In the development phase, new casting products are often characterized by low quantities and tight delivery schedules. Traditional mold-making for casting is time-consuming and costly. The rapid prototyping of sand cores using 3D printing casting overcomes these drawbacks of conventional methods. Our company recently developed a large sprocket hub, where the early stage of blank development was not suitable for investment in permanent molds. To shorten the research and development cycle and save mold fabrication costs, we adopted 3D printing casting technology for sand core preparation and trial production. This product is made of ductile iron QT500-7 and is applied in construction machinery. The product structure is complex, and the key areas must be free of shrinkage porosity, as well as defects such as sand holes, gas holes, cold shuts, and cracks. The product parameters for the sprocket hub are listed in Table 1, and the structural characteristics are described below.
| Parameter | Value |
|---|---|
| Material Grade | QT500-7 |
| Overall Dimensions (mm) | Φ800 × 541 |
| Weight (kg) | 295 |
| Maximum Wall Thickness (mm) | 50 |
| Minimum Wall Thickness (mm) | 10 |
| Main Body Wall Thickness (mm) | 20 |
The sprocket hub is a rotational body structure. According to the structural characteristics and quality requirements of the product, we selected a vertical pouring position with the large end facing downward and the small end facing upward. The thick flange area is located in the middle, which facilitates the placement of feeders on the flange for feeding and chill blocks at the bottom for chilling. For the product structure, this pouring position is also conducive to the structural uniformity of the product, especially suitable for rotational body structures like wheel hubs. The 3D printing casting approach allows us to design the core without constraints from mold withdrawal or parting lines.
3D Printing Casting Process Design
Parting Line Selection for 3D Printed Sand Cores
Although 3D printed sand cores are not limited by mold stripping, and the parting line can be set arbitrarily, we still need to consider the need for cleaning loose sand and applying coatings after printing. Therefore, to facilitate these operations, we set the parting line at the middle flange plane. This approach maximizes the cleaning opening, making it easier to remove loose sand and apply coating, avoiding issues such as incomplete cleaning leading to dimensional inaccuracies and sand inclusion during pouring. Furthermore, inadequate coating application can cause metal penetration, which complicates later cleaning. The 3D printing casting technology provides the flexibility to choose the core design that best suits the post-processing steps.
Process Parameters for 3D Printed Sand Cores
The 3D printed sand cores are made of furan resin self-hardening sand. The process parameters are selected according to the standard for 3D printing furan resin sand processes. The high dimensional accuracy of 3D printing casting allows for the use of tight clearances typically associated with metal molds. The selected process parameters are presented in Table 2.
| Parameter | Value | Selection Basis |
|---|---|---|
| Shrinkage Allowance | 1.0% | 3D printed sand cores use furan resin self-hardening sand with high strength, and the casting shrinks under restraint, hence a 1.0% shrinkage rate. |
| Machining Allowance (mm) | Top face: 7 mm; Others: 5 mm | Maximum product dimension is 800 mm. Machining allowance grade is G. The top face has a greater tendency for shrinkage and is prone to sand/slag inclusion defects, so a 7 mm allowance is used; other faces get 5 mm. |
| Clearances (mm) | Fitting surfaces: 0.3 mm; Non-fitting surfaces: 0.5 mm; Anti-chafing surfaces: 1.0 mm | High precision of 3D printing casting cores allows for 0.3 mm on fitting surfaces, 0.5 mm on non-fitting surfaces, and 1.0 mm on surfaces prone to chafing during core assembly. |
| Draft Angle | 0° | 3D printed cores are not constrained by mold stripping; no draft angle is required. |
Gating System Design for 3D Printing Casting
Given the hollow structure of the product, and to minimize the sand core volume and reduce the sand-to-metal ratio, the sprue was designed to be positioned at the center of the shaft hole. This arrangement in 3D printing casting maximizes the reduction of the core assembly volume, thereby decreasing resin usage and saving costs. The ingates were placed at the bottom of the casting, forming a bottom gating system. The choke was located at the bottom of the gating system, between the filter and the ingates. This system is a closed-open gating system. The initial closed section aids in slag trapping, while the subsequent open section ensures smooth filling. According to the choke area formula:
$$A_{\text{direct}} : A_{\text{choke}} : A_{\text{ingate}} = 28\ \text{cm}^2 : 21\ \text{cm}^2 : 30\ \text{cm}^2 = 1.3 : 1 : 1.4$$
After pouring, the ingates remain full, ensuring that the gating system fills completely. The molten metal fills the mold smoothly, which is beneficial for cavity venting and prevents splashing and gas entrapment that could lead to oxide slag defects. The ingates are flat and thin, with a total of four arranged in a cross pattern to ensure smooth filling and uniform temperature distribution. Due to the product structure leading to a high sprue, which causes high impact and flow velocity of the molten metal, we designed two levels of foam ceramic filters (100 mm × 100 mm × 22 mm, 10ppi) in the sprue. This structure filters the molten metal while buffering the impact, reducing flow velocity to avoid sand erosion and minimize turbulence, resulting in a more stable filling process. The pouring cup uses a simple funnel shape, effectively reducing molten metal consumption and improving process yield. This efficient gating system is a direct benefit of the design freedom offered by 3D printing casting.
Feeder and Chill Block Space Design in 3D Printing Casting
Initially, a preliminary numerical simulation was conducted without feeders or chills. The solidification mode of ductile iron is pasty solidification. However, due to the significant wall thickness variations in this product, the modulus ratios of the thick sections (top, middle flange, and bottom ring) to the adjacent thin sections were all greater than 2. The thin sections solidify rapidly, isolating the three thick sections into separate hot spots. This cuts off the feeding channels, preventing these sections from receiving molten metal from other parts and from sharing the feeding effect of graphitization expansion. Therefore, they need to be calculated independently. Since the top thick section is a non-critical area, with shrinkage porosity located centrally, and its size does not affect product performance, only riser vents were placed at the top for venting. The middle flange and bottom ring are critical areas requiring no shrinkage porosity. The feeder modulus and feeding volume were calculated based on the hot spot location and modulus.
For the middle flange section, the casting modulus is 1.8 cm, and the volume is 7,200 cm³. Assuming a liquid contraction of 4% for the casting, the required feeding volume for the flange area is:
$$V_{\text{feed, flange}} = 7,200\ \text{cm}^3 \times 0.04 = 288\ \text{cm}^3$$
The feeder modulus needs to be greater than 1.2 times the casting modulus. The TELE140R feeder is a high-exothermic sleeved feeder with a modulus of 2.3 cm and an effective feeding volume of 80 cm³. The modulus requirement is satisfied: $$M_{\text{feeder}} > 1.2 \times M_{\text{casting}}$$ The number of feeders required is:
$$n_{\text{flange}} = \frac{288\ \text{cm}^3}{80\ \text{cm}^3} = 3.6 \approx 4$$
Therefore, four TELE140R feeders were placed. These feeders have an exothermic neck with a diameter of only 23 mm, ensuring smooth feeding and easy removal during cleaning due to the small contact area with the casting.
For the bottom ring section, the casting modulus is 1.5 cm, and the volume is 4,300 cm³. The required feeding volume is:
$$V_{\text{feed, ring}} = 4,300\ \text{cm}^3 \times 0.04 = 172\ \text{cm}^3$$
A 7/10 insulation sleeve feeder with a modulus of 2.0 cm and an effective feeding volume of 100 cm³ was selected. The modulus condition is satisfied. The required number of feeders is:
$$n_{\text{ring}} = \frac{172\ \text{cm}^3}{100\ \text{cm}^3} = 1.7 \approx 2$$
Two 7/10 insulation sleeve feeders were placed. The modulus ratio for the casting, feeder neck, and feeder was designed as $$M_{\text{casting}} : M_{\text{neck}} : M_{\text{feeder}} = 1 : 1.1 : 1.2$$. The minimum cross-section of the feeder neck was calculated to be a rectangle of 45 mm × 25 mm, with an effective modulus of 1.65 cm. The length of the feeder neck was set to 35 mm, which is the average of the neck’s cross-sectional dimensions. The neck was designed with a 60° transition towards the feeder and a 45° transition towards the casting.
Six venting risers (35 mm × 5 mm) were placed at the top to ensure cavity venting and prevent gas porosity. External chills were placed on the casting bottom and between adjacent feeders on the flange to extend the end zone, improve the effective feeding distance of the feeders, and prevent mutual interference between feeders. Since the printed sand cores cannot have feeders and chills pre-installed, their positions were reserved during the 3D printing casting design phase, and the feeder sleeves and chills were placed after the cores were printed. This strategic placement of feeders and chills is a key step in optimizing the 3D printing casting process.
Numerical Simulation and Optimization of 3D Printing Casting Process
We used MAGMA software for numerical simulation analysis. The simulation results from the final optimized process showed that the mold filling was rapid and stable, with a good liquid metal front rise and no gas entrapment. The filling time was 16 seconds, the pouring weight was 400 kg, and the average filling rate was 25 kg/s. From the liquid phase solidification analysis, the chilling effect of the chills was evident. The liquid phase shrinkage was consistently directed towards the feeders, which were the last to solidify. The feeder modulus and liquid volume were appropriate, ensuring good feeding. The process yield reached 82%. The porosity analysis indicated no shrinkage porosity at the machined positions of the product, satisfying the product requirements. The process was finalized for development based on this plan, and the sand cores were designed for 3D printing casting.
Design of 3D Printed Sand Cores
Although 3D printed sand cores are not constrained by parting lines or mold draft, cleaning loose sand and applying coatings after printing remain necessary. Therefore, the overall core was divided into several parts. The core assembly was split into an outer skin core and an inner core. Because the outer skin core was large and difficult to transport and clean, it was further divided into upper, middle, and lower sections. The inner core was split into two parts. The gating system was integrated into the inner core and the top feeder core. After cleaning the loose sand, the cores were coated with an alcohol-based coating. Finally, the cores were assembled. To reduce resin consumption and shorten printing time, non-critical areas were hollowed out for weight reduction, while key areas were reinforced with ribs to enhance strength. This approach in 3D printing casting reduced the core weight while ensuring strength. The core surfaces were coated with an alcohol-based refractory coating to prevent metal penetration and sand sintering. The cores were assembled using rabbets and concave-convex positioning pins for precise alignment. Before assembly, the filters, chills, and feeders were placed. The core prints were sealed with adhesive during assembly. Due to the high mold weight and significant buoyancy force from the molten metal, the upper and lower outer skin cores were fastened with bolts. The outer skin cores fixed the inner core in place. After assembly, the whole mold was a single unit, preventing core shift and mold expansion, effectively ensuring the dimensional accuracy of the product.
This image illustrates the type of complex sand core geometry achievable through 3D printing casting, which is integral to the success of the sprocket hub production.
Melting and Pouring Process for 3D Printing Casting
The pouring process employed a steel shot embedding method to ensure the strength of the sand core, preventing mold expansion and core shift. A medium-frequency induction furnace was used for melting. The raw materials included pig iron, scrap steel, and return scrap. The carbon and silicon contents in the molten metal promote graphitization. Under certain conditions, increasing the carbon equivalent can enhance the degree of graphitization. The carbon content was controlled at 3.4%–3.7%, and the silicon content at 2.5%–2.8%. Manganese can lower the eutectoid transformation temperature and refine and stabilize pearlite. Typically, manganese content should be below 0.5%, controlled at 0.45% ± 0.05%. Other chemical composition control points are listed in Table 3. The tapping temperature was controlled between 1,460 °C and 1,500 °C, and the pouring temperature between 1,360 °C and 1,380 °C. Strict control of the melting and pouring parameters is crucial for the success of any 3D printing casting project.
| Element | Range / Target |
|---|---|
| C | 3.4 – 3.7 |
| Si | 2.5 – 2.8 |
| Mn | 0.45 ± 0.05 |
| P | ≤ 0.05 |
| Cu | < 0.2 |
| Ti | ≤ 0.05 |
| S | ≤ 0.015 |
| Sn | 0.025 ± 0.005 |
| Cr | ≤ 0.1 |
| Mg | 0.048 ± 0.008 |
Inspection and Machining of 3D Printing Casting Components
After the sand cores were printed, they were cleaned, coated, and assembled into a mold. The casting was then poured, cooled, and removed from the mold for shakeout and grinding. Due to the high dimensional accuracy of the 3D printing casting process and the reduced number of cores, the casting exhibited significantly fewer flash and fins compared to conventionally tooled castings. The surface quality was good, requiring only simple grinding and cleaning. The casting subsequently passed UT inspection and machining verification. With thorough planning and strict process control, the 3D printing casting and core assembly approach led to a successful first-time development. The internal quality of the casting was qualified, with good graphite morphology and a qualified matrix structure. The mechanical properties and microstructure are presented in Table 4.
| Property | Value |
|---|---|
| Tensile Strength (MPa) | 589 |
| Yield Strength (MPa) | 345 |
| Elongation (%) | 10.6 |
| Hardness (HBW) | 196 |
| Nodularity (%) | 95 |
| Pearlite Content (%) | 55 |
Conclusions on 3D Printing Casting Application
(1) 3D printing casting offers great process flexibility, not being limited by mold stripping or product structure. However, care must be taken in selecting the parting line positions to facilitate loose sand removal and coating application. The design for manufacturability of the cores themselves is a critical step in 3D printing casting.
(2) 3D printing casting eliminates the need for molds, saving both mold fabrication time and cost. Furthermore, when product modifications or upgrades are needed, there is no concern about mold rework. This makes the process highly suitable for the rapid fabrication of castings, providing strong support for the early-stage research and development of the sprocket hub.
(3) The entire process, from process design to finished product machining, took 15 days. Compared to the traditional mold-making process, this saved 30 days in mold manufacturing time and saved 200,000 CNY in mold costs. A total of four sprocket hubs were successfully produced and installed in the final machinery, providing strong support for the overall machine’s research and development, thereby saving both time and costs. The successful application of 3D printing casting in this project demonstrates its significant potential for accelerating product development cycles and reducing upfront investment in the foundry industry.