I focused my research on rapid sand mould casting of a complex aluminium cylinder head, because this component combines thin walls, internal water jackets, intake passages, exhaust passages, oil galleries, and several thick sections that must be fed during solidification. My objective was to connect CAD modelling, casting process design, SLA pattern manufacturing, resin-bonded sand moulding, inverse heat-transfer analysis, and numerical simulation into one practical route for rapid sand mould casting. I wanted the route to be useful not only for a single prototype but also for low-volume production and for process optimisation before metal is poured. In my work, rapid sand mould casting was not treated as a simple replacement of a wooden pattern or a metal pattern. Instead, I treated it as an integrated manufacturing method in which the pattern, core boxes, gating system, risers, moulding material, and simulation parameters all influence the final quality of the cylinder head.

1. Motivation for Rapid Sand Mould Casting of a Cylinder Head
I selected a cylinder head because it is one of the most demanding parts in an engine. It must withstand thermal cycling, mechanical loading, pressure testing, and strict dimensional requirements. In conventional development, several rounds of pattern making, moulding, pouring, inspection, and correction are usually required. This makes the development cycle long and expensive. Rapid sand mould casting offers a different logic. A digital model can be converted into an SLA pattern or core box, and the sand mould can then be produced directly around that pattern. I found this especially valuable for a cylinder head because the internal passages are too complex for simple moulding and require multiple sand cores. The use of rapid sand mould casting allowed me to shorten the design loop, reduce dependence on a metal pattern, and test several gating concepts before committing to a final layout.
I also considered the broader context of automotive aluminium castings. Aluminium alloys are widely used for cylinder heads, cylinder blocks, intake manifolds, wheels, and structural components because of their low density, good thermal conductivity, and acceptable castability. However, aluminium alloys also have a relatively narrow freezing range in some compositions and can be sensitive to turbulence, oxide films, shrinkage, and gas entrapment. In sand mould casting, the mould has low thermal conductivity and good collapsibility, which helps complex shapes to be formed, but the process still requires careful control of filling and feeding. My research therefore combined the flexibility of rapid sand mould casting with the predictive power of CAE simulation.
| Feature of the cylinder head | Manufacturing requirement | Consequence for rapid sand mould casting |
|---|---|---|
| Complex internal water jacket | Accurate core assembly and core stability | Multiple sand cores must be designed, located, and vented |
| Thin walls and ribs | Complete filling without cold shut | Gating must be smooth and filling time must be controlled |
| Thick bosses and bearing zones | Feeding during solidification | Risers and chills or local feeding paths may be needed |
| Intake and exhaust ports | Dimensional accuracy and surface quality | Core boxes and pattern tolerances are critical |
| Pressure tightness | Low porosity and no connected shrinkage | Numerical simulation of filling and solidification is essential |
2. Process Chain I Established
I established a process chain for rapid sand mould casting that began with a three-dimensional CAD model. I used a parametric CAD environment to build the cylinder head geometry, and then I imported the model into a casting process design system. The design system allowed me to set parting surfaces, add machining allowance, apply shrinkage, define cores, design the gating system, place risers, and evaluate the assembly. After the process design was complete, the model was prepared for CAE simulation. The mesh was generated, boundary conditions were assigned, and the filling and solidification stages were computed. Finally, the results were used to select the best gating system and to manufacture the sand mould for trial casting.
| Step | Action | Output |
|---|---|---|
| 1 | Create 3D cylinder head model | Solid model with main external and internal features |
| 2 | Analyse casting process | Parting direction, core division, feeding strategy |
| 3 | Design sand cores | Water jacket core, intake core, exhaust core, oil gallery cores |
| 4 | Design gating system | Sprue, runner, ingate layout and area ratio |
| 5 | Design risers and vents | Feeding paths and gas removal paths |
| 6 | Generate mesh and simulate | Filling sequence, temperature field, solid fraction, shrinkage prediction |
| 7 | Manufacture patterns and mould | SLA patterns, core boxes, resin sand mould |
| 8 | Pour and inspect | Prototype cylinder head casting |
3. Three-Dimensional Modelling and Process Design
I built the cylinder head model as a thin-walled box-type component. The overall dimensions were approximately 426 mm by 200 mm by 131 mm, the volume was about 4581 cm³, and the mass was about 12.4 kg. The minimum wall thickness was about 4 mm. These values made the part suitable for sand mould casting but also made it vulnerable to premature freezing, incomplete filling, and deformation. I used a machining allowance of 4 mm for general surfaces and 5 mm near the top face where a riser was connected. I applied a linear shrinkage of 1% to compensate for contraction during cooling. I divided the sand core system into several functional cores: lower left oil gallery core, lower right oil gallery core, water jacket core, upper oil gallery core, intake port core, and exhaust port core. The core assembly had to maintain accurate relative positions, so I introduced core prints, locating blocks, and core supports.
| Core or mould part | Main function | Critical issue |
|---|---|---|
| Water jacket core | Forms internal cooling passages | Thin sections, core strength, gas venting |
| Intake port core | Forms intake passages | Surface roughness and dimensional accuracy |
| Exhaust port core | Forms exhaust passages | Thermal resistance and core stability |
| Oil gallery cores | Forms lubrication channels | Core location and clearance |
| Bottom core | Supports lower oil gallery cores | Seating and alignment |
| Top core | Locates upper oil gallery core | Three-point positioning |
For the gating system, I used an open-type layout to reduce the risk of turbulence and oxide entrainment. The area ratio was selected as:
$$
A_{\text{sprue}} : A_{\text{runner}} : A_{\text{ingate}} = 1 : 2 : 4
$$
The ingate area was estimated from the pouring mass, pouring time, density, friction coefficient, and average static head:
$$
A_{\text{in}} = \frac{G_L}{\rho_L \mu t \sqrt{2gH_p}}
$$
where \(G_L\) is the mass of liquid metal passing through the choke, \(\rho_L\) is the liquid density, \(\mu\) is the flow coefficient, \(t\) is the pouring time, \(g\) is gravitational acceleration, and \(H_p\) is the average static head. I also used an empirical pouring time relation:
$$
t = \delta \sqrt{G_L}
$$
For the riser design, I used the modulus method. The modulus of a casting or a riser is the ratio of its volume to its heat-transfer surface area:
$$
M = \frac{V}{A}
$$
The solidification time can then be related to the modulus by Chvorinov’s rule:
$$
t_c = \left(\frac{M_c}{K_c}\right)^2
$$
$$
t_r = \left(\frac{M_r}{K_r}\right)^2
$$
For effective feeding, the riser must remain liquid longer than the casting section it feeds:
$$
t_r \ge t_c
$$
I placed two standard kidney-shaped open risers on the top face. Their positions were chosen to feed the thick sections while allowing gas to escape. I also added vents and vent plugs in the sand cores and mould surfaces to avoid gas defects in the rapid sand mould casting process.
4. Material Properties and Boundary Conditions
The cylinder head was cast from ZL105 aluminium alloy, an Al-Si-Mg alloy. Its chemical composition is summarised below. I used this alloy because it combines good castability with moderate strength and thermal stability. The liquidus temperature was about 622 °C and the solidus temperature was about 536 °C.
| Element | Si | Cu | Mg | Al |
|---|---|---|---|---|
| Mass fraction (%) | 4.5–5.5 | 1.0–1.5 | 0.4–0.6 | Balance |
The thermal properties of ZL105 aluminium alloy vary with temperature. I imported these data into the simulation so that the solid fraction, density, enthalpy, and thermal conductivity changed continuously during cooling. The solid fraction was especially important because it controlled the transition from liquid feeding to isolated liquid regions. The density decreased as temperature increased, and the slope became steeper in the liquid-solid two-phase region. The enthalpy curve included latent heat release, and the thermal conductivity changed with temperature. These data made the numerical simulation more realistic than a constant-property model.
| Property | Value or trend |
|---|---|
| Liquidus temperature | 622 °C |
| Solidus temperature | 536 °C |
| Density | Decreases with temperature; steeper change in two-phase region |
| Enthalpy | Increases with temperature and includes latent heat |
| Thermal conductivity | Temperature-dependent |
| Initial mould temperature | 20–30 °C |
| Gravity | 9.8 m/s² |
I set the pouring temperature in the range of 690–720 °C and the pouring rate in the range of 0.75–1.5 kg/s during preliminary trials. The mould and core material was phenolic urethane resin-bonded sand. The boundary conditions included the interface heat-transfer coefficient between the casting and the sand mould, the initial temperatures of the casting and mould, the gravity vector, and the thermal conditions at the outer mould surface. These parameters were not treated as arbitrary values. I measured the interfacial heat-transfer coefficient for the ZL105 aluminium alloy and the resin-bonded sand because this coefficient strongly affects the predicted cooling rate and the predicted shrinkage defect locations.
5. Inverse Determination of Interfacial Heat Transfer Coefficient
I designed a separate experiment to determine the interfacial heat transfer coefficient for the rapid sand mould casting of ZL105 aluminium alloy in phenolic urethane resin-bonded sand. I simplified the geometry to a symmetrical cylindrical casting so that the heat flow could be approximated as one-dimensional during the early stage. I placed thermocouples at several distances from the casting-mould interface. One thermocouple measured the casting temperature close to the interface, and several thermocouples measured the sand temperature at distances of 5 mm, 10 mm, 15 mm, and 20 mm from the interface. I insulated the top and bottom of the assembly to reduce heat loss to the environment.
The interfacial heat flux and the interfacial heat-transfer coefficient were defined by:
$$
h(t) = \frac{q(t)}{T_1(t) – T_2(t)}
$$
where \(q(t)\) is the heat flux through the interface, \(T_1(t)\) is the casting surface temperature, and \(T_2(t)\) is the mould surface temperature. I used an inverse heat conduction method to estimate the heat flux from the measured temperatures. The one-dimensional heat conduction equation in cylindrical coordinates was written as:
$$
\frac{\partial T}{\partial t}
= \alpha \left( \frac{\partial^2 T}{\partial r^2} + \frac{1}{r}\frac{\partial T}{\partial r} \right)
$$
I used a finite-difference formulation to relate the nodal temperatures. The Fourier number was defined as:
$$
F_0 = \frac{\alpha \Delta t}{(\Delta r)^2} = \frac{k \Delta t}{\rho c (\Delta r)^2} \le 0.5
$$
The correction to the assumed heat flux was obtained from the difference between measured and calculated temperatures:
$$
q_{\text{corr}} = q + \Delta q
$$
$$
\frac{\Delta q}{q} \le \varepsilon
$$
The measured cooling curves showed that the casting centre cooled quickly at first, then more slowly as the solidification front advanced. The mould temperature increased rapidly near the interface and then gradually approached the casting temperature. The estimated heat flux reached a maximum of about \(2.56 \times 10^5 \, \mathrm{W/m^2}\) in the first stage and then decreased rapidly. After about 500 s, the heat flux decreased to a few thousand watts per square metre and then stabilised.
| Temperature condition | Physical state | Interfacial heat-transfer coefficient |
|---|---|---|
| About 690 °C | Liquid metal just poured | About 811 W/(m²·K) |
| Above 622 °C | Liquid region | About 790 W/(m²·K) |
| 622–536 °C | Liquid-solid two-phase region | Decreases markedly |
| Below 536 °C | Solid region | About 310 W/(m²·K) |
| Later cooling | Solid casting and warm mould | About 230 W/(m²·K) |
I validated the inverse result by using the obtained heat-transfer coefficient in a forward simulation. The calculated temperatures at the selected points were compared with the measured temperatures. The maximum difference was about 11 °C, and the trends were consistent. This confirmed that the inverse method was reliable enough for the rapid sand mould casting simulation of the cylinder head.
6. Mesh Generation and Simulation Setup
I used tetrahedral elements for the mesh because tetrahedral meshes can follow the complex internal passages of the cylinder head. The mesh size was chosen so that the smallest wall thickness was represented by at least two or three elements. The total number of nodes and elements depended on the gating design. The mesh quality was checked by dihedral angle, radius ratio, and edge length ratio. I avoided highly skewed elements because they can cause poor convergence and inaccurate temperature gradients.
| Gating scheme | Nodes | Elements |
|---|---|---|
| Top unilateral injection | 107,385 | 471,675 |
| Top symmetrical injection | 107,035 | 470,289 |
| Bottom unilateral injection | 115,990 | 508,256 |
I used a virtual mould in the simulation to reduce computation time. The virtual mould was defined by a rectangular box around the casting. The mould material was the same resin-bonded sand used in the experiment. The interface between the casting and the mould used the temperature-dependent heat-transfer coefficient obtained from the inverse method. The outer boundary of the virtual mould was treated as adiabatic. This approach allowed me to simulate the rapid sand mould casting process without meshing the entire sand box.
7. Governing Equations for Filling and Solidification
The filling process in sand mould casting involves momentum, mass, and energy transport. I used the continuity equation to express mass conservation:
$$
\frac{\partial u}{\partial x} + \frac{\partial v}{\partial y} + \frac{\partial w}{\partial z} = 0
$$
The momentum equations in the three coordinate directions were written as:
$$
\rho \left( \frac{\partial u}{\partial t} + u\frac{\partial u}{\partial x} + v\frac{\partial u}{\partial y} + w\frac{\partial u}{\partial z} \right)
= -\frac{\partial p}{\partial x} + \mu \left( \frac{\partial^2 u}{\partial x^2} + \frac{\partial^2 u}{\partial y^2} + \frac{\partial^2 u}{\partial z^2} \right) + \rho g_x
$$
$$
\rho \left( \frac{\partial v}{\partial t} + u\frac{\partial v}{\partial x} + v\frac{\partial v}{\partial y} + w\frac{\partial v}{\partial z} \right)
= -\frac{\partial p}{\partial y} + \mu \left( \frac{\partial^2 v}{\partial x^2} + \frac{\partial^2 v}{\partial y^2} + \frac{\partial^2 v}{\partial z^2} \right) + \rho g_y
$$
$$
\rho \left( \frac{\partial w}{\partial t} + u\frac{\partial w}{\partial x} + v\frac{\partial w}{\partial y} + w\frac{\partial w}{\partial z} \right)
= -\frac{\partial p}{\partial z} + \mu \left( \frac{\partial^2 w}{\partial x^2} + \frac{\partial^2 w}{\partial y^2} + \frac{\partial^2 w}{\partial z^2} \right) + \rho g_z
$$
The energy equation included convection, conduction, and the source term from latent heat:
$$
\frac{\partial T}{\partial t} + u\frac{\partial T}{\partial x} + v\frac{\partial T}{\partial y} + w\frac{\partial T}{\partial z}
= \frac{\lambda}{\rho c_p} \left( \frac{\partial^2 T}{\partial x^2} + \frac{\partial^2 T}{\partial y^2} + \frac{\partial^2 T}{\partial z^2} \right) + \frac{Q}{\rho c_p}
$$
For solidification, the heat conduction equation was written as:
$$
\rho c_p \frac{\partial T}{\partial t}
= \frac{\partial}{\partial x}\left(\lambda \frac{\partial T}{\partial x}\right)
+ \frac{\partial}{\partial y}\left(\lambda \frac{\partial T}{\partial y}\right)
+ \frac{\partial}{\partial z}\left(\lambda \frac{\partial T}{\partial z}\right) + Q
$$
The solid fraction was approximated from the liquidus and solidus temperatures:
$$
f_s = \frac{T_L – T}{T_L – T_S}
$$
This relation was used to identify isolated liquid regions and to predict shrinkage porosity. I also used the cooling rate and temperature gradient to evaluate feeding conditions:
$$
G = \max \left( \frac{T_i – T_e}{\Delta l_i} \right)
$$
$$
R = \frac{T_e^{t+\Delta t} – T_e^t}{\Delta t}
$$
The ratio \(G/R\) is often used as a thermal criterion for shrinkage defects. When the ratio falls below a critical value, the local region may not be fed adequately.
8. Comparison of Three Gating Systems for Sand Mould Casting
I designed and compared three gating systems. The first was a top unilateral injection system with ingates on one side of the top face. The second was a top symmetrical injection system with ingates distributed symmetrically on both sides of the top face. The third was a bottom unilateral injection system with ingates located near one side of the bottom region. All three systems were evaluated for filling time, filling stability, temperature distribution, solid fraction evolution, and shrinkage defect location.
| Scheme | Gating position | Advantages | Risks |
|---|---|---|---|
| Scheme 1 | Top unilateral injection | Simple moulding; good filling energy; favourable thermal gradient from top | Strong impact on mould; possible turbulence, oxidation, sand inclusion, gas entrapment |
| Scheme 2 | Top symmetrical injection | More uniform ingate distribution; relatively smoother filling than scheme 1 | Metal streams meet in the middle; possible entrapment, turbulence, and cold shut |
| Scheme 3 | Bottom unilateral injection | Smooth filling; reduced direct impact on sand; stable rising of liquid metal | Lower hydrostatic head during early filling; possible incomplete filling if not controlled |
8.1 Top Unilateral Injection
In the top unilateral injection scheme, the metal entered the mould from several ingates on one side of the top face. The total filling time was about 15.69 s. The metal filled the cavity completely, and no obvious misrun was observed. In the early stage, the metal near the ingates reached the bottom first, while the opposite side filled later. After about 8 s, the liquid level became more uniform and rose steadily. However, the initial impact was relatively strong, and there was a risk of sand erosion and oxide entrainment. During solidification, the ingate and riser regions remained hot for a short period and provided some feeding. After about 120 s, the lower thick section began to form isolated liquid regions. The final shrinkage prediction showed several concentrated shrinkage pores in the lower thick part and some dispersed porosity away from the ingates. Therefore, scheme 1 was not the best choice for rapid sand mould casting of this cylinder head.
8.2 Top Symmetrical Injection
In the top symmetrical injection scheme, eight ingates were distributed symmetrically. The total filling time was about 13.98 s. The filling process was more balanced than in scheme 1, but the two streams met near the middle of the mould, which increased the possibility of turbulence, gas entrapment, and oxide film entrapment. The temperature field showed that the ingate and riser regions remained hot, but the lower thick section cooled too quickly for the risers to feed effectively. After about 120 s, isolated liquid regions appeared in the lower part. The final shrinkage prediction showed one concentrated shrinkage region and some dispersed porosity. The defect volume was slightly smaller than in scheme 1, but the risk of turbulence-related defects was still significant. Thus, scheme 2 also required further improvement.
8.3 Bottom Unilateral Injection
In the bottom unilateral injection scheme, the metal entered the mould from the bottom side. The total filling time was about 15.43 s. The filling process was the smoothest of the three schemes. The metal rose steadily after the initial stage, and the mould was filled completely. Because the metal entered from the bottom, direct impact on the sand cores was reduced. The temperature field during solidification showed a favourable gradient from the bottom ingate and the top risers. The ingate remained hot for a longer time, and the risers fed the upper thick sections. After about 220 s, the casting was nearly solid, while the ingate still remained in the liquid-solid two-phase state. The final shrinkage prediction showed only dispersed porosity rather than a large concentrated shrinkage cavity. On the basis of these results, I selected the bottom unilateral injection scheme for the rapid sand mould casting of the cylinder head.
| Scheme | Filling time | Filling stability | Shrinkage defect tendency | Overall evaluation |
|---|---|---|---|---|
| Top unilateral | 15.69 s | Moderate; initial impact | Concentrated shrinkage in lower thick part | Not optimal |
| Top symmetrical | 13.98 s | Moderate; central stream collision | One concentrated shrinkage region and dispersed porosity | Needs improvement |
| Bottom unilateral | 15.43 s | Best; smooth rising | Only dispersed porosity | Selected |
9. Optimisation of Pouring Parameters
After comparing the gating systems, I optimised the pouring temperature and pouring speed. If the pouring temperature is too low, the metal may freeze before the thin sections are filled. If the pouring temperature is too high, oxidation, gas absorption, and sand burn-on can increase. I found that a pouring temperature near 690 °C gave a good balance between fluidity and mould protection. For the pouring speed, I evaluated several values. A very high pouring speed increased turbulence and the risk of sand erosion. A very low pouring speed increased the risk of cold shut and incomplete filling. The selected pouring speed was about 0.22 m/s. At this speed, the metal filled the mould smoothly, the free surface rose steadily, and no obvious turbulence or interruption was observed.
| Parameter | Range studied | Selected value | Reason |
|---|---|---|---|
| Pouring temperature | 690–720 °C | About 690 °C | Good fluidity with lower oxidation risk |
| Pouring speed | 0.75–1.5 kg/s equivalent trials | About 0.22 m/s | Smooth filling without turbulence |
| Mould initial temperature | 20–30 °C | 20–30 °C | Typical workshop condition |
| Heat-transfer coefficient | 200–1000 W/(m²·K) | Temperature-dependent from inverse method | Improved simulation accuracy |
10. Moulding Material for Rapid Sand Mould Casting
I selected phenolic urethane resin-bonded sand as the moulding and core material for rapid sand mould casting. This material offers good flowability, rapid hardening, high strength, and good collapsibility. It is suitable for manual moulding and for complex core assembly. The original sand was natural washed silica sand with a high SiO₂ content and a controlled grain shape. I used a grain size of about 40–70 mesh. The sand had to have low moisture, low clay content, and a suitable temperature. A high moisture content would slow curing and reduce strength. A high clay content would increase binder demand and reduce permeability. A low sand temperature would retard hardening, while an excessively high temperature would reduce final strength.
I performed orthogonal tests to determine the best ratio of the two resin components. The two components were mixed with the original sand in different proportions. The tensile strength of the cured sand specimens was measured. When component I or component II was below about 0.8% of the original sand mass, the moulding strength was insufficient. When one component was sufficient but the other was too low, the sand specimen also failed to reach the required strength. The tensile strength increased with increasing resin content, but the rate of increase became smaller above about 2% for component I and above about 2.2% for component II. For rapid sand mould casting, excessive resin content makes sand removal difficult and increases cost. I therefore selected a practical ratio.
| Component I mass fraction in sand | Component II mass fraction in sand | Ratio | Tensile strength tendency | Moulding result |
|---|---|---|---|---|
| <0.8% | Any | Not balanced | Too low | Failed |
| 0.8–1.0% | 0.8–1.0% | About 1:1 | About 1.1 MPa | Suitable |
| 2.0% | 2.0% | About 1:1 | Higher but less efficient | Excess resin |
| >2.5% | >2.5% | About 1:1 | High | Difficult to clean and costly |
The optimal ratio of component I to component II was approximately 1:1, and the mass fraction of component I in the original sand was about 0.8%–1.0%. This ratio gave a tensile strength of about 1.1 MPa, which was sufficient for the rapid sand mould casting of the cylinder head. It also reduced resin consumption and improved sand collapsibility after pouring.
11. Experimental Production and Verification
I used the selected gating system, riser layout, and moulding material to produce a trial cylinder head by rapid sand mould casting. The sand cores were assembled in the correct sequence. The water jacket core, intake port cores, exhaust port cores, and oil gallery cores were located by core prints and locating blocks. Venting paths were kept open to allow gas to escape during pouring. The mould was closed, weighted, and poured at about 690 °C. After cooling, the casting was cleaned and inspected. The trial casting filled completely and showed good external contour. The internal passages were formed correctly, and no large shrinkage cavity was found. Some dispersed porosity was predicted by simulation and was consistent with the experimental observations. The trial confirmed that the selected rapid sand mould casting route was practical for a complex cylinder head.
| Experimental stage | Observation | Conclusion |
|---|---|---|
| Pattern and core box preparation | SLA patterns and core boxes were produced | Rapid sand mould casting reduced pattern lead time |
| Core assembly | Cores located and vented | Core positioning was stable |
| Mould filling | Mould filled completely | Selected bottom gating worked well |
| Solidification | No large concentrated shrinkage | Riser and ingate design was acceptable |
| Final casting | Good external shape and internal passages | Rapid sand mould casting route was validated |
12. Discussion
My research showed that numerical simulation is highly valuable for rapid sand mould casting of a cylinder head, but its accuracy depends on the boundary conditions. The interfacial heat-transfer coefficient between the aluminium alloy and the resin-bonded sand was one of the most sensitive parameters. If I had used a constant value, the predicted cooling rate and shrinkage defect locations would have been less reliable. The inverse method provided a temperature-dependent coefficient that improved the agreement between measured and simulated temperatures. I believe this step is essential for any serious sand mould casting simulation involving ZL105 aluminium alloy and phenolic urethane resin sand.
The comparison of the three gating systems also showed that filling stability and feeding behaviour must be considered together. Top gating can provide good filling energy and favourable thermal gradients in some castings, but for this cylinder head it produced strong initial impact and led to concentrated shrinkage in the lower thick sections. Symmetrical top gating improved the distribution of metal, but the meeting of two streams increased the risk of turbulence and oxide entrapment. Bottom gating produced the smoothest filling and the best feeding behaviour because the ingate remained hot longer and the risers fed the upper thick sections. The final defect prediction showed only dispersed porosity rather than a large shrinkage cavity. This result supports the use of bottom unilateral gating for rapid sand mould casting of similar thin-walled, complex components.
The moulding material also played an important role. Phenolic urethane resin-bonded sand provided the necessary strength for complex cores and the necessary collapsibility for easy removal. The orthogonal tests showed that a balanced ratio of the two resin components was more important than simply increasing the total resin content. A ratio of about 1:1 with component I at 0.8%–1.0% of the sand mass gave sufficient tensile strength while avoiding excessive resin consumption. This finding is useful for rapid sand mould casting because it reduces cost and improves sand reclamation.
13. Conclusions
I developed and verified a rapid sand mould casting route for a complex aluminium cylinder head. The main conclusions are as follows.
| Area | Conclusion |
|---|---|
| Process design | A complete CAD-based rapid sand mould casting route was established, including parting, core design, gating, riser design, and venting. |
| Heat transfer | The interfacial heat-transfer coefficient between ZL105 aluminium alloy and resin-bonded sand was determined by an inverse method. It was about 811 W/(m²·K) at 690 °C, decreased through the two-phase region, and stabilised near 230 W/(m²·K) after solidification. |
| Simulation | Three gating systems were compared. Bottom unilateral injection gave the smoothest filling and the least concentrated shrinkage. |
| Pouring parameters | A pouring temperature of about 690 °C and a pouring speed of about 0.22 m/s gave stable filling without turbulence. |
| Moulding material | Phenolic urethane resin-bonded sand with a component ratio of about 1:1 and component I at 0.8%–1.0% of the sand mass gave adequate strength and good collapsibility. |
| Experimental validation | The trial casting filled completely and showed acceptable internal quality, validating the rapid sand mould casting approach. |
Overall, I found that rapid sand mould casting is a practical and efficient method for developing complex engine components such as a cylinder head. By combining CAD, CAE, inverse heat-transfer analysis, SLA patterns, and resin-bonded sand, I was able to reduce trial-and-error, predict defects before pouring, and produce a functional prototype. I also confirmed that the selection of the gating system and the control of the interfacial heat-transfer coefficient are decisive for the quality of the final sand mould casting. The approach can be extended to other thin-walled, high-integrity aluminium castings, especially when the production volume is low and the design is still evolving.
