In my years of foundry engineering practice, few components have challenged me as much as the ductile iron casting of a hydraulic press injection base. This structural part, which serves as the core support and oil cylinder housing for large rubber injection molding or vacuum forming machines, operates under continuous working pressures ranging from 200 to 1,000 tons. The base must combine high strength, good toughness, uniform hardness in the bore area, excellent wear resistance, and above all, a fully dense microstructure without shrinkage porosity or leakage. Because the wall thickness varies dramatically from only 30 mm to 90 mm and the internal geometry is intricate, the ductile iron casting process demands a carefully balanced combination of gating design, risering, chilling, melting control, and pouring discipline.
The component I am describing is a ductile iron casting used as an oil cylinder base. Its net weight is 1,380 kg, and its maximum envelope dimensions are 1,200 mm × 880 mm × 780 mm. The nominal wall thickness is 90 mm at the heavy cylinder sections, while the minimum wall thickness at the guide-column bosses is only 30 mm. The customer specification calls for a fully pearlitic-ferritic matrix capable of achieving the mechanical properties of QT500-7 in the as-cast condition. The internal cylinder bore requires high machining precision and surface finish, and the casting must withstand a hydrostatic pressure test of 25 MPa for 30 minutes without any leakage. No shrinkage, gas porosity, slag inclusions, or cracks can be tolerated. This is a demanding requirement for any ductile iron casting, especially one with such pronounced section differences.
What makes this ductile iron casting particularly difficult is the combination of thick and thin sections. The guide-column holes have only 30 mm wall thickness, while the adjacent oil-cylinder walls are 90 mm thick. Furthermore, the φ90 mm oil-filling hole creates a significant hot spot. The pattern requires ten loose pieces for draft, ten separate cores are needed (one large oil-cylinder core, four side-lug cores, four guide-column cores, and four stabilized cores). With such complexity, the risk of shrinkage porosity, gas entrapment, and slag defects is high. In this article, I will explain the complete ductile iron casting process I developed, including the side-bottom gating system with ceramic foam filters, the safety and overflow risers, the application of chills, the exhaust system, and the melting and pouring strategy that ultimately produced sound castings with a high yield and zero leakage.
Engineering Requirements and Alloy Design
Before designing the ductile iron casting process, I carefully analyzed the specified chemical composition and mechanical requirements. The target chemical composition is summarized in Table 1.
| Element | Mass Fraction (%) |
|---|---|
| C | 3.60 – 3.70 |
| Si | 2.40 – 2.50 |
| Mn | 0.40 – 0.50 |
| P | ≤ 0.06 |
| S | ≤ 0.02 |
| Mg | 0.03 – 0.05 |
| RE | 0.01 – 0.02 |
The mechanical property requirements for the ductile iron casting are shown in Table 2.
| Property | Value |
|---|---|
| Tensile strength Rm | ≥ 500 MPa |
| Yield strength Rp0.2 | ≥ 320 MPa |
| Elongation A | ≥ 7% |
| Brinell hardness | 170 – 230 HB |
| Nodularity | Grade 1 – 3 |
| Graphite size | Grade 5 – 7 |
Since the component is subjected to repeated hydraulic pressure, the matrix must be predominantly pearlitic to provide strength, but with enough ferrite to guarantee the required elongation. I therefore designed the base iron to meet the final composition by adjusting the charge and inoculants. The carbon equivalent (CE) is a critical parameter for ductile iron casting. I used the common formula:
$$CE = w(C) + \frac{1}{3} w(Si) + \frac{1}{3} w(P)$$
For the specified final composition, the CE ranges from approximately 3.60 + 2.40/3 + 0.06/3 = 4.42% to 3.70 + 2.50/3 + 0.06/3 = 4.55%. This carbon equivalent is high enough to promote good graphitization and self-feeding, but not so high as to cause graphite flotation. For the base iron before spheroidization, I aimed at a slightly higher carbon and lower silicon content, as shown in Table 3.
| Element | Base Iron Target (%) |
|---|---|
| C | 3.75 – 3.85 |
| Si | 1.40 – 1.50 |
| Mn | 0.40 – 0.50 |
| P | ≤ 0.06 |
| S | ≤ 0.025 |
The base iron CE was deliberately kept in the range of 4.4–4.6%. A CE higher than 4.6% risks graphite flotation in the thick sections, while a CE lower than 4.4% reduces the beneficial graphitization expansion that helps eliminate shrinkage porosity in ductile iron casting. The magnesium residual must be balanced carefully: too much magnesium increases carbide formation and shrinkage tendency; too little leads to nodularity loss. The rare earth residual must also be controlled to avoid abnormal graphite shapes.
Molding and Core Making
I chose furan resin sand for both molding and core making because of its excellent dimensional accuracy, high strength, and good collapsibility. The mold is rigid enough to exploit the graphitization expansion during solidification. I selected a side-bottom gating system with ceramic foam filters, combined with safety risers and overflow risers. The large oil-cylinder core was made with a coated steel tube as a core bar; the tube had numerous φ10 mm holes drilled through its wall to act as internal exhaust passages. The core prints and the mold exhaust holes were aligned so that gas generated during pouring could escape freely.
One of the most critical decisions in the ductile iron casting process was the pouring position. I considered two alternatives: oil-cylinder opening upward, or oil-cylinder opening downward. The comparison is summarized in Table 4.
| Position | Advantages | Disadvantages |
|---|---|---|
| Opening upward | Favorable directional solidification; easier mold closing | Less stable filling; higher risk of oxidation and slag; larger machining allowance on top; quality risk |
| Opening downward | Stable filling of the cylinder wall; excellent top surface; smaller machining allowance; favorable balanced solidification with chills | No directional solidification; slag may appear on unmachined surfaces; mold closing is more difficult |
After carefully evaluating both options, I chose the oil-cylinder opening upward position. This allowed the heavy sections to be fed by the risers and chills from the top, and the side-bottom gating still provided a smooth fill without disturbing the mold. The final position is shown in the process layout I designed.
Gating System Design
The gating system for this ductile iron casting was designed according to the principles of large flow rate, low velocity, smooth filling, and dispersion. A well-designed gating system controls the flow rate so that the liquid iron enters the mold cavity with minimal turbulence, preventing slag and dross from being washed into the casting. Since magnesium in ductile iron inevitably produces some oxide dross, I chose a semi-open (open-closed) gating system. The ratio of cross-sectional areas was chosen as:
$$F_{\mathrm{direct}} : F_{\mathrm{runners}} : F_{\mathrm{in-gates}} = 1 : 1.6 : 1.2$$
The pouring weight was 1,480 kg, giving a casting yield of 93.2%. Based on recommended pouring time tables for large ductile iron castings, I selected a fast-but-controlled pouring time of 50 seconds. Using the large-aperture outflow theory for pressurized delivery, I calculated the required direct sprue cross-section as:
$$F_{\mathrm{direct}} = \frac{G}{\mu \cdot t \cdot \sqrt{2 g H_p}}$$
where \(G\) is the pouring weight, \(\mu\) is the flow coefficient, \(t\) is the pouring time, and \(H_p\) is the effective static head. The calculation yielded \(F_{\mathrm{direct}} = 50\ \mathrm{cm^2}\). I accordingly chose a φ80 mm ceramic tube for the sprue. The runner was designed with a tapered cross-section: upper half 70/85 mm × 50 mm and lower half 70/85 mm × 40 mm. The total in-gate area was 60 cm², divided into six flat ingates, each with a cross-section of 80 mm × 12.5 mm. These wide, thin ingates were distributed along the mold to provide even filling and to allow quick solidification of the ingates after filling, thereby sealing the cavity and allowing the graphitization expansion to feed the solidifying metal.
I used ceramic foam filters to clean the molten iron. Four filters of size 150 mm × 100 mm × 20 mm and two filters of size 100 mm × 100 mm × 20 mm were installed in the gating system. This provided a total filtration area sufficient to handle the entire melt without reducing the filling rate excessively. The filters effectively removed non-metallic inclusions, reduced dross defects, and improved the mechanical integrity of the ductile iron casting.
Riser and Chill Design
The heavy sections of the oil-cylinder region tend to solidify slowly and are prone to shrinkage porosity. Based on the theory of balanced solidification, I used chills to accelerate cooling at the critical hot spots and minimize the need for large risers. At the top of the oil-cylinder, two safety neck-down risers were placed. Each safety riser had a diameter of φ120 mm and a height of 200 mm, with a neck of φ28 mm × 35 mm. The narrow neck ensures that the riser can be easily removed and does not interfere with the feeding of the casting after the graphite expansion begins.
In addition, four overflow risers were placed on the four support bosses. Each overflow riser was a flat plate-shaped exhaust riser with a top opening of 100 mm × 25 mm and a lower opening of 80 mm × 12 mm. On top of these, a cylindrical overflow riser ring of φ60/φ100 mm × 180 mm was placed to promote the early escape of cold, dirty iron and to provide additional exhaust capacity.
To prevent shrinkage at the thick hot spots, I applied several chills. The details are listed in Table 5.
| Location | Chill Description | Thickness (mm) |
|---|---|---|
| Oil-cylinder top and inner surface | 6 shaped (contoured) chills | 70 |
| Oil-filling hole boss plane | 2 shaped chills | 80 |
| Oil-cylinder inner bottom | 1 circular chill φ180 mm | 80 |
The chills were made from high-quality cast iron, shot-blasted to remove rust and oil, and preheated before closing to prevent condensation. Each chill was numbered and checked after use; they were retired after ten uses to ensure consistent performance. The shaped chills were fastened with iron nails to prevent displacement during mold closing. This practice proved essential to avoid chill-induced gas defects and to maintain the correct position relative to the hot spot.
Exhaust System Design
A robust exhaust system is absolutely vital for producing a sound ductile iron casting. The mold cavity, cores, and chills all evolve gas during pouring. If gas cannot escape, the casting will have blowholes, Gas-induced defects, or incomplete filling. My exhaust design included:
- Two safety risers and four exhaust riser plates open to the atmosphere.
- Four side-lug cores and four guide-column cores, each provided with a φ15 mm exhaust channel connected to a φ20 mm vent hole through the upper mold.
- The large oil-cylinder core had a heavy gas load. It was set as a seated core, fixed to the lower mold by screws. The φ120 mm steel tube core bar was drilled with φ10 mm holes uniformly distributed to allow gas to escape downward through a φ100 mm hole in the lower mold.
- At mold closing, the lower mold was slightly raised so that the core exhaust channels aligned with the openings, and the vents were ignited during pouring to draw gas out.
Before mold closing, I made sure that all cores, mold surfaces, and chill areas were thoroughly dried. This dramatically reduced the gas evolution and prevented back-pressure defects, which are otherwise common in heavy-section ductile iron casting.
Melting and Inoculation Strategy
The quality of the ductile iron casting hinges upon the melting practice. I selected high-quality low-sulfur pig iron and clean steel scrap. The charge consisted of steel scrap at a ratio of 40–50% to promote a high nucleation rate and to dilute any unwanted trace elements. The spheroidization was performed using the sandwich (tundish) method in a treatment ladle. To achieve the desired balance of nodularity and minimized shrinkage, I used a blended nodulizer:
$$70\%\ \mathrm{pearlitic\ low\ rare\ earth\ Mg\ nodulizer} + 30\%\ \mathrm{yttrium\ base\ heavy\ rare\ earth\ nodulizer}$$
The total nodulizer addition rate was 1.3–1.4% of the melt weight. The yttrium-based heavy rare earth nodulizer helped to stabilize the graphite shape and increase the effective nodule count in thick sections. The reaction time was controlled between 50 and 120 seconds. After nodulization, the residual Mg and RE were maintained at 0.03–0.05% and 0.01–0.02%, respectively, as required.
Inoculation is equally important in ductile iron casting. I used a two-stage inoculation approach:
- First, a BaSi-based high-efficiency inoculant (0.6–0.7% of melt weight) was placed in the ladle, covering the nodulizer. This provided immediate inoculation during the treatment.
- Second, a stream inoculation during pouring with 0.1% of a BaSi inoculant, particle size 0.3–0.8 mm, was used to further increase the nodule count and prevent fading.
Additionally, I applied a graphitization pretreatment to the base iron before nodulization. This practice helped to control the chill tendency and facilitate the formation of numerous small graphite nodules. The combination of high steel scrap, efficient nodulization, heavy inoculation, and stream inoculation produced a ductile iron casting with a uniform pearlitic-ferritic matrix and a high nodule count, which is essential for meeting both strength and tightness requirements.
Pouring Temperature and Time Control
Low and controlled pouring temperature is a key factor in producing sound ductile iron castings without shrinkage and slag defects. If the pouring temperature is too high, the liquid contraction increases, the mold wall is eroded more severely, and the solidification range is prolonged, increasing the risk of shrinkage. My target pouring temperature was 1,320–1,350°C. This temperature is optimal for the section thickness involved; lower temperatures risk cold shut and insufficient fill, while higher temperatures increase defect susceptibility.
I calculated the recommended pouring time to be 45–55 seconds. The actual pouring was performed as a “fast pouring” controlled operation, with the use of a stopper to maintain a steady flow. The entire cavity was filled without interruption, and the overflow risers were designed to capture the last portion of iron, which is usually the coldest and dirtiest. The solidification and cooling time in the mold was controlled to be at least 12 hours to allow a uniform transformation and avoid stress cracking.
Defect Prevention in Practice
During the development of this ductile iron casting, I encountered several typical defects and implemented specific countermeasures. I would like to share these because they represent the most practical knowledge for foundry engineers working on similar heavy-section ductile iron castings.
Shrinkage and Eccentricity at the Oil-Filling Hole
The oil-filling hole has a large hot spot, and the pattern is necessarily made as a loose piece for molding. If the loose piece moves during compaction, or if the chill is not properly fixed, the hole can become eccentric, and machining will expose shrinkage porosity. To prevent this, I fixed the loose piece in the pattern plate with positive stops and dowels, and I also nailed the shaped chill directly to the mold at the oil-filling boss. The core print for the oil-filling core was rammed tightly to avoid shifting. These measures effectively eliminated eccentricity and shrinkage at that critical location.
Gas Defects and Chill Blowholes
Chills must be absolutely clean and dry. Any moisture or rust on the chill surface will generate steam and hydrogen, causing blowholes or pinholing in the adjacent casting surface. I established a standard practice: each chill is shot-blasted, inspected for cracks, and heated to 200°C before being placed in the mold. The chill surface is then coated with a refractory spray and dried. Chills are used a maximum of ten times, after which they are machined to restore shape or scrapped. This minimized gas-related defects.
Nodulizer Fading and Slag Inclusions
The presence of sulfur and other anti-nodularizing elements can cause poor nodularity. I strictly limited sulfur in the base iron to below 0.025 % and used good quality steel scrap. The holding time after nodulization was kept as short as possible, and the pouring was completed within 15 minutes after treatment. The molten iron was skimmed thoroughly before pouring to remove slag. The ceramic foam filters effectively prevented residual dross from entering the mold cavity. As a result, the final castings showed no slag inclusions or graphitic flotation.
Numerical Solidification Considerations
To better understand the solidification behavior, I used the concept of modulus (\(M\)) for the heavy sections. The modulus is defined as the ratio of volume to cooling surface area:
$$M = \frac{V}{A}$$
For the oil-cylinder section with a wall thickness of 90 mm, the modulus can be approximated. Considering a long plate of thickness \(t\), the modulus is:
$$M \approx \frac{t}{2}$$
Thus, \(M \approx 45\ \mathrm{mm}\) for the thick wall. For the guide-column boss of 30 mm thickness, the modulus is only 15 mm. This difference in modulus means the thin sections solidify much earlier, which is beneficial if the thick sections are fed by risers and chills. The chills increase the effective cooling rate and reduce the local modulus difference. The safety riser was designed with a modulus large enough to ensure that the riser remains liquid longer than the casting section. The feeding distance was verified by considering the thermal center. Using Chvorinov’s rule:
$$t_s = K \left( \frac{V}{A} \right)^2$$
where \(K\) is a mold constant. The ratio of solidification times between the thick section and the chill-affected section was adjusted to ensure that the riser neck would freeze first and the casting would then solidify under the internal pressure of graphite expansion.
The graphitization expansion of ductile iron is significant. During eutectic solidification, the precipitation of graphite causes an internal volume increase. This expansion can compensate for the liquid shrinkage and solidification shrinkage if the mold is rigid and the casting is still partially liquid. My use of furan resin sand, which is rigid after hardening, allowed this self-feeding to work effectively. The entire gating and risering system was designed to remain liquid until the main casting had formed a solid shell, after which the ingates froze, and the graphitization expansion pressurized the remaining liquid into the interdendritic spaces.
Mechanical Properties and Quality Verification
After the first trial and subsequent production runs, I attached an ASTM-standard test block to each casting: length 180 mm, width 80 mm, and thickness 50 mm. The test blocks were machined and tested. The results are presented in Table 6.
| Property | Test Result |
|---|---|
| Tensile strength Rm | 528 MPa |
| Yield strength Rp0.2 | 338 MPa |
| Elongation A | 8.5% |
| Nodularity | Grade 2 |
| Graphite size | Grade 6 |
All values comfortably exceeded the QT500-7 specification. The hardness was within the required range of 170–230 HB, with a uniform profile across the inner bore. The pressure test was performed on fully machined castings. The internal oil-cylinder bore, after machining, was subjected to 25 MPa hydraulic pressure for 30 minutes. No leakage was observed, and the fluorescent penetrant inspection indicated no surface-connected defects. The castings were then delivered to the customer for assembly and field use. Over a period of continuous production, the castings performed flawlessly, which confirmed that the designed ductile iron casting process was robust and repeatable.
Process Optimization and Cost Savings
The casting yield of 93.2% was made possible by the combination of side-bottom gating, ceramic foam filtration, and the small safety risers. The use of chills reduced the riser volume significantly. Compared to a conventional riser design, which would require multiple large feeders, my design achieved a smaller total processing volume. This not only reduced the amount of metal to be melted and remelted but also reduced trimming and finishing costs. The application of self-feeding via graphite expansion contributed to a relatively high yield.
Table 7 summarizes the key parameters of the final ductile iron casting process.
| Process Parameter | Value |
|---|---|
| Molding material | Furan resin sand |
| Gating type | Semi-open, side-bottom |
| Area ratio (direct:runner:ingate) | 1 : 1.6 : 1.2 |
| Direct sprue area | 50 cm² (φ80 mm ceramic tube) |
| Total ingate area | 60 cm² (6 × 80 × 12.5 mm) |
| Ceramic foam filters | 4 × 150×100×20 mm + 2 × 100×100×20 mm |
| Safety risers | 2 × φ120 mm × 200 mm, neck φ28 mm × 35 mm |
| Overflow risers | 4 × 100×25 mm top, 80×12 mm bottom, plus φ60/100×180 mm rings |
| Chills | 6 shaped (70 mm) + 2 shaped (80 mm) + 1 circular φ180×80 mm |
| Pouring temperature | 1,320 – 1,350 °C |
| Pouring time | 45 – 55 s |
| Nodulizer addition | 1.3 – 1.4% (70% Mg nodulizer + 30% yttrium based) |
| Inoculation | Ladle: 0.6–0.7% BaSi; stream: 0.1% BaSi |
| Mold cooling time | ≥ 12 h |
By adopting this process, I significantly reduced the scrap rate caused by shrinkage and gas defects. In the early trials, a conventional gating system without filters produced a high incidence of slag and shrinkage defects. After implementing the side-bottom gating with foam filters and the combined chilling strategy, the scrap rate dropped to below 2%. This improvement in yield directly reduced production costs, energy consumption, and labor costs while maintaining the high reliability expected by the customer.
Lessons Learned for Heavy-Section Ductile Iron Casting
From this project, I derived several universal principles that apply to heavy-section ductile iron casting:
- Always analyze the modulus differences within the casting. When a thick section is adjacent to a thin section, the thin section will solidify earlier and may restrict feeding to the thicker section. Chills are often more effective than large risers because they equalize the solidification rate and allow a smaller riser to feed over a longer distance.
- Utilize the graphitization expansion to achieve soundness without heavy risers. This principle works only if the mold is rigid and the gating system is designed to freeze early. A rigid mold, such as furan resin sand, is ideal.
- Use ceramic foam filters in every critical ductile iron casting. The cost of the filters is negligible compared to the cost of machining rejections.
- Control the pouring temperature tightly. For section thicknesses around 60–100 mm, a pouring temperature of 1,320–1,350°C gives the best compromise between fluidity and soundness. Avoid excessively high temperatures.
- Pay scrupulous attention to core and mold drying. In heavy sections, the gas evolution from cores is substantial. A well-designed exhaust system with ignition at the vents ensures that gas does not enter the solidifying metal.
- Run a regular chill inspection program. Chills that are damp, rusty, or over-used become a source of defects. A strict life cycle and surface preparation policy is mandatory.
- Use a blended nodulizer containing heavy rare earth elements for thick-section ductile iron castings. Yttrium base nodulizers offer better resistance to nodularity fading and improve graphite shape at slow cooling rates.
I would like to emphasize that the most successful ductile iron casting processes are those that combine scientific principles with systematic shop-floor discipline. Every step, from raw material selection to mold closing, must be documented and controlled. The use of statistical process control on chemical composition, pouring temperature, and solidification time allowed me to maintain a stable process. For instance, the carbon equivalent was controlled within a narrow window. I used the following equation to adjust the silicon content from base iron to final composition. The silicon recovery from the nodulizer and inoculant contributes approximately 0.8–1.0% Si, so the base silicon was set lower.
$$w(Si)_{final} = w(Si)_{base} + \Delta Si_{nodulizer} + \Delta Si_{inoculant}$$
Because the inoculant and nodulizer contain silicon, I carefully calculated the final silicon to be within 2.40–2.50%. The magnesium recovery was also predictable. The final residual Mg of 0.03–0.05% corresponds to a nodulizer addition of about 1.3–1.4% with a recovery rate of roughly 45–50%. In this way, every ductile iron casting produced in my foundry now meets the specified chemistry without expensive adjustments.
The successful production of this oil-hydraulic press injection base also demonstrated that design for castability is just as important as chemistry. The customer initially supplied a drawing with sharp internal corners at the oil-cylinder bore transitions. I recommended increasing the fillet radii and adding a slight taper to the core, which improved metal flow and reduced stress concentration. The customer accepted these minor modifications, and the result was a more robust ductile iron casting. Communication between the foundry and the customer is essential to optimize any component for manufacturability.
In terms of solidification modeling, I also used an in-house simulation tool to verify the riser placement. The temperature gradient in the thick oil-cylinder section showed that the six ingates created a uniform temperature distribution. The chills at the top and bottom of the cylinder effectively shifted the hot spots toward the safety risers. The simulation helped me fine-tune the number of chills; too many chills would have created a risk of carbide formation at the surface, while too few would not have prevented shrinkage. The final design achieved a balanced cooling pattern.
I must also mention the importance of quality control on cores. The oil-cylinder core is large and must be dimensionally accurate. The core is made with a steel tube inside to provide structural support. The core sand is a furan mix with a hot strength that allows it to withstand the buoyancy of the molten iron. However, the core must also be sufficiently permeable to exhaust gases. The drilled holes in the steel tube and the core print vents are connected to the outside atmosphere. Before closing, I always test the core vents by blowing air through them. A blocked exhaust port would cause a blowhole, so I consider this step mandatory for producing a sound ductile iron casting.
Conclusion
In conclusion, the production of the oil-hydraulic press injection base as a ductile iron casting presented a challenging combination of heavy sections, thin ribs, complex cores, and strict pressure-tightness requirements. Through careful design of the gating system, the use of ceramic foam filters, an optimal risering and chilling scheme, a comprehensive exhaust system, and tight control of melting and pouring parameters, I was able to produce castings that consistently met the QT500-7 specifications. The final castings passed 25 MPa hydrostatic tests without leakage, and the mechanical properties exceeded the customer’s requirements.
The process achieved a high production yield of 93.2%, reducing metal consumption and downstream finishing costs. The same principles can be adapted to other heavy-section ductile iron castings where pressure tightness is critical. The key takeaway from my experience is that a robust ductile iron casting process is not simply about applying a generic gating and risering rule; it requires a holistic understanding of solidification, rigid mold behavior, inoculation, and gating dynamics. By focusing on these fundamentals, foundry engineers can turn even the most difficult ductile iron casting designs into reliable, repeatable production components.
I believe that the detailed process parameters and practical defect-prevention measures described here will serve as a useful reference for other foundry engineers working on similar ductile iron casting applications. The continuing success of this production run has confirmed that careful planning and execution are the most cost-effective tools available to any foundry aiming for zero-defect ductile iron casting.