Study on size error transfer from wax pattern to casting

The best wax dipping process was adopted, and the sample after wax dipping was used as wax pattern. The ceramic shell was prepared by using silica sol as binder. Silica sol (W > 30%) and zircon powder (325 mesh) are used to prepare the coating for the surface layer. Zircon sand (80 ~ 120 mesh) is used for the surface layer, kaolinite clinker (220 mesh) is used for the transition layer, and kaolinite clinker (16 ~ 30 mesh) is used for the back layer. The number of coating layers was 7-8, and the shell was prepared. After the shell is fully dried, it is heated at 600 ℃ for 30min, then at 1000 ℃ for 30min, and finally at 1115 ℃ for preheating, 1600 ℃ for pouring, and 2kg / s for pouring. After the shell is fully cooled, the gate is cut off. The corresponding measuring points of the casting were measured and the average value of each position was calculated.

Among them, the dimensional accuracy of metal castings is the relative error of castings relative to wax patterns, which is calculated by the following formula:


ε 2 – relative error of casting size relative to wax pattern;

L3 – casting size, mm.

It is found in Fig. 1 that the relative error of the castings obtained in this shell making process is quite different from that of the wax pattern, and the overall performance is negative. The results show that the linear size reduction occurs during the solidification process and the expansion occurs during the preheating process. The relative error of the specimen size is negative under the combined action of the two factors. There is no obvious change in the X direction, but it changes in a certain range with the change of the test position, which indicates that the corresponding shrinkage of the alloy is basically the same when the X direction size is between 20 mm and 100 mm. The relative error of the average size in the X direction is 0.75%, and there is a small relative error in the Y direction at the test position (1) and (2), while in other test positions, the curve regresses to – 0.75%, which indicates that the Y direction has a small relative error The dimension relative error will change with the change of Z-direction dimension. Specifically, when Z-direction increases, Y-direction plane also gradually increases, and the shrinkage of corresponding metal castings increases, but the final Y-direction dimension relative error will approach – 0.75%. The variation trend of relative error in Z direction is the same as that in Y direction. Specifically, the relative error of metal casting is small at test position (1) and (2), which can be regarded as no shrinkage. At test position (3), the shrinkage rate of metal casting increases gradually with the increase of Z dimension. Especially when the Z dimension increases to 22mm, the relative error curve of Z dimension gradually approaches – 0.75%. To sum up, it is not difficult to find that when the wax mold size is less than 20 mm, the shrinkage of metal castings is not obvious, which can be regarded as no shrinkage. When the wax mold size is greater than 20 mm, the shrinkage of metal castings is obvious, with an average shrinkage rate of 0.75%.

It can be seen that the relative errors of X, y and Z dimensions are all negative, which indicates that there is a negative deviation in the casting process of the alloy, and the dimensions in all directions are reduced. Therefore, the transfer law from metal casting to wax mold can be obtained.

The law of X-direction transfer is as follows

The transfer law of Y direction is as follows

The transfer law of Z direction is as follows

Where, ε X3, ε Y3 and ε Z3 are relative errors of metal castings relative to x, y and Z average dimensions of wax pattern. In conclusion, through the rapid manufacturing of stepped linear shrinkage parts, it provides a theoretical basis for the size compensation of diffuser impeller in later rapid investment casting.