed contributing to the failure. Thus, a suitable regression model is created and validated, then, it is found the optimum values of this model through the response surface method. Therefore, by the end of this study it is recommended and tested optimum values for the manufacturing variables with the goal to meet the required surface profile tolerance as well as all further tolerances specified which ensures the warpage will be into an acceptable range for the part use and it avoids useless mold reworks.
- Pages
- 8
- Citation
- França,, D., "Polymer injection molding parameters optimization into computational analysis using design of experiments methodologies," SAE Technical Paper 2018-36-0093, 2018, https://doi.org/10.4271/2018-36-0093.