Dynamic reverse compensation to increase the accuracy of the rapid prototyping system

Abstract Rapid prototyping (RP) is a fast and feasible technique for prototype construction. However, the phenomenon of volume shrinkage is unavoidable with any material or building method. The volume shrinkage and curl distortion are the main reason for the inaccuracy of the built prototype. The curl distortion changes with the laser scan path; the curl distortion increases with both scanning length and part height. Consequently, the error is serious for built prototypes with parts that have large height–length dimensions. Subsequently, more expensive equipment is used to improve the precision of the dimension and volume shrinkage on the market. Also, obtaining better processes parameters via trial and error in the RP process is expensive and inefficient. In order to improve dimension precision, and reduce the processing cost and the frequency of trial and error, this study first induces the concept of CAE into the processing of RP, which uses a dynamic finite element simulation code to simulate the photopolymerization process, to reduce the selection time for the processing parameters and obtain the distortion data. Second, dynamic reverse compensation employed to obtain a new CAD model, which is then loaded into a RP machine for practical prototype processing, to increase the accuracy of the process. Finally, to confirm this method and restriction in experimental equipment, stereolithography process and simple laser scanning path are chosen as examples for experimental comparison. The results of the simulation and experiment demonstrate the effectiveness of the proposed method. The proposed method cannot only reduce the equipment cost but also can simultaneously enhanced precision of the dimensions of the final parts. Besides, this method is believed to be applicable to other materials or build-up methods used in RP fabrication.