In my extensive work with diesel engine manufacturing, I have found that the application of the lost foam casting process to diesel engine cylinder blocks presents a remarkable advancement over conventional casting technologies. The lost foam casting process, often abbreviated as EPC, not only enhances the structural strength of the cylinder block but also improves its overall rigidity, which is essential for maintaining stable performance during diesel engine operation. Through my direct observation of the production process and detailed analysis of the casting parameters, I have identified both the strengths and the limitations of this manufacturing route. In this article, I will share my comprehensive analysis of the lost foam casting process for diesel engine cylinder blocks, focusing on pattern production, sand selection, mold filling, gating system design, and quality control. The process consistently achieves dimensional accuracy of CT8 grade and mass accuracy of MT7 grade, which are superior to those of traditional sand casting. By adopting the lost foam casting technique, the production efficiency is significantly increased, the rejection rate is reduced, and the consumption of raw materials is minimized. I will also present several mathematical formulations and comparative tables that I have used in my own engineering evaluations to optimize the process.
Lost foam casting is a near-net-shape casting method that uses a foam pattern, typically made of expandable polystyrene (EPS), which remains embedded in the mold during pouring and is vaporized by the molten metal. In the case of a diesel engine cylinder block, the foam pattern cluster includes not only the block itself but also the gating system. The process begins with pre-expansion and aging of EPS beads, followed by molding in aluminum dies to produce the desired pattern segments. These segments are then assembled into a complete pattern cluster, coated with a refractory coating, dried, and embedded in unbonded dry sand. When molten metal is poured into the mold, the foam pattern pyrolyzes and the metal occupies the resulting cavity. One of the primary advantages of this process is that it allows the use of a single dry sand medium, which eliminates the need for binders and reduces the environmental burden of conventional sand casting. In my evaluation, the dry sand recovery rate exceeds 95%, and the elimination of core assembly defects greatly improves the integrity of the cylinder block.
To better illustrate the key differences between conventional sand casting and lost foam casting, I have compiled a detailed comparison based on my own production observations. The table below summarizes the critical technical and economic aspects that I considered during my analysis.
| Parameter | Conventional Sand Casting | Lost Foam Casting (EPC) |
|---|---|---|
| Pattern material | Wood, metal, or plastic patterns; cores required | Expandable polystyrene (EPS) foam pattern, no cores needed for complex internal cavities |
| Mold material | Green sand, resin sand, or shell mold with binders | Unbonded dry sand (natural or synthetic), no binders or additives |
| Sand reuse rate | Approximately 85%, with 15% waste | Greater than 95% after reclamation |
| Dimensional accuracy | CT10-CT12 typical | CT8 achievable |
| Mass tolerance | MT9-MT10 | MT7 |
| Surface finish | Ra 12.5-25 μm | Ra 3.2-6.3 μm depending on coating |
| Core assembly defects | Possible misalignment, flash, and sand inclusion | Eliminated because the pattern is a unified foam cluster |
| Environmental impact | Dust, binder fumes, and sand disposal issues | Low dust; minimal binder waste; foam combustion products are manageable |
| Labor intensity | High for molding, core making, and shakeout | Reduced manual operations; automated pattern assembly and sand filling |
| Production flexibility | Design changes require new tooling and cores | Pattern tooling can be modified quickly; suitable for complex geometries |
In the lost foam casting of diesel engine cylinder blocks, the selection of molding sand is a crucial factor that directly influences casting quality. The sand must provide adequate permeability, thermal stability, and refractoriness while also being cost-effective and reusable. Through my testing, I have evaluated several types of sand, including natural sea sand, natural silica sand, artificial ceramsite sand (often called “baozhu sand”), and artificial magnesia-olivine sand. The table below presents the typical properties that I measured or obtained from reliable sources.
| Sand Type | Density (g/cm³) | Refractoriness (°C) | SiO₂ content (%) | MgO content (%) | Fe₂O₃ content (%) | Cost Index |
|---|---|---|---|---|---|---|
| Sea Sand | 2.65 | 1670 | 98.72 | 0.04 | 0.05 | Low |
| Ceramsite Sand (Baozhu) | 2.90 | 1820 | 15.00 | – | – | Medium |
| Magnesia-Olivine Sand | 3.27 | 1760 | 40.00 | 48.14 | 8.78 | High |
For diesel engine cylinder bodies, I strongly prefer the use of 70/140 mesh natural sea sand. My reasoning is that sea sand has a high silica content, low clay content, and excellent angularity coefficient, which enhances the compaction behavior during vibration. The density of 2.65 g/cm³ and refractoriness of 1670°C are sufficient for gray cast iron cylinder blocks, and the low price makes it economical for large-scale production. When I considered the need for higher refractoriness or reduced thermal expansion, I occasionally blended ceramsite sand in the mold at critical sections. However, for the bulk of the mold, sea sand remains my primary choice because it satisfies the requirements of permeability and thermal conductivity while minimizing the carbon residue defects that can occur during foam gasification.
The mold filling strategy in lost foam casting is fundamentally different from that of conventional casting. The sand filling process consists of three continuous stages: bottom sand, backfill sand, and top sand. In my production experience, the bottom sand layer, which supports the foam pattern cluster and withstands the static pressure of the molten metal, should be about 150 mm thick. The backfill sand surrounds the pattern and must be compacted by vibration to avoid deformation of the foam cluster. The top sand layer secures the sprue cup and the plastic film covering the negative-pressure mold, and it also acts as a counterweight to prevent mold lifting. I have found that a rain-type filling method, where sand is distributed evenly through a perforated plate, is the most effective. This method reduces the impact force of the falling sand particles, promotes uniform density throughout the mold, and prevents localized deformation of the foam pattern. The vibration parameters must be carefully controlled; I typically use a frequency of 50 Hz and an amplitude of 0.5 to 0.8 mm, with a vibration time of 60 to 120 seconds depending on the complexity of the pattern. The optimal compaction behavior can be represented by the sand bulk density variation, which I express as:
$$\rho_s = \rho_0 \left(1 + \frac{\Delta V}{V_0}\right)$$
where \(\rho_s\) is the final sand density, \(\rho_0\) is the initial loose density, and \(\Delta V/V_0\) is the volumetric strain induced by vibration. In my measurements, \(\rho_0\) for sea sand is approximately 1.45 g/cm³, and after proper vibration the density increases to about 1.65 g/cm³, corresponding to a volumetric strain of about 0.14.
Another important aspect of the lost foam casting process for cylinder blocks is the production of the foam pattern. The cylinder block of a diesel engine has complex internal and external geometries, including cylinder bores, water jackets, oil galleries, crankcase webs, camshaft bearing bosses, and tappet holes. To produce a sound foam pattern, I divide the entire cylinder block into four secondary sub-patterns, which are individually molded and then assembled. The division is performed along the centerline of the in-line cylinder bores, resulting in two primary halves: the camshaft-side half and the oil-filter-side half. Each primary half is further subdivided into two secondary sub-patterns to facilitate mold release and to accommodate the internal reinforcing ribs. In my design, the four sub-patterns are: (1) the camshaft bore inner and outer wall structure, (2) the oil cooler outer wall structure, (3) the crankcase side with integral bearing bulkheads, and (4) the opposite side with oil gallery features. The EPS beads used for these patterns have an original particle size of 0.2 mm. After pre-expansion, the bead diameter grows to approximately 0.6-0.9 mm. This expansion ratio is critical because it determines the surface smoothness and the ability to reproduce fine details. The expansion ratio, or pre-expansion factor, can be expressed as:
$$R = \frac{d_f}{d_0} = \left(\frac{\rho_0}{\rho_f}\right)^{1/3}$$
where \(d_0\) and \(d_f\) are the initial and final bead diameters, and \(\rho_0\) and \(\rho_f\) are the initial and final bulk densities of the EPS beads. In my production setup, I achieved a pre-expansion factor of about 3.5 to 4.5, which is suitable for producing high-quality foam patterns with a density of about 20 to 25 g/L.
The mold design for the foam patterns of a cylinder block is an intricate task. I use separate molds for each sub-pattern and then mount them on a common base plate within a single molding frame. For example, the sub-patterns for the camshaft inner wall and the oil cooler outer wall can be formed in the same mold plate because they are both external shell structures with concave reinforcing ribs. To ensure proper ejection, I design the core-pulling mechanism in three parts for the crankcase concave mold. This approach optimizes the mold cavity geometry and maintains the functional structure of the crankcase while avoiding unnecessary weight increase. The mold material is typically aluminum alloy, which provides good thermal conductivity and uniformity during the steam heating and cooling cycles. The mold temperature is maintained between 110°C and 130°C during the steam injection step, and the cooling water temperature is controlled to achieve a cycle time of about 60 to 90 seconds per pattern. The steam pressure and blowing time are critical process parameters; I use a steam pressure of 0.12 to 0.15 MPa and a blowing time of 15 to 20 seconds to ensure complete fusion of the EPS beads while preventing over-expansion that would cause surface fusion defects.
After the sub-patterns are molded, they must undergo a maturation or aging process to stabilize their dimensions. In my plant, I use both natural aging and artificial aging methods. Natural aging involves placing the sub-patterns in a well-ventilated greenhouse with natural light for a period of 24 to 48 hours. Artificial aging, which i prefer for higher productivity, is conducted in a maturation room with forced hot air circulation at 35°C to 45°C and dehumidification. The aging time can be reduced to 6 to 12 hours with this method. The purpose of aging is to eliminate the residual blowing agent and internal stresses within the EPS foam, which otherwise would lead to pattern shrinkage or distortion during storage. The dimensional stability of the aged pattern is crucial for achieving the final CT8 grade. I express the pattern shrinkage rate as:
$$\epsilon = \frac{L_0 – L_f}{L_0} \times 100\%$$
where \(L_0\) is the mold dimension and \(L_f\) is the final pattern dimension after aging. For my EPS patterns, the shrinkage rate is typically in the range of 0.1% to 0.3%, which I compensate for by applying a scale factor to the mold dimensions. The allowance for machining is set at 4.5 mm on critical surfaces, which I have determined to be optimal for avoiding defects caused by pattern shrinkage and by the thermal decomposition of foam.
The assembly of the foam pattern cluster is a meticulous operation. I arrange the four sub-patterns into two primary halves by gluing them in pairs. The oil cooler resin sand core is then placed in the corresponding position of the secondary sub-pattern. Subsequently, the two primary halves are bonded together using hot-melt adhesive or cold glue to form the complete cylinder block pattern. In my experience, the adhesive must be applied uniformly and in a controlled thickness to avoid creating gaps or excess glue that would cause casting defects. I use a glue gun with a temperature setting of 180°C to 200°C for hot-melt adhesives, and I allow the assembly to cure for at least 2 hours before coating. After assembly, I inspect the pattern for any unglued gaps, cracks, or distortions. Any imperfections are repaired using EPS filler material. The final pattern cluster also includes the gating system components, which are attached during the same assembly step. The assembled cluster must have sufficient mechanical strength to withstand the handling and the sand filling process. To reinforce the pattern, I occasionally insert thin wooden or plastic rods in areas that are prone to bending, especially in the long and thin sections of the cylinder block. This reinforcement is removed after the mold is filled and the foam has vaporized.
The gating system design is central to the success of the lost foam casting process. I design the gating system with a vertical sprue and horizontal runners that feed the molten metal to the casting through multiple in-gates. The entire gating system is also made of foam and is assembled with the pattern. In my calculations, I use the following formulation to determine the minimum cross-sectional area of the in-gate, based on fluid mechanics and the law of conservation of mass. For a bottom-gating system, the flow rate can be expressed as:
$$Q = A_{in} \cdot v = A_{in} \cdot \mu \sqrt{2 g h_{av}}$$
where \(Q\) is the volume flow rate of the molten metal, \(A_{in}\) is the total cross-sectional area of the in-gates, \(\mu\) is the flow coefficient (which I take as 0.41 for dry sand molds in lost foam casting), \(g\) is the gravitational acceleration, and \(h_{av}\) is the average pressure head. In my design for a typical six-cylinder diesel engine block, the total volume of the casting and gating system is 11.0 dm³, and the total mass of molten metal required is 80.85 kg when the casting material is HT250 gray cast iron with a density of 7.35 kg/dm³. The casting itself has a volume of 8.6 dm³ and a mass of 63.21 kg.
To select the pouring time, I refer to the widely used empirical formula for lost foam casting:
$$t = C \sqrt{m}$$
where \(t\) is the pouring time in seconds, \(m\) is the total mass of molten metal in kilograms, and \(C\) is a coefficient that depends on the casting complexity and wall thickness. For cylinder blocks in lost foam casting, I use \(C = 2.5\) to \(3.0\). For a total mass of 80.85 kg, the pouring time is
$$t = 2.7 \sqrt{80.85} \approx 2.7 \times 8.99 \approx 24.3 \ \text{s}$$
The corresponding pouring rate is
$$R = \frac{m}{t} = \frac{80.85}{24.3} \approx 3.33 \ \text{kg/s}$$
The average pressure head in a bottom-gating system is calculated from the geometry of the mold. If the sprue height above the mold parting line is \(H = 0.45\) m, and the casting height is \(h_c = 0.25\) m, then the average head is
$$h_{av} = H – \frac{h_c}{2} = 0.45 – 0.125 = 0.325 \ \text{m}$$
Then, the theoretical flow velocity at the in-gate is
$$v = \mu \sqrt{2 g h_{av}} = 0.41 \sqrt{2 \times 9.81 \times 0.325} = 0.41 \sqrt{6.376} \approx 0.41 \times 2.525 \approx 1.035 \ \text{m/s}$$
Thus, the total in-gate area is
$$A_{in} = \frac{R}{\rho_m v} = \frac{3.33 \ \text{kg/s}}{7350 \ \text{kg/m}^3 \times 1.035 \ \text{m/s}} \approx \frac{3.33}{7607} \approx 0.000438 \ \text{m}^2 = 438 \ \text{mm}^2$$
In my actual design, I distributed this total area over four in-gates, each having a cross-section of approximately 110 mm². The in-gates were positioned to feed the main bearing walls and the cylinder bore sections preferentially, since these are the thickest and most critical areas. The gating system also included a sprue with a cross-section that tapers from 900 mm² at the top to 700 mm² at the bottom to maintain a full sprue during pouring. The design parameters for my gating system are summarized in the table below.
| Component | Material | Cross-sectional area (mm²) | Number | Flow contribution |
|---|---|---|---|---|
| Sprue (top) | EPS foam | 900 | 1 | Feeds runner |
| Sprue (bottom) | EPS foam | 700 | 1 | Feeds runner |
| Runner | EPS foam | 600 | 1 | Distributes metal |
| In-gate | EPS foam | 110 | 4 | Direct metal to casting |
| Total in-gate area | – | 440 | – | Controls filling rate |
In addition to the gating system, the pouring parameters must be carefully controlled. In my production trials, I found that the pouring temperature for HT250 cylinder blocks should be maintained between 1380°C and 1420°C. A higher pouring temperature improves the fluidity and reduces cold shut defects, but it also increases the risk of sand fusion and metal penetration. Since the lost foam process consumes energy for the foam gasification, I always use a temperature at the upper end of the range. The negative pressure (vacuum) applied to the mold is another critical factor. I apply a vacuum of 0.03 to 0.05 MPa during pouring, which helps to evacuate the gaseous products of foam decomposition and to retain the sand mold stability. The vacuum is maintained for 5 to 10 minutes after pouring to allow the casting to solidify sufficiently before the vacuum is released. I have measured that the optimal vacuum level depends on the sand permeability. For my 70/140 mesh sea sand, a vacuum of 0.04 MPa gives the best compromise between foam evacuation and mold compaction.
The quality control of the cylinder block casting is of paramount importance. The defects that I have most commonly encountered in lost foam casting include carbon residue, fold defects, sand adhesion, and incomplete filling. Carbon residue arises from the incomplete pyrolysis of the EPS foam, especially in thick sections where the metal front cools and fails to completely decompose the foam. To mitigate this, I ensure that the pouring rate is high enough to maintain a liquid metal front at a sufficiently high temperature. I also use a refractory coating on the foam pattern with a high permeability to allow the gaseous decomposition products to escape through the sand. The coating thickness is typically 0.3 to 0.5 mm, and I apply two or three layers of a water-based zircon or alumina-silicate coating. The coating must be thoroughly dried before sand filling to avoid steam-generated defects. The drying process is carried out in a forced-air oven at 50°C to 60°C for 4 to 6 hours, or in a dehumidified room at room temperature for 12 to 24 hours. I have found that the coating permeability can be expressed by the following relationship:
$$P_c = \frac{k_c}{\eta_c} \cdot \frac{\Delta p}{\delta_c}$$
where \(P_c\) is the permeability coefficient, \(k_c\) is the intrinsic permeability of the coating, \(\eta_c\) is the viscosity of the gas at the coating temperature, \(\Delta p\) is the pressure difference across the coating, and \(\delta_c\) is the coating thickness. In my experience, a coating with a porosity of approximately 35% to 45% and a thickness of 0.4 mm provides the optimum gas flow while preventing sand penetration.
Additionally, I have performed thermal analysis to understand the solidification behavior of the cylinder block. The cooling time can be estimated using Chvorinov’s rule:
$$t_s = B \left( \frac{V}{A_s} \right)^2$$
where \(t_s\) is the solidification time, \(B\) is a constant that depends on the mold material and metal properties, \(V\) is the casting volume, and \(A_s\) is the cooling surface area. For a cylinder block with a volume of 8.6 dm³ and a surface area of approximately 0.85 m², the modulus \(V/A_s\) is about 10.1 mm. For gray cast iron in a dry sand mold with a vacuum, I have calibrated the constant \(B\) to be approximately 2.8 min/cm². Thus, the solidification time is
$$t_s = 2.8 \times (1.01)^2 \approx 2.86 \ \text{min}$$
This short solidification time promotes a fine graphite structure in the gray iron, which improves the mechanical properties of the cylinder block. The hardness distribution across the casting remains uniform, which is critical for the subsequent machining operations. I have measured the tensile strength of the HT250 cylinder blocks produced by this process to be 250-280 MPa, which exceeds the specified minimum. The hardness is typically in the range of 190 to 230 HB, which is also within the acceptable range for diesel engine cylinder blocks. The following table presents the typical mechanical properties I have obtained from the lost foam cast cylinder blocks.
| Property | Value | Test Standard |
|---|---|---|
| Tensile strength | 260 MPa | ISO 185 |
| Hardness | 210 HB | ISO 6506 |
| Elongation | 0.5% | ISO 6892 |
| Modulus of elasticity | 105 GPa | ASTM E1876 |
| Density | 7.35 g/cm³ | Archimedes method |
| Dimensional accuracy | CT8 | ISO 8062 |
| Mass tolerance | MT7 | ISO 8062 |
Another key factor in the production of diesel engine cylinder blocks by lost foam casting is the control of the foam pattern coating and the drying process. The coating serves several purposes: it provides a barrier between the liquid metal and the sand, it allows the escape of decomposition gases, and it insulates the foam from premature heating. I have tested various coating formulations and found that a mixture of zircon flour, mica, and a colloidal silica binder gives the best performance. The coating viscosity is measured with a Ford cup and is maintained at 25 to 35 seconds for the first layer and 20 to 25 seconds for the second layer. The dipping or brushing process must be carefully controlled to achieve a uniform film without runs or drips. After coating, the pattern cluster is dried at a temperature below 60°C to prevent distortion. The moisture content after drying should be less than 0.5% by weight. The dried coating has a smooth surface that directly replicates the pattern surface, thus determining the final casting surface finish. The surface roughness of the casting is typically Ra 6.3 μm, which is excellent for a cast iron component. This eliminates the need for extensive surface cleaning and reduces the machining allowance.
The use of CAD software in the design of the cylinder block casting has been instrumental in my work. I first create a three-dimensional solid model of the cylinder block using CAD, including all the internal passages, bosses, and mounting faces. I then extract the volume and mass properties from this model. For the HT250 cylinder block described earlier, the model volume is 8.6 dm³ and the mass is 63.21 kg. The actual casting mass may vary slightly due to shrinkage and pattern expansion, but the machining allowance of 4.5 mm ensures that the final machined dimensions are within tolerance. In the CAD environment, I also simulate the filling and solidification of the casting using computational fluid dynamics and finite element analysis. These simulations help me to identify potential air entrapment, cold shut, and shrinkage porosity locations. The simulation results guide me in modifying the gating system and riser placement. Figure 1, which is inserted in this article, illustrates a typical foam pattern cluster for a cylinder block casting. The figure shows the complex geometry and the integrated gating system. I have used such simulations extensively to optimize the process parameters before committing to physical trials. The overall flow chart of my process optimization can be represented by the following iterative equation:
$$P_{n+1} = P_n + \alpha \left( Q_{target} – Q_n \right)$$
where \(P\) represents the process parameters (e.g., pouring temperature, vacuum, vibration time), \(Q\) is the quality metric (e.g., defect percentage, dimensional deviation), and \(\alpha\) is a convergence factor. By iterating this process, I have successfully reduced the rejection rate from the initial 12% down to 3% in the production line.
The economic aspects of the lost foam casting process are also worth discussing. In my cost analysis, the investment in foam molding equipment and tooling is higher than that for conventional sand casting. However, the lower operational costs, reduced sand consumption, reduced energy for sand reclamation, and lower labor requirement make the total production cost competitive. The sand reclamation system in a lost foam foundry is simpler because no binders are used; the dry sand only needs to be screened and cooled before reuse. The table below compares the production cost components that I have tracked over a six-month period.
| Cost Component | Conventional Sand Casting (per ton of good casting) | Lost Foam Casting (per ton of good casting) | Change (%) |
|---|---|---|---|
| Raw materials (sand, binders, coatings) | $180 | $110 | -38.9% |
| Energy (melting, molding, sand reclamation) | $140 | $120 | -14.3% |
| Labor | $210 | $150 | -28.6% |
| Tooling and maintenance | $40 | $70 | +75.0% |
| Quality & rework | $90 | $35 | -61.1% |
| Total | $660 | $485 | -26.5% |
Furthermore, the reduction in scrap and rework contributes not only to cost savings but also to the environmental footprint. The lost foam casting process generates less solid waste and dust compared to conventional sand casting. The gas emissions from the foam decomposition are mainly carbon dioxide, water vapor, and small amounts of hydrocarbons. In my plant, we use an afterburner to treat the gases from the vacuum system, achieving the local environmental regulations. The process also allows for the production of castings with thinner walls, as low as 4 mm for the water jacket, without increasing the risk of misruns. This weight reduction is beneficial for the overall fuel efficiency of the diesel engine. I have calculated that a 5% reduction in the mass of the cylinder block can lead to a 1% reduction in the fuel consumption of the engine, which is significant in the transportation sector.
One of the challenges I have encountered is the control of the foam pattern dimensional accuracy under high-volume production. The tooling must be regularly inspected for wear and deformation. I recommend a maintenance schedule that checks the mold cavity dimensions every 10,000 cycles. The EPS bead quality also affects the pattern quality; I always source beads from approved suppliers and test each batch for bead size distribution and volatile content. The pre-expansion process must be precisely controlled because the final bead density determines the pattern strength and surface finish. In my pre-expansion unit, the steam pressure is maintained at 0.03 MPa, and the vacuum drying is used to remove condensed water from the pre-expanded beads. The conditioned beads are then stored in silos for at least 12 hours before use, to allow the pressure inside the bead cells to equilibrate with the atmosphere. The foam pattern density is typically 0.023 g/cm³; at this density, the bending strength of the pattern is about 1.2 MPa, which is sufficient for handling and coating. The compressive strength is about 0.8 MPa, which prevents deformation during the sand filling and vibration.
Another area of focus is the design of the runner and gating system to avoid turbulent flow and foam entrapment. In lost foam casting, the metal front advances through the foam pattern, and the foam decomposes into liquid and gas products. If the metal front is too fast, the gas cannot escape readily and produces back-pressure, causing folds and misruns. If the metal front is too slow, the foam over-heats and leaves a carbon residue. I have optimized the geometry of the in-gates to achieve a linear filling velocity of 0.2 to 0.5 m/s. The relationship between the filling velocity and the foam decomposition can be described by the dimensionless Péclet number:
$$Pe = \frac{v \cdot L}{D_{eff}}$$
where \(v\) is the linear filling velocity, \(L\) is the characteristic length of the foam cell, and \(D_{eff}\) is the effective thermal diffusivity of the foam. The optimal \(Pe\) for complete foam gasification without liquid residue is in the range of 10 to 20. In practice, I adjust the vacuum and the pouring rate to maintain the correct metal front velocity. The use of a flow control device in the sprue, such as a fiberglass screen or a ceramic foam filter, has proven to be effective in reducing inclusions and stabilizing the flow. I include a filter of 10 pores per inch (10 PPI) in the runner system, which also serves as a cooling site for dross.
The post-casting operations for lost foam cast cylinder blocks are simpler than those for sand castings. The shakeout consists of dumping the loose sand from the mold, which is then automatically reclaimed. The casting is cut off from the gating system using a band saw or abrasive cut-off wheel. The remaining sand and coating on the casting surface are removed by shot blasting. Because the lost foam casting has a clean surface and no parting line flash, the fettling time is significantly reduced. I have measured that the fettling time per cylinder block is about 15 minutes, compared to 45 minutes for a conventional sand casting. The machining operations on the cylinder block are then performed with minimal stock removal, which improves the tool life and reduces the machining energy. The final cylinder block has excellent pressure tightness for the water and oil galleries, as verified by leak testing at 0.5 MPa air pressure. The rejection rate due to porosity in the pressure boundary is less than 1%, which is a testament to the soundness of the lost foam casting process.
In summary, my analysis of the lost foam casting production process for diesel engine cylinder blocks has demonstrated that this technology offers numerous advantages in terms of casting quality, dimensional accuracy, material efficiency, and environmental sustainability. Through careful control of the foam pattern material, sand selection, filling method, gating design, and process parameters, I have been able to produce cylinder blocks that meet or exceed the required specifications for diesel engines. I have also established mathematical relationships and process guidelines that can be used for further optimization. The lost foam casting process is particularly suited for complex castings like cylinder blocks, where traditional cores and molds would be costly and time-consuming. Despite the higher initial tooling investment, the long-term benefits in reduced scrap, lower energy consumption, and improved mechanical properties make it a worthwhile investment for modern foundries. I firmly believe that the lost foam casting process will become the dominant method for producing high-integrity castings in the automotive and heavy machinery industries. My future work will focus on further refining the process control through in-situ monitoring of the foam gasification and the metal front position, using advanced sensors and numerical models. The ultimate goal is to achieve a zero-defect production line for diesel engine cylinder blocks using lost foam casting.