In this paper, a simple and accurate method called 3D scanning method was adopted。 The measuring system called naviSCAN3D was used as shown in Fig。 5, which was a combination of a portable coordinate measuring machine, Metronor DUO, with a Breuckmann stereoSCAN3D white light scanner system。 While the Metronor DUO

Fig。 6 Deformation pattern obtained by 3D scanning

was able to cover an extensive scanning volume, the stereoSCAN3D was equipped to scan areas of up to one square meter per shot at high level of speed and accuracy。 The consolidation of the inpidual shots was carried out seamlessly。 During the entire scanning sequence, the system monitored the position of both scanner and measuring objects and identified possible interfering movements。 This technique ensured accurate and exact measurement results。 After scanning, the deformation pattern of sample could be gotten automatically as shown in Fig。 6。 Meanwhile, we chose 9 measurement points and recorded their distortion。 We  tested 8 simples and listed the results in Fig。   7。

Fig。 7 Deformation pattern obtained by 3D scanning

4。 Finite Element Analysis

4。1 Analysis approach

Residual stress was introduced into the plastic part produced by the solidification and cooling process of the plastic part。 Moldflow was used to estimate the residual stress in a molded product。 But, it was a lack of detailed structural analysis on how the residual stress affect the final shape and performance of the product。 Therefore, Abaqus was employed for this purpose, and the Abaqus Interface for Moldflow provided for the transfer of stress results from a Moldflow analysis to an Abaqus analysis。 But it could not predict the residual stress of insert part accurately。 To overcome this defect, the stress of metal insert was studied with the aid of elastic-plastic and thermal-mechanical coupled FEM software, Deform 3D。 The thermal stress was generated and changed from the beginning of polymer filling process to the end of product cooling down to the room temperature。 Therefore, the process chain analysis for the distortion prediction was carried out as Fig。 8。 The procedure for performing a distortion analysis on a molded product began with a simulation of injection molding process using Moldflow where the magnesium alloy was set as an insert part。 The Moldflow analysis results included descriptions of the material properties, geometry data and the distribution of residual stress in the solidified part。 The Abaqus Interface for Moldflow was then used to translate these data into a format that could be used by Abaqus。 These data that contained mesh information, residual stress results, and material propersities of the plastic were then used in subsequent Abaqus analysis for distortion and the effects of residual stresses。12 At the same time, the transient temperature distribution and the thermal stress were simulated by Deform 3D, using user routine codes。 The thermal stress simulated during the period of packing and cooling was approximate to the thermal stress generated actually。 And then the data after simulation exported and converted to Abaqus format data by a FORTRAN program coded by ourselves。 After assembling the geometries of polymer and metal-insert and mapping the residual stresses, the Abaqus input file was obtained。 Using tie constraint and two material models,   deformation

after ejection was obtained by Abaqus  analysis。

4。2 Modeling and simulation

For the injection molding, the model was used as shown in Fig。 3。 Boundary and initial conditions were taken from process data required during the production as Table 1。 The 3D element (C3D4) was employed

Fig。 8 Analysis process for prediction of distortion