As for specimen selection, Wang Danni of Wuhan University of science and technology used finite element analysis and BP neural network to simulate the sintering process and used cylinder specimen for experimental verification; Yan Ran of Chongqing University of science and technology used response surface method to study the influence of interaction between process parameters on forming quality, and used rectangular specimen for data measurement and analysis [23]. The above research methods often need to measure a large number of data to support and verify. Due to pure manual measurement, not only the process is complicated, but also the requirements for the measurer in the process of data acquisition are relatively high. Different machines and equipment often have different process parameters, and the sintering model selected is too single. The first mock exam of sintering process is the result of sintering shrinkage, secondary sintering, Z axis surplus and powdered spheroidizing. The shrinkage rate of different models is different. Therefore, the single model can not accurately calculate the shrinkage of the model. In this paper, the traditional investment casting wax ladder linear test pieces were used as the test pieces to study the shrinkage of the test pieces at different steps. Fig. 1 is a wax model ladder type linear test piece.
The SLS prototype was treated with the optimal process parameters, and the residual wax drops and flow marks were removed after simple mold repair. Combined with the investment casting process, through adding gating system, tree assembly, slurry sticking, sand hanging, shell making, shell baking, shell preheating, alloy pouring and other processes, the corresponding metal castings are prepared after the castings are cooled fully, the shell is removed and the gating system is removed. The corresponding castings were prepared and the next step of the size error transfer law was analyzed. The size error of the specimen in the three steps of CAD model SLS prototype wax mold casting was calculated respectively.
The sls300 printer produced by Shaanxi Hengtong Intelligent Machine Co., Ltd. was used to make the corresponding prototype. The rapid prototyping equipment is shown in Figure 2.