In recent years, the application of 3D printing technology in foundry processes has become increasingly widespread, particularly for producing complex sand castings. This technology allows for the creation of intricate sand molds and cores without the need for traditional patterns or core boxes, enabling one-piece molding and simplifying operations. However, as we have observed in our production experience, 3D printed sand molds often exhibit high-temperature strength and poor collapsibility, leading to a significant increase in crack defects in sand castings. These cracks, which can manifest as hot tears or cold cracks, result in scrap parts and reduced efficiency. In this article, I will analyze the mechanism behind crack formation in sand castings due to 3D printed molds and propose a control method based on optimizing sand core design to enhance collapsibility. Throughout, I will emphasize the importance of addressing these issues for high-quality sand castings.
The fundamental problem stems from the inherent properties of 3D printed sand molds. Unlike conventional resin-bonded sand molds, 3D printed molds are fabricated layer by layer using binders that cure to form a rigid structure. This process often results in sand castings with molds that have excessive high-temperature strength and limited溃散性 (collapsibility). Collapsibility refers to the ability of a sand mold to yield or disintegrate as the casting contracts during solidification and cooling. When collapsibility is poor, the mold restricts the free contraction of sand castings, inducing internal stresses. If these stresses exceed the material’s strength at elevated temperatures, hot cracks form; if they exceed strength at lower temperatures, cold cracks occur. Mathematically, the stress ($\sigma$) induced in sand castings due to hindered contraction can be expressed as:
$$ \sigma = E \cdot \alpha \cdot \Delta T \cdot f(C) $$
where $E$ is the modulus of elasticity of the casting material, $\alpha$ is the coefficient of thermal expansion, $\Delta T$ is the temperature drop during contraction, and $f(C)$ is a function representing the constraint factor imposed by the mold’s collapsibility. For sand castings, $f(C)$ approaches zero when collapsibility is ideal, but in 3D printed molds, it can be significantly high due to their structural integrity.
To quantify the collapsibility issue, we conducted experiments measuring the溃散层 (collapsed layer) thickness of various sand molds. The溃散层 is the depth from the mold surface that disintegrates upon exposure to molten metal, allowing for退让性 (yieldability). Table 1 compares typical collapsibility parameters between traditional resin sand molds and 3D printed sand molds used for sand castings.
| Mold Type | High-Temperature Strength (MPa) | Collapsed Layer Thickness (mm) | Typical Application in Sand Castings |
|---|---|---|---|
| Traditional Resin Sand Mold | 1.0 – 2.0 | 10 – 20 | General-purpose sand castings with moderate complexity |
| 3D Printed Sand Mold | 3.0 – 6.0 | 2 – 5 | Complex sand castings requiring integral molding |
As shown, 3D printed sand molds have high-temperature strengths 2–3 times that of traditional molds, but their collapsed layer thickness is only about 10–25% of traditional molds. This poor collapsibility directly increases the constraint factor $f(C)$ in the stress equation, raising the risk of cracks in sand castings. In our observations, sand castings with thin sections or high structural integrity, such as machine tool beds or engine blocks, are particularly susceptible. The cracks often appear at stress concentration points like sharp corners or transitions in wall thickness, as the mold’s resistance prevents uniform contraction.
To address this, we developed an optimized design for 3D printed sand cores aimed at improving collapsibility for sand castings. The core idea is to create a hollow structure that reduces the overall mass and rigidity of the mold, allowing it to yield more easily during casting contraction. Our design consists of three key parts: the casting structure section, the hollow section, and the central support section, forming a frame-like “回” shape. This approach balances the need for mold strength during handling and pouring with enhanced退让性 during cooling.
The casting structure section is the part that directly forms the geometry of the sand castings. Its thickness must be sufficient to withstand the metallostatic pressure and erosion from molten metal, yet not so thick as to hinder collapsibility. Based on our trials, we recommend a thickness range of 20–50 mm, which exceeds the typical collapsed layer thickness of 3D printed molds (2–5 mm) while providing adequate strength. The hollow section is critical for improving collapsibility. In this region, the core is designed to be empty during printing, filled only with loose, unbonded sand. After printing, we open cleaning holes to remove part of this loose sand, creating a void that reduces internal resistance. The hollow section thickness should be at least 5 mm to ensure effective退让性. The central support section connects along the axis of handling (e.g., lifting direction) and is linked to the outer contour via tensile ribs, forming a框架结构 (frame structure) that maintains overall core integrity during use.
We formulated a mathematical model to guide the design parameters. The退让性 (yieldability, $Y$) of the hollow core can be approximated as:
$$ Y = \frac{V_h}{V_t} \cdot \frac{1}{S_m} $$
where $V_h$ is the volume of the hollow section, $V_t$ is the total core volume, and $S_m$ is the high-temperature strength of the mold material. For optimal performance in sand castings, we aim to maximize $Y$ while ensuring structural stability. Table 2 summarizes the design parameters and their recommended values for producing crack-free sand castings.
| Parameter | Symbol | Recommended Value | Rationale |
|---|---|---|---|
| Casting Structure Thickness | $t_c$ | 20–50 mm | Balances strength and collapsibility for sand castings |
| Hollow Section Thickness | $t_h$ | ≥5 mm | Ensures sufficient void volume for退让性 |
| Central Support Rib Width | $w_r$ | 10–20 mm | Provides frame stability without excessive rigidity |
| Hollow Volume Ratio | $V_h/V_t$ | 0.3–0.5 | Maximizes yieldability while maintaining core integrity |
In practice, the design process involves calculating these parameters based on the specific geometry of the sand castings. For instance, for a large机床件 (machine tool bed) casting, we might set $t_c = 30$ mm, $t_h = 10$ mm, and $w_r = 15$ mm, resulting in $V_h/V_t \approx 0.4$. This design ensures that during the contraction of sand castings, the hollow section allows the outer shell to deform or fracture locally, providing the necessary退让空间 (yield space) for stress relief.
To validate this method, we conducted production trials on a critical component: a machine tool bed (referred to as Component A) that had previously exhibited a 100% crack defect rate when using conventional 3D printed sand cores. We selected 10 sand cores for this component and optimized 4 of them using the hollow design approach. The remaining 6 cores served as controls with the original solid design. All other process parameters—such as printing settings, core assembly, melting, and pouring—were kept identical to ensure a fair comparison. The molten iron was poured at 1,350°C, and the sand castings were allowed to cool in the mold for 24 hours before shakeout.
After cooling, we inspected the sand castings for cracks using dye penetrant testing and dimensional checks. The results were striking: the 4 sand castings produced with hollow cores showed no visible cracks, while all 6 control sand castings exhibited cracks similar to those previously observed. We repeated the trial with two consecutive production runs, and the hollow core design consistently yielded crack-free sand castings. This confirms that the optimized design effectively mitigates crack formation by improving mold collapsibility. Table 3 presents a summary of the trial results, highlighting the impact on sand castings quality.
| Core Design | Number of Sand Castings Produced | Crack Defect Rate | Observations on Sand Castings Quality |
|---|---|---|---|
| Original Solid Design | 6 | 100% | Cracks appeared at stress concentration points; scrap parts |
| Optimized Hollow Design | 4 | 0% | No cracks detected; dimensional accuracy within tolerance |
| Optimized Hollow Design (Repeat Runs) | 2 | 0% | Consistent crack-free performance in sand castings |
The success of this method can be further analyzed through stress calculations. Using the earlier stress equation, we estimate that for sand castings with hollow cores, the constraint factor $f(C)$ is reduced due to increased退让性. Assuming an ideal case where the hollow section allows near-free contraction, $f(C)$ approaches zero, leading to minimal stress. In contrast, for solid cores, $f(C)$ is high, causing stress to exceed the fracture strength of the casting material. The fracture strength ($\sigma_f$) of gray iron at high temperatures can be modeled as:
$$ \sigma_f = \sigma_0 \cdot e^{-k(T – T_0)} $$
where $\sigma_0$ is the strength at room temperature, $k$ is a material constant, $T$ is the temperature, and $T_0$ is a reference temperature. For typical sand castings made of gray iron, $\sigma_0 \approx 250$ MPa and $k \approx 0.01$ K⁻¹. At a contraction temperature drop of $\Delta T = 500$ K, the induced stress in hindered sand castings can surpass $\sigma_f$, leading to cracks. Our hollow design reduces this risk by lowering $f(C)$.
Moreover, we considered the economic and operational aspects. The hollow design does not significantly increase printing time or material cost for sand castings, as the reduced material in the hollow section offsets the additional support structures. In fact, by preventing crack defects, it reduces scrap rates and improves overall productivity for sand castings production. We have since implemented this design across all 3D printed sand cores in our foundry, and after continuous production of over 100 sand castings, no crack issues have recurred. This demonstrates the robustness of the solution for various types of sand castings, from engine blocks to pump housings.
In conclusion, crack defects in sand castings caused by 3D printed sand molds are primarily due to poor collapsibility from high-temperature strength. By optimizing the sand core design to include a hollow structure with a框架支撑 (frame support), we can significantly enhance退让性, allowing sand castings to contract freely and minimizing internal stresses. Our production trials validate that this method effectively eliminates cracks, providing a reliable control strategy for high-quality sand castings. Future work could focus on refining the design parameters using finite element analysis to simulate stress distributions, or exploring alternative binder systems for 3D printing that offer better collapsibility. Nonetheless, the hollow design approach represents a practical and proven solution for the foundry industry, ensuring that 3D printing technology can be leveraged to produce complex sand castings without compromising integrity. As we continue to advance in additive manufacturing, such innovations will be crucial for meeting the growing demand for precision sand castings in automotive, aerospace, and machinery sectors.