Flow directions in gas assisted injection molding when cavities of square flat plates and pipes are involved 2. Development of time-dependent flow model

Abstract

For such conditions that (H/R₀)² is replaced by ε (that is the order of 10^-1) and$\hat \theta ^2$ is the order of one, the rule of thumb for an approximated flow model was introduced in part 1 of the paper to show, in qualitative way, whether the resistance of the relatively thick cavity of two square plates might affect the gas direction in GAIM under the aforesaid geometry. Subsequently, various simulations were performed by using Moldflow (version of MPI 4.0) under the conditions that all dimensions of cavity of two square plates and pipes were fixed except for the diameters of pipes, and the results of simulation were compared with the results of a rule of thumb (RT1) containing the approximated flow model as well as those of another rule of thumb (RT2) without the resistance of the relatively thick cavity of two square plates. There were some exceptional cases where RT1 or RT2 were not consistent with the simulation results (i.e., flow directions). Thus such a developed model as time-dependent model was required to describe transient behavior of the interface between gas phase and resin phase instead of comparison of initial velocities in upper side and lower side of the configuration, which was proposed and utilized to compare with the results of Moldflow in this 2^nd part of the paper. The predictions of the developed flow model were so quite consistent with the results of simulation that the proposed time-dependent flow model may be referred to describe very well the transient behavior of the movement of the interface of gas and melt-resin in the cavities. In addition, a timedependent model was also established and was used to compare with the results of Moldflow when cavities of pipes and runners were involved in configuration. It is amazing that the proposed developed model was able to predict exactly the cross-over between the trajectories of interface of upper and lower side, and it is also surprising to describe the time dependent behavior so well that the result of the predictions by the developed model were quite consistent to the results of simulation by Moldflow.