Effect of the Cooling Capability Difference Between Additively Manufactured Sand Molds on Shrinkage Defect in A356 Alloy Castings

Abstract

Three-dimensional sand mold printing is a novel manufacturing method that can produce complicated shapes based on a 3D CAD model. Various types of material systems can be employed to produce 3D-printed sand molds. The thermal and mechanical properties of these sand molds have been extensively studied, but few studies on the effect of the type of 3D-printed sand mold on the shrinkage porosity of the resultant castings have been reported. Shrinkage porosity is a concern that can limit the capability for advanced castings process using 3D-printed sand molds. In this study, aluminum cast alloy A356.0 was cast into typical 3D-printed silica sand molds and an 3D-printed mullite artificial sand mold. The results indicated that differences in cooling capability among these sand molds did not affect the hardness of the A356.0 castings but significantly influenced the location of shrinkage porosities. In particular, sound castings were not obtained from 3D-printed silica sand molds owing to their shrinkage porosity. Thus, a thermal solidification simulation was performed according to the cooling capability of the 3D-printed silica sand mold. The riser shape of the 3D-printed silica sand mold was modified based on the simulation result, and the sound A356.0 casting was made resultantly. This study demonstrated that the use of a riser shape according to the cooling capability of a 3D-printed sand mold is necessary to produce sound castings for the advance casting process using the 3D-printed sand mold.

Appendix 1

Boundary conditions and thermal properties used in the casting solidification simulation

The heat transfer equations in the casting solidification simulation were solved using finite difference method software. The casting temperature $T$ as a function of time $t$ was calculated using the following equation.

$$ \rho \frac{\partial H}{{\partial t}} = \nabla \cdot \left( {{\uplambda }\nabla T} \right) $$

(2)

$$ H = \mathop \smallint \limits_{0}^{T} C_{p} dT + L\left( {1 - f_{s} } \right) $$

(3)

where, $\lambda$, $C_{p}$, $\rho$, $f_{s}$, $H$, and $L$ are the thermal conductivity, specific heat, density, solid fraction, enthalpy, and latent heat, respectively. The heat flow ${\text{q}}$ at the boundaries was solved using the following equations.

$$ q = \left\{ {\begin{array}{*{20}c} {h\left( {T_{sur, cast} - T_{air} } \right)} \\ {h\left( {T_{sur, mold} - T_{air} } \right)} \\ {h\left( {T_{int, cast} - T_{int,mold} } \right) } \\ \end{array} } \right. $$

(4)

where $h$, $T_{sur, cast}$, $T_{sur, mold}$, $T_{air}$, $T_{int, cast}$, and $T_{int,mold}$ are the heat transfer coefficient (HTC), casting surface temperature, mold surface temperature, air temperature, casting temperature at the interface, and mold temperature at the interface, respectively. Table

Table 3 Thermal Properties of the A356.0 Casting and 3D-Printed Catalyst-Pre-mixed Silica Sand Mold Inputted in the Solidification Simulation

3, Figures

Figure 18

Temperature-dependent thermal conductivity, specific heat, and density of the A356.0 casting

18,

Figure 19

Solid fraction of the A356.0 casting

19, and

Figure 20

Initial and boundary conditions inputted in the solidification simulation (simulation model with the modified riser)

20 show the parameters inputted in the casting solidification simulation.