In my years of work with heavy-section castings, I have learned that a successful ductile iron casting is not simply the result of pouring good molten metal into a mold. It depends on the interaction among mold rigidity, cooling rate, feeding behavior, nucleation control, and inoculation practice. This article describes a systematic development route for a high-pressure hydraulic cylinder body produced as a ductile iron casting. The component is used in the clamping system of an injection molding machine and has to withstand an internal oil pressure of 20 MPa without leaking or sweating. This is a demanding application for ductile iron, because the section thickness is large, the cooling rate is low, and the tendency for shrinkage porosity, graphite degeneration, and nonmetallic inclusions is high.
Ductile iron casting combines the process flexibility of gray iron with mechanical properties approaching those of steel. It is therefore widely used for hydraulic pump bodies, valve bodies, cylinder blocks, and large machine frames. In this particular case, the high-pressure hydraulic cylinder body has an outer diameter of 990 mm, a height of 625 mm, a mass of 1.3 t, and a maximum section thickness of 135 mm. The hydraulic chamber is machined to a surface roughness of Ra0.8 μm and must remain leak-tight at a working pressure of 20 MPa. The traditional ductile iron casting process often used a sand core made of chromite sand, special shaped chills at the hot spots, and large top risers. Although that approach could produce acceptable castings, the process yield was low, the amount of finishing work was heavy, and the reject rate remained high because of slag inclusions, gas defects, and internal unsoundness.
The purpose of this work was to develop a more robust ductile iron casting process for this hydraulic cylinder body. I focused on four main aspects: high-stiffness mold construction, a controlled gating system, a preconditioned and well-treated melt, and a microstructure optimized by silicon carbide additions and microalloying with antimony. The final process produced castings that met all ultrasonic and magnetic particle inspection requirements, satisfied the 20 MPa hydraulic test, and achieved the expected mechanical properties. In this article, I will present the process design, the metallurgical principles, and the validation results from the trial production of this ductile iron casting.
1. Design Challenges for Heavy-Section Ductile Iron Casting
Heavy-section ductile iron casting is inherently more difficult than casting small components because solidification time increases with section thickness. The slow cooling of a 135 mm wall section gives graphite sufficient time to diffuse and grow, but it also creates conditions for chunky graphite, graphite flotation, carbide segregation, and microshrinkage. The thick section also promotes mold dilation. If the mold wall moves outward during eutectic solidification, the volume increase caused by graphite precipitation is consumed by the mold instead of by compensating internal shrinkage. The result is a porous casting that may pass static pressure tests but fails under dynamic oil pressure or machining.
The solidification modulus of a casting is often used to evaluate the difficulty of feeding:
$$M = \frac{V}{A}$$
where V is the volume of the solidifying region and A is the cooling surface area. A large modulus means a long solidification time and a greater need for feed metal. For the hydraulic cylinder body, the large thickness at the junction between the hydraulic chamber and the flange creates a very high local modulus. In the old process, chills were placed at those hot spots to accelerate cooling. The use of many shaped chills, however, made the process complicated and the cooling effect was often uneven. In addition, the hydraulic chamber was usually cast with the cavity facing upward. This orientation trapped nonmetallic inclusions on the underside of the chamber bottom because those inclusions floated upward and then became embedded under the upper mold surface. After machining, those inclusions opened to the oil pressure and caused leakage.
The revised approach was therefore designed around the following principles:
First, the mold or core that forms the hydraulic chamber had to be extremely rigid and strong so that the casting would not swell during solidification. Second, a controlled cooling medium had to replace the traditional shaped chills in order to produce a firm and uniform chilling effect. Third, the gating system had to deliver clean iron smoothly into the thickest part of the casting without turbulence or vortex entrainment. Fourth, the melt had to be preconditioned with silicon carbide, spheroidized with a rare-earth magnesium nodulizer, microalloyed with a small amount of antimony, and inoculated at multiple stages so that the graphite would be fine, spherical, and uniformly distributed.
The table below summarizes the main component requirements that guided the process design.
| Parameter | Requirement |
|---|---|
| Component | High-pressure hydraulic cylinder body |
| Outer diameter / height | 990 mm / 625 mm |
| Mass | 1.3 t |
| Maximum section thickness | 135 mm |
| Hydraulic pressure | 20 MPa |
| Machined surface roughness | Ra 0.8 μm |
| Critical internal quality | No cracks, cold shuts, shrinkage, porosity, slag inclusions, or gas holes |
| Inspection | Ultrasonic testing and magnetic particle inspection |
2. High-Stiffness Mold Construction with Steel Pipe Reinforcement
In the conventional ductile iron casting process for this component, the sand core used to form the hydraulic chamber was made of chromite sand. Chromite sand has a high thermal conductivity and can act as a natural chill, but it is expensive and its performance depends on the quality of the sand core. When the core was not sufficiently compacted or when the binder degraded, the sand surface could be eroded by the flowing iron. This caused sand inclusions, gas generation, and a poor surface on the hydraulic chamber. The roughness of the chamber after machining was difficult to control, and leakage occurred in service.
To solve this problem, I designed a high-stiffness mold structure in which the hydraulic chamber is no longer formed by a separate conventional sand core. Instead, the chamber profile is formed by a compact mold assembly reinforced with steel pipes. The steel pipes are arranged along the contour of the hot zones and are packed with molding sand around them. The pipes serve two functions. First, they support the molding sand and strongly increase the rigidity of the mold. Second, they act as internal cooling channels. After pouring, the heat of the solidifying ductile iron casting is conducted from the mold sand to the steel pipe surface and then removed by air flowing through the pipe bore. This produces a strong, uniform, and controllable chilling effect without the need for dedicated metal chills.
The heat flow through the steel pipe system can be described by Newton’s law of cooling:
$$\dot{Q} = h A_{\mathrm{pipe}} \left(T_{\mathrm{surface}} – T_{\mathrm{air}}\right)$$
where h is the convective heat transfer coefficient between the inner pipe wall and the air, Apipe is the inner surface area of the pipe, Tsurface is the inner wall temperature, and Tair is the temperature of the air inside the pipe. If radiation is included, the equation becomes:
$$\dot{Q} = h A_{\mathrm{pipe}} \left(T_{\mathrm{surface}} – T_{\mathrm{air}}\right) + \epsilon \sigma A_{\mathrm{pipe}} \left(T_{\mathrm{surface}}^4 – T_{\mathrm{air}}^4\right)$$
In practice, the convective term dominates because the mold sand and pipe wall are not extremely hot for a long period, but the radiation term cannot be completely ignored during the first stage of cooling.
The high stiffness provided by the steel pipes is especially important for ductile iron casting. During eutectic solidification, graphite nodules precipitate and generate internal pressure. If the mold is rigid, this pressure can feed the solidification shrinkage. If the mold is soft, the mold wall moves away and the internal pressure drops. In that case, the shrinkage porosity develops and ductile iron castings leak under pressure. By reinforcing the sand mold with a steel pipe skeleton, we increased the mold stiffness and reduced the risk of mold dilation.
The effective solidification modulus of the casting after pipe cooling can be expressed as:
$$M_{\mathrm{eff}} = \frac{V}{A_{\mathrm{surface}} + \eta A_{\mathrm{pipe}}}$$
where η is a heat-transfer efficiency factor for the pipe, Asurface is the normal mold surface area in contact with the casting, and Apipe is the effective area of the steel pipe surface within the mold. Because Apipe increases the effective cooling area, Meff is reduced. The result is a faster solidification, finer graphite, and a denser matrix. This is a major advantage for high-pressure hydraulic components produced by ductile iron casting.
This new mold structure offers several important benefits when compared with the traditional process:
| Aspect | Traditional ductile iron casting process | Improved ductile iron casting process |
|---|---|---|
| Core material | Chromite sand core | Mold assembly reinforced with steel pipes |
| Chilling method | Dedicated shaped metal chills | Uniform cooling through steel pipe air channels |
| Rigidity of mold | Moderate | High |
| Risk of sand erosion | Higher | Lower |
| Process yield | Lower due to large top risers | Higher because feeding demand is reduced |
| Finishing work | More weld repair and grinding | Smoother surface and fewer defects |
The elimination of separate chromite-sand cores also reduces gas generation and lowers the cost of the ductile iron casting process. The molding sand only needs to fill the space between the steel pipes and the pattern, which reduces the total amount of sand used. The venting capacity of the mold is improved because the steel pipes provide natural channels for gas to escape. During pouring, the chance of erosion, sand drop, and gas entrainment is significantly reduced.
3. Gating System Design for Clean and Quiescent Filling
For a high-pressure ductile iron casting, the gating system must fulfill two contradictory requirements. It must fill the mold quickly enough to avoid cold shuts, but it must also fill gently enough to avoid turbulence, slag entrainment, and oxide film formation. In the improved process, I placed the ingates directly at a thick platform on the underside of the hydraulic chamber bottom. This location is one of the main hot spots in the casting. By introducing the iron directly into the thick section, we provide a steep thermal gradient and promote directional solidification toward the riser system. At the same time, the iron enters the cavity in a more stable way because it first travels through a series of transition runners.
The main elements of the gating system are the sprue, the first runner, the second runner, the first ingates, and the final ingates. The cross-sectional area ratios are critical in ductile iron casting because they determine the filling velocity and pressure distribution in the mold. The ratios used in this project were:
$$\Sigma S_{\mathrm{sprue}} : \Sigma S_{\mathrm{runner1}} : \Sigma S_{\mathrm{runner2}} : \Sigma S_{\mathrm{ingate1}} : \Sigma S_{\mathrm{ingate}} = 1.0 \sim 1.1 : 1.3 \sim 1.5 : 1.3 \sim 1.5 : 1.0 : 1.0$$
The actual ranges used in the production mold are shown in the following table.
| Gating element | Symbol | Relative total area range |
|---|---|---|
| Sprue exit area | ΣSsprue | 1.0 – 1.1 |
| First runner area | ΣSrunner1 | 1.3 – 1.5 |
| Second runner area | ΣSrunner2 | 1.3 – 1.5 |
| First ingate area | ΣSingate1 | 1.0 |
| Final ingate area | ΣSingate | 1.0 |
The average filling rate and the gate velocity can be estimated by using the basic orifice equation:
$$Q = \mu S_{\mathrm{choke}} \sqrt{2 g h_{\mathrm{p}}}$$
where μ is a flow coefficient, Schoke is the choke area, g is the gravitational acceleration, and hp is the effective metal head height. The average gate velocity is then:
$$v_{\mathrm{gate}} = \frac{Q}{\Sigma S_{\mathrm{ingate}}}$$
Because the total ingate area is larger than the runner areas in the downstream portion, the velocity is reduced before the iron enters the mold cavity. This is precisely what we want for a high-quality ductile iron casting. The lower filling velocity prevents the liquid metal from splashing and prevents the surface oxide film from being folded into the bulk metal.
I also placed a filter between the first transition runner and the second transition runner. The filter removes slag, dross, and nonmetallic inclusions before they can reach the ingates. The second transition runner was positioned 20 to 30 mm higher than the second runner. This height difference creates a favorable condition for slag flotation. Dense particles settle at the bottom of the runner, while the lighter slag particles rise into the transition runner and remain trapped there. This extra measure is very useful for ductile iron casting, because the spheroidization reaction always produces a certain amount of magnesium silicate and other reaction products.
A good gating system also limits the formation of entrained oxide films. In ductile iron casting, the main surface film is made of silicate. When the metal surface is disturbed, this film can be folded into the interior of the casting and cause cracks or leaks. By keeping the flow laminar or nearly laminar, the improved gating system greatly reduces this risk. The Reynolds number at the ingate can be estimated as:
$$\mathrm{Re} = \frac{\rho v D_{\mathrm{h}}}{\eta}$$
where ρ is the density of the liquid iron, v is the flow velocity, Dh is the hydraulic diameter of the ingate, and η is the dynamic viscosity. In the final design, the ingate velocity was kept low enough to maintain the Reynolds number below the critical value for severe surface turbulence.
4. Melting Procedure and Charge Composition
High-quality ductile iron casting requires reliable control of the base melt chemistry. The large thermal mass of the hydraulic cylinder body makes the melt particularly sensitive to spheroidization fade and inoculation fade. Therefore, I designed a melting procedure that started with a clean charge and included silicon carbide preconditioning.
The charge composition used for the production trial is shown in the table below.
| Charge material | Mass fraction / % |
|---|---|
| Pig iron | 50 |
| Steel scrap | 30 |
| Return scrap | 20 |
| Silicon carbide, percentage of metallic charge | 0.7 |
| Recarburizer, percentage of metallic charge | 0.8 |
All the silicon carbide, pig iron, steel scrap, and return scrap were charged into the melting furnace. The recarburizer was added once during the charging step. After the charge was fully melted, 0.6% FeSi75 silicon iron was added to adjust the silicon content. The molten iron was then superheated to 1452 °C. After superheating, the base melt composition was targeted as shown in the following table.
| Element | Base melt target / mass % |
|---|---|
| C | 3.75 |
| Si | 1.55 |
| Mn | 0.25 |
| P | ≤ 0.025 |
| S | ≤ 0.023 |
| Fe | Balance |
Silicon carbide plays an important role in ductile iron casting. SiC has a melting point well above the temperature of molten iron. When SiC is added to the furnace charge, it does not completely melt during the melting stage. Instead, it participates in a reaction that can be written as:
$$\mathrm{SiC} + \mathrm{Fe} \rightarrow \mathrm{FeSi} + \mathrm{C}$$
This reaction releases silicon and carbon atoms into the iron melt. Since silicon increases the carbon activity, the carbon equivalent of the melt is stabilized and the precipitation of graphite becomes more favorable. The undissolved SiC particles remaining in the melt act as heterogeneous nucleation substrates for graphite. This increases the graphite nodule count and reduces the tendency for carbide formation during solidification. Because the SiC particles are often covered by a thin SiO2 layer, the dissolution reaction is slowed down. This provides a prolonged preconditioning effect and prevents premature homogenization of carbon clusters in the liquid iron. The result is a more uniform graphite distribution in the final ductile iron casting.
After the base melt reached the desired temperature and composition, I performed spheroidization treatment using the sandwich method. The treatment ladle was prepared with a dam. On one side of the dam, the nodulizer and inoculant were placed in layers and compacted. The charge was added as follows:
First, 0.006% pure antimony, based on the mass of the original molten iron, was placed in the treatment pocket. The antimony content in the final ductile iron casting was about 0.0055%. Antimony is a powerful microalloying element in heavy-section ductile iron casting. It helps suppress chunky graphite formation, stabilizes carbides or pearlite in local regions, and improves the morphology of graphite nodules. At this low concentration, antimony does not cause unwanted carbides but improves the uniformity of the structure.
After the antimony, a granular inoculant with a particle size of 3 to 8 mm was added and compacted. The nodulizer used was a rare-earth magnesium alloy with the composition shown below. The nodulizer addition was 1.2% of the original melt mass, and the magnesium flash reaction lasted about 125 seconds. The inoculant was a barium-bearing silicon inoculant added at 0.62% of the original melt mass.
| Alloy | Mg / % | RE / % | Si / % | Ca / % | Ba / % | Al / % | S / % | Fe / % | Addition / % |
|---|---|---|---|---|---|---|---|---|---|
| Nodulizer | 5.8 | 1.49 | 42.8 | 2.45 | – | 0.85 | – | Balance | 1.20 |
| Inoculant | – | – | 72 | 1.0 | 2.0 | 0.70 | 0.015 | Balance | 0.62 |
After spheroidization, the melt was stirred, slagged, and allowed to stand for a short period so that the reaction products could separate from the melt. The final composition of the treated iron was:
| Element | Final melt / mass % |
|---|---|
| C | 3.55 |
| Si | 2.45 |
| Mn | 0.25 |
| P | ≤ 0.025 |
| S | 0.0098 |
| Mg | 0.037 |
| RE | 0.008 |
| Sb | 0.0055 |
| CE | 4.38 |
The carbon equivalent is calculated using the usual expression for ductile iron:
$$\mathrm{CE} = \mathrm{C} + \frac{1}{3}\left(\mathrm{Si} + \mathrm{P}\right)$$
By substituting the final values, we obtain:
$$\mathrm{CE} = 3.55 + \frac{1}{3}\left(2.45 + 0.025\right) = 4.38$$
The magnesium recovery can be estimated from the final magnesium content and the amount of magnesium added with the nodulizer:
$$\eta_{\mathrm{Mg}} = \frac{0.037}{1.2 \times 0.058} \times 100\% \approx 53\%$$
This magnesium recovery is reasonable for a sandwich treatment in a heavy-section ductile iron casting. The sulfur removal efficiency is also meaningful. The final sulfur content decreased from about 0.023% in the base melt to 0.0098% after spheroidization, which indicates that magnesium and rare earth reacted with sulfur and removed it from the melt. This desulfurization reaction can be written symbolically as:
$$\mathrm{Mg} + \mathrm{S} \rightarrow \mathrm{MgS}$$
Magnesium sulfide is less dense than the iron melt and therefore rises into the slag. This is one of the reasons why efficient slagging is so important after spheroidization.
The pouring temperature was fixed at 1295 °C. When the molten iron reached this temperature, it was poured into the prepared mold. During pouring, a fine powder inoculant was added to the metal stream as stream inoculation. The stream inoculation addition was 0.12% of the melt mass. This late inoculation is essential because it counteracts inoculation fade and ensures a large number of active nucleation particles are present at the moment of solidification.
5. Microstructure of the Cast-On Test Block
In order to evaluate the internal quality of the ductile iron casting, I examined a test block cast together with the hydraulic cylinder body. The test block was mounted on the casting and had dimensions of 70 mm × 105 mm × 210 mm. A sample was cut from the test block, ground, polished, and etched with 4% nital solution. The etched specimen was observed using scanning electron microscopy. The microstructure was mainly ferritic, with a uniform distribution of spherical graphite nodules. The nodularity was measured as 94.2%, and the graphite size was classified as grade 6. Locally, fine lamellar pearlite was present, which gave additional strength to the ductile iron casting without harming its elongation.
The EDS analysis of selected points on the test block is summarized in the table below. The values are expressed in mass percent.
| Test point | Si / % | C / % | O / % | Fe / % |
|---|---|---|---|---|
| Point 1 | 3.05 | 22.38 | 26.94 | 47.63 |
| Point 2 | 2.70 | 1.69 | 0.98 | 94.63 |
| Point 3 | 6.24 | 4.10 | 8.95 | 80.71 |
| Point 4 | 1.45 | 32.33 | 0.78 | 65.44 |
Point 1 and point 4 are clearly carbon-rich regions, which correspond to graphite nodules or graphite-like phases. Point 2 is a typical metallic matrix region with a high iron content and small amounts of silicon. Point 3 has an elevated silicon and oxygen content, which suggests the presence of a small silicate inclusion or an oxide particle. Although such particles are present, they are finely dispersed and do not impair the hydraulic integrity of the casting.
The nucleation of graphite in ductile iron casting is a key metallurgical phenomenon. Graphite precipitates when the local carbon concentration exceeds the solubility limit in austenite or liquid iron. The free energy change for the homogeneous nucleation of graphite can be written as:
$$\Delta G_{\mathrm{hom}} = -\frac{4}{3}\pi r^3 \Delta G_v + 4\pi r^2 \gamma_{\mathrm{sl}}$$
where ΔGv is the volume free energy change per unit volume, r is the radius of the nucleus, and γsl is the solid-liquid interfacial energy. The critical homogeneous nucleus radius is:
$$r^* = -\frac{2\gamma_{\mathrm{sl}}}{\Delta G_v}$$
Because ΔGv is negative below the equilibrium temperature, r* is positive. However, homogeneous nucleation is rare in liquid iron. In commercial ductile iron casting, graphite nucleates on heterogeneous substrates such as MgO, SiO2, SiC, sulfides, and other complex compounds. The energy barrier for heterogeneous nucleation is reduced by a factor f(θ):
$$\Delta G_{\mathrm{het}} = \Delta G_{\mathrm{hom}} f(\theta)$$
where θ is the wetting angle between the nucleus and the substrate and f(θ) is:
$$f(\theta) = \frac{\left(2 + \cos\theta\right)\left(1 – \cos\theta\right)^2}{4}$$
When the substrate is effective, the wetting angle is small, f(θ) is small, and the nucleation barrier is drastically reduced. The undissolved SiC particles and the reaction products formed during spheroidization provide exactly such effective substrates. That explains why the graphite nodule count in this ductile iron casting was high and why the graphite distribution was so uniform.
The use of silicon carbide is especially beneficial in heavy-section ductile iron casting. As I described earlier, the partial dissolution of silicon carbide releases carbon in a highly reactive form and leaves behind undissolved particles. This promotes a large number of graphite nuclei. The same effect also refines the pearlite because undissolved SiC particles can act as heterogeneous substrates for Fe3C. The ability of silicon carbide to promote multiple phases is very useful when a balance between strength and elongation is needed.
The addition of a small amount of antimony also contributes to the microstructure. Antimony concentrates at the graphite-matrix interface and prevents the growth of degenerate graphite. In heavy sections, chunky graphite often forms because the growth interface becomes unstable. A residual antimony content of about 0.0055% is sufficient to suppress this instability while keeping the material ductile. The final micrograph of the cast-on sample clearly showed that the graphite nodules were round, well dispersed, and surrounded by a ferritic matrix with only occasional fine pearlite patches.
6. Mechanical Properties and Quality Validation
The acceptance criteria for the hydraulic cylinder body ductile iron casting required a combination of tensile strength, yield strength, elongation, and hardness that would ensure reliable service under pressure. The requirements and the actual values obtained from the cast-on test block are given in the following table.
| Property | Required value | Measured value |
|---|---|---|
| Tensile strength / MPa | ≥ 390 | 405 |
| Yield strength / MPa | ≥ 260 | 278 |
| Elongation after fracture / % | ≥ 8.0 | 18.5 |
| Brinell hardness / HBW | 140 – 190 | 148 |
The measured tensile strength of 405 MPa, yield strength of 278 MPa, and elongation of 18.5% all exceeded the requirements. The high elongation demonstrates that the matrix is predominantly ferritic and that the graphite is well spheroidized. A ductile iron casting with such a high elongation value is less sensitive to local stress concentrations and is therefore more reliable for a high-pressure hydraulic component.
The hardness value of 148 HBW is in the middle of the specified range. This hardness is suitable for machining and gives a good balance between wear resistance and machinability. The machined surface of the hydraulic chamber achieved the required Ra0.8 μm finish without exposing pinholes or microshrinkage cavities. This confirms that the solidification structure is dense enough to be machined to a smooth surface.
The trial-produced castings were also subjected to non-destructive examination. The ultrasonic testing met the Level 1 acceptance requirement according to the applicable standard. The magnetic particle inspection met Level 2 acceptance. The hydraulic test at 20 MPa did not show any leakage, sweating, or seepage through the pressure chamber. This performance is the ultimate proof that the improved ductile iron casting process can produce sound, pressure-tight components.
I also observed that the process yield was noticeably higher than with the old process. The large top risers used in the conventional ductile iron casting method were reduced, and the amount of slag and sand defects was much lower. The amount of grinding and weld repair was significantly decreased. This is important not only for cost but also for process stability. A process that depends on heavy finishing operations after casting is inherently difficult to control. The new ductile iron casting process is much more reliable because the quality is built into the mold design and melting practice, not patched after solidification.
7. Discussion of the Key Process Factors
Several factors contributed to the success of this ductile iron casting. The first is mold rigidity. In ductile iron, the eutectic solidification involves both austenite and graphite. Graphite precipitation produces an internal expansion force. If the mold wall is not sufficiently rigid, this expansion is lost, and the feeding path closes prematurely. The steel-pipe skeleton in the mold provides a strong frame that resists dilation. As a result, the internal shrinkage of the casting is controlled much more effectively. This is perhaps the single most important factor for a heavy-section high-pressure ductile iron casting.
The second factor is the controlled cooling rate. The steel pipes are not simply placed in the mold as mechanical supports; they are positioned exactly in the zones where the original process used shaped chills. The air inside the pipes continuously removes heat from these zones. This removes the need for separate metal chills and avoids the thermal shock and local over-chilling caused by conventional chills. The cooling effect is more gentle but also more uniform. The uniform cooling prevents the formation of carbides near the chills and reduces the risk of cracks.
The third factor is the gating system. The direct filling of the thick platform under the hydraulic chamber bottom means that the hottest metal is placed at the largest thermal center. The transition runners and filters remove slag before it enters the cavity. The height difference between the transition runner and the runner provides a slag trap. The cross-sectional area ratios ensure a controlled filling velocity. Together, these measures create a clean, stable filling pattern that avoids oxide entrainment. In high-pressure ductile iron casting, oxide inclusions are one of the most dangerous defects because they act as initiation sites for leaks under pressure.
The fourth factor is melt treatment. The combination of silicon carbide preconditioning, a rare-earth magnesium nodulizer, pure antimony microalloying, barium inoculant, and stream inoculation gives excellent control of graphite nucleation and growth. Silicon carbide provides a long-lasting source of nucleation sites. The magnesium and rare earth combine with sulfur and oxygen and create clean metal. Barium in the inoculant improves the resistance to inoculation fade. Antimony suppresses chunky graphite. The final result is a microstructure with high nodule count, high nodularity, and a ferritic matrix capable of high elongation.
It is useful to compare the expected behavior of this ductile iron casting with the requirements of a high-pressure hydraulic component. A pressure-tight casting must have low porosity, low inclusion content, and high resistance to crack propagation. The measured elongation of 18.5% is a good indicator of crack resistance. A ductile iron casting with this level of elongation can tolerate a certain amount of local stress concentration without fracturing. The fine graphite size and high nodularity also help to distribute stress uniformly. The 20 MPa hydraulic test confirmed that the component is able to withstand the pressure without leaking.
One of the most interesting aspects of this project is the role of silicon carbide. The reaction:
$$\mathrm{SiC} + \mathrm{Fe} \rightarrow \mathrm{FeSi} + \mathrm{C}$$
is deceptively simple. In practice, the beneficial effects of silicon carbide cannot be explained only by the final silicon and carbon contents. The local microstructural memory of the melt is important. When SiC particles dissolve slowly, they create local fluctuations in carbon concentration that favor graphite nucleation. The surface SiO2 film also forms silicate inclusions that can act as nuclei. The same inclusions, if they are fine and well dispersed, do not harm the mechanical properties. This is why a small addition of silicon carbide can have a much stronger effect than simply adding ferroalloys.
The antimony microaddition is also worthy of attention. In heavy-wall ductile iron casting, a residual antimony content of between 0.004% and 0.006% is often sufficient to stabilize the graphite structure. At this level, antimony does not form carbides or reduce elongation excessively. In the present ductile iron casting, the final antimony content was 0.0055%. The measured elongation remained as high as 18.5%, which proves that the antimony concentration was well balanced.
The pouring temperature of 1295 °C was also chosen carefully. A higher pouring temperature would reduce the risk of cold shuts but would increase the risk of spheroidization fade, inoculation fade, and mold erosion. A lower pouring temperature would improve graphite nucleation but might lead to mistun. For this heavy section, 1295 °C provides a good compromise. The large mass of the casting means that the iron cools slower than for a light casting, so the pouring temperature can be slightly lower without increasing cold-shut risk. This lower temperature helps to preserve the spheroidization and inoculation effects during the long solidification time.
8. Production Experience and Process Stability
During the trial production, I monitored the process carefully. The first important observation was that the steel-pipe reinforced mold assembly was easy to assemble. The steel pipes can be cut, bent, and welded to form a skeleton that follows the contour of the casting. The molding sand is filled around the skeleton and compacted by conventional molding equipment. After the mold has been assembled, the pipes form a network that remains in the sand structure. Because the pipes have open ends, they connect the interior of the mold to the outside atmosphere. This provides excellent venting.
The second observation was that the pouring behavior was much cleaner than in the traditional process. The metal rose smoothly in the mold. The filters captured a significant amount of dross and slag. After shakeout, the surface of the ductile iron casting was smooth, especially on the underside of the hydraulic chamber bottom. This area was previously a source of inclusions and poor surface quality. With the new gating design and the direct filling of the thick platform, this area became one of the cleanest parts of the casting.
The third observation was the reduction of surface defects on the machined hydraulic chamber. In the earlier process, the machined surface often revealed small oxide streaks, pinholes, or microshrinkage cavities. These defects were not always visible before pressure testing, but they could cause leakage after the chamber was subjected to 20 MPa oil pressure. In the new process, the machined surface was consistently clean. This gave us confidence that the ductile iron casting would be reliable in service.
The ultrasonic testing result, Level 1, indicates the absence of significant internal porosity and inclusions. The magnetic particle testing result, Level 2, indicates that the casting is free of unacceptable surface cracks or linear indications. These results are essential for a safety-related hydraulic component. They also show that the improvements in mold stiffness and cooling have succeeded in eliminating the typical shrinkage defects found in heavy-section ductile iron castings.
From a cost perspective, the new process is more attractive than the traditional process. The removal of large risers reduces the cutting cost and the recycling cost of the riser metal. The elimination of chromite sand reduces the material cost. The reduction of finishing and repair welding lowers labor cost. The higher process yield also reduces the number of rejected castings and the amount of energy consumed per good casting. In a foundry producing several of these ductile iron castings each year, the savings are considerable.
9. Conclusions
In this article, I have presented a complete process design and microstructure control strategy for a heavy-section high-pressure hydraulic cylinder body produced as a ductile iron casting. The work led to the following main conclusions:
First, a high-stiffness mold structure reinforced by steel pipes is an effective replacement for the conventional chromite-sand core and dedicated metal chills. The steel pipes provide both mechanical rigidity and uniform cooling. This reduces mold dilation, improves feeding, and produces a dense casting without external chills.
Second, the gating system with the specific area ratio:
$$\Sigma S_{\mathrm{sprue}} : \Sigma S_{\mathrm{runner1}} : \Sigma S_{\mathrm{runner2}} : \Sigma S_{\mathrm{ingate1}} : \Sigma S_{\mathrm{ingate}} = 1.0 \sim 1.1 : 1.3 \sim 1.5 : 1.3 \sim 1.5 : 1.0 : 1.0$$
together with a filter and a slag-trapping height difference between the transition runner and the runner, provides clean and quiescent mold filling. Placing the ingates directly at the thick platform under the hydraulic chamber bottom creates a favorable thermal gradient and reduces the risk of shrinkage and slag defects.
Third, the melting process based on 50% pig iron, 30% steel scrap, and 20% return scrap, with 0.7% silicon carbide and 0.8% recarburizer, followed by FeSi75 adjustment and superheating to 1452 °C, gives the desired base melt composition. The addition of 0.006% antimony, 1.2% rare-earth magnesium nodulizer, 0.62% barium inoculant, and 0.12% stream inoculation ensures a high nodule count, high nodularity, and a ferritic matrix with high elongation.
Fourth, the final composition of 3.55% C, 2.45% Si, 0.25% Mn, 0.0098% S, 0.037% Mg, 0.008% RE, and 0.0055% Sb, with a carbon equivalent of 4.38%, is suitable for this heavy-section ductile iron casting. The carbon equivalent formula:
$$\mathrm{CE} = \mathrm{C} + \frac{1}{3}\left(\mathrm{Si} + \mathrm{P}\right)$$
is a useful tool for evaluating the solidification behavior and graphite morphology of such castings.
Fifth, the cast-on test block showed a nodularity of 94.2%, graphite size grade 6, and a predominantly ferritic matrix with fine local pearlite. The measured tensile strength was 405 MPa, yield strength was 278 MPa, elongation was 18.5%, and hardness was 148 HBW. These values satisfy the requirements for a pressure-tight ductile iron casting.
Finally, the trial-produced hydraulic cylinder bodies passed ultrasonic testing at Level 1, magnetic particle inspection at Level 2, and the 20 MPa hydraulic pressure test without leakage. The improved ductile iron casting process is therefore suitable for stable production and quality control of high-pressure hydraulic cylinder bodies.
This ductile iron casting development demonstrates that high-pressure hydraulic components do not have to be forged from steel. With careful attention to mold rigidity, cooling control, gating design, and melt treatment, ductile iron casting can deliver the required strength, ductility, and pressure tightness. The process described here offers a reliable alternative for injection molding machines and other hydraulic systems where cast iron is often the most cost-effective choice.
Further work can be directed toward optimizing the number and position of steel pipes through solidification simulation, measuring the local cooling curves in different sections of the casting, and extending the same process to other large hydraulic components. The principles described here are not limited to this particular hydraulic cylinder body. They can be applied to valve bodies, pump housings, cylinder blocks, and other ductile iron castings where internal soundness and pressure tightness are critical. In every case, the goal is the same: to combine the design freedom of casting with the mechanical reliability of a forged or fabricated steel component. The success of this ductile iron casting project shows that the goal is achievable when engineering and metallurgy are carefully integrated.