The pursuit of internal soundness in ductile iron castings while simultaneously maximizing yield and minimizing production costs represents a fundamental challenge in foundry engineering. Traditional casting theory often dictates the use of risers for feeding, especially for smaller castings with low geometric modulus. This case study details a first-person development journey to produce a sound, small-modulus mandrel casting using a riserless approach, challenging conventional wisdom. The subject is a head mandrel with a specified material grade equivalent to QT500-7. The casting has a rough weight of 68 kg and, critically, a calculated geometric modulus (M) of only 1.69 cm. According to established handbooks, this low modulus typically necessitates riser supplementation to compensate for shrinkage during solidification. Initial calculations of the casting’s natural Perimetrischem Quotient (Qm)—a parameter relating volume to cooling surface area—yielded a value of approximately 17.9 kg/cm³, which also falls short of some recommended thresholds for successful riserless casting of ductile iron castings.
The primary obstacle was the inherent糊状凝固 (mushy solidification) characteristic of ductile iron castings. The extensive solidification range and graphite expansion can lead to mold wall movement, making riser feeding inefficient and often ineffective at eliminating microscopic shrinkage porosity. The goal was to harness the expansive force of graphite precipitation to counteract the volumetric shrinkage of the iron, a principle central to riserless or feeding-neutral designs. The hypothesis was that even a casting with an unfavorable natural modulus and Qm could be made sound by artificially modifying its solidification dynamics to promote what is known as “均衡凝固” (balanced solidification), where shrinkage and expansion phases overlap.
Process Evolution: From Conventional to Riserless Design
The development progressed through three distinct process stages, each providing critical insights.
| Process Design | Orientation & Feeding | Key Features | Calculated Yield | Internal Soundness Result |
|---|---|---|---|---|
| Process 1 (Baseline) | Vertical pouring, horizontal molding. Top risers. | Two Ø85mm x 120mm necked risers on the flange. Bottom gating. | 73% | Unacceptable shrinkage porosity present. |
| Process 2 (Riserless with Chills/Fins) | Vertical pouring, horizontal molding. No risers. | Riserless. External fins/chills added to the shaft body. Bottom gating. | 82% | No shrinkage defects detected. |
| Process 3 (Optimized Riserless) | Vertical pouring, horizontal molding. No risers. | Riserless. Strategic external chill plates placed on shaft body. Bottom gating. | 90% | No shrinkage defects detected (optimal). |
Process 1: Conventional Riser-Dependent Approach
The initial design adhered to traditional methods. To address potential core buoyancy in the long, thin bore during horizontal pouring, a vertical pouring orientation was selected. Two substantial risers were placed on the top (flange) section to act as a feed metal reservoir. Despite a relatively low yield of 73%, sectioning revealed internal shrinkage defects. This confirmed the limitation of relying solely on riser feeding for this specific geometry of ductile iron castings, where the expansive phase was not effectively harnessed or synchronized with the shrinkage demand.
Process 2: The Shift to Riserless and the Concept of Artificial Qm
Abandoning the riser-based philosophy, the focus shifted to creating conditions for balanced solidification. Since the casting’s natural modulus and Qm were low, the strategy was to artificially increase its effective cooling surface area and thereby alter its solidification sequence. This was achieved by adding thin longitudinal fins or “chill ribs” along the length of the cylindrical body. These fins acted as intense cooling channels, creating localized regions of rapid solidification (thin sections) adjacent to the slower-solidifying main body (thick section).
The effect is quantified by recalculating the Perimetrischem Quotient. The original Qm is given by:
$$ Q_m = \frac{W}{V/A} = \frac{W}{M} $$
where \(W\) is the casting weight, \(V\) is volume, and \(A\) is surface area. By adding fins, the total surface area for heat extraction \(A\) increases significantly. While the weight \(W\) increases slightly due to the added metal in the fins, the dominant effect is the reduction in the volumetric modulus \(V/A\), leading to a higher effective \(Q_m\). In this case, the effective \(Q_m\) was raised to approximately 37 kg/cm³.
Production trials were conducted, pouring castings from the same melt using both Process 1 and 2 molds. The riserless Process 2 castings were found to be completely sound. This demonstrated a pivotal insight: the successful riserless casting of ductile iron castings is not strictly governed by a fixed minimum modulus, but by the ability to engineer a solidification pattern where expansion compensates for shrinkage. The fins created a deliberate thermal gradient, prompting the thin sections to solidify and begin their graphite expansion phase earlier, which could help feed the still-solidifying thicker sections.
Process 3: Optimization with Strategic Chill Plates
Building on the principle of Process 2, an optimization was made. Instead of continuous fins, discrete external chill plates were strategically placed on the casting’s outer surface. This offered a more controlled and efficient method for achieving the same goal. The chills provide an even greater local cooling intensity than fins, creating very pronounced “thin-wall” zones. The pattern of alternating chilled and unchilled areas along the casting length effectively segments the casting into a series of closely coupled thin and thick regions, ideal for promoting balanced solidification.
The solidification progression can be modeled conceptually. Let \( t_s(thin) \) and \( t_s(thick) \) represent the solidification times of the chilled and unchilled regions, respectively, with \( t_s(thin) \ll t_s(thick) \). The volumetric change \( \Delta V(t) \) in a region is the sum of liquid contraction \( \Delta V_{cont}(t) \) and graphite expansion \( \Delta V_{exp}(t) \). In a well-designed riserless system for ductile iron castings, the sequence aims for:
$$ \Delta V_{exp}^{thin}(t_1) \approx |\Delta V_{cont}^{thick}(t_1)| $$
at an intermediate time \( t_1 \), where the thin section is expanding while the thick section is still contracting. Later, the expansion of the thick section \( \Delta V_{exp}^{thick}(t_2) \) compensates for its own remaining micro-shrinkage tendency. The chills precisely accelerate the solidification of select areas to orchestrate this timing.
This final process achieved a yield of 90% and consistently produced defect-free ductile iron castings. The internal soundness was validated through extensive sectioning.

Theoretical Framework and Design Principles
The success of this project hinges on several key principles specific to the metallurgy and solidification of ductile iron castings.
1. Geometric Modulus vs. Effective Cooling Modulus:
Traditional modulus calculations \( M = V/A \) consider the casting’s inherent geometry. For riserless feasibility, a high modulus is often cited as favorable. However, the effective modulus during solidification can be radically altered by applied cooling. The use of chills or fins increases \(A\) in the calculation, thereby decreasing the local effective modulus \(M_{eff}\) of specific regions. The design goal is not a uniformly high modulus, but a controlled differential in effective modulus across the casting to drive directional solidification and time the expansion phases.
2. The Role of the Perimetrischem Quotient (Qm):
Qm serves as a useful indicator of a casting’s “feeding demand” relative to its ability to generate internal pressure via expansion. A low Qm indicates a high surface-area-to-volume ratio, meaning heat dissipates quickly, shrinkage occurs rapidly, and the subsequent graphite expansion may be too weak or ill-timed to be effective. The empirical threshold often mentioned is Qm ≥ 20 kg/cm³ for riserless casting. The methodology here proves this threshold is not a barrier for ductile iron castings if the casting’s effective Qm is artificially increased through added cooling surfaces. The formula becomes:
$$ Q_{m_{effective}} = \frac{W_{casting} + W_{added\,metal}}{(V_{casting}+V_{added\,metal}) / (A_{casting} + A_{added\,cooling})} $$
where \(A_{added\,cooling}\) is the enhanced surface area from fins/chills, which is the dominant factor.
3. Balanced Solidification and Expansion Pressure:
The core mechanism is the management of the internal pressure \(P_{int}\) generated by graphite expansion within the coherent solid network. This pressure must overcome the metallostatic pressure and any resistance from mold wall movement to compress micropores. The pressure development is a function of time, carbon equivalent, inoculation efficiency, and cooling rate. By using chills, the cooling rate in specific zones is increased, which can accelerate the onset of eutectic solidification and graphite expansion in those zones. The pressure generation \(P_{exp}(t)\) in the early-solidifying zones can be conceptually related to the fraction of solid \(f_s\) and the rate of graphite precipitation:
$$ \frac{dP_{exp}}{dt} \propto \eta \cdot \frac{df_s}{dt} \cdot \frac{d[\%C_{graphite}]}{dt} $$
where \(\eta\) is an efficiency factor accounting for mold rigidity. The chills increase \(df_s/dt\) locally, aiming to align the peak of \(dP_{exp}/dt\) in the chilled areas with the period of maximum feeding demand in the adjacent thicker areas.
| Parameter | Symbol/Formula | Role in Riserless Casting | Influence of Chills/Fins |
|---|---|---|---|
| Geometric Modulus | $$ M = \frac{V}{A} $$ | Indicates natural cooling rate. Low M suggests fast cooling/high feeding demand. | Decreases the local effective modulus (Meff) where applied. |
| Perimetrischem Quotient | $$ Q_m = \frac{W}{M} $$ | Empirical indicator of feeding demand vs. expansion potential. | Increases the effective Qm by significantly increasing A. |
| Solidification Time Gradient | $$ \nabla t_s = \frac{t_s^{thick} – t_s^{thin}}{d} $$ | Drives directional solidification and timing for expansion compensation. | Sharply increases \(\nabla t_s\) by drastically reducing \(t_s^{thin}\). |
| Graphite Expansion Pressure | \( P_{exp}(f_s, \dot{C}_g) \) | Internal pressure that seals shrinkage porosity. | Promotes earlier and potentially more synchronized \(P_{exp}\) generation in thin zones. |
Conclusions and Broader Implications
This developmental study conclusively demonstrates that small-modulus ductile iron castings can be successfully produced using a riserless casting technique, yielding significant benefits in material yield and cost reduction. The key learning points are:
1. Paradigm Shift: The primary condition for riserless casting of ductile iron castings is not achieving an arbitrary minimum geometric modulus, but rather engineering the solidification sequence to ensure the overlapping of shrinkage and expansion phases (balanced solidification).
2. Artificial Modification of Cooling Characteristics: When a casting’s natural geometry is unfavorable (low M, low Qm), its effective solidification behavior can be favorably altered by adding external cooling features. Attaching chill fins or strategically placing chill plates effectively increases the casting’s cooling surface area, raises its effective Perimetrischem Quotient, and creates the necessary thermal differentials.
3. Design Principle: The optimal design involves creating a deliberate pattern of alternating regions with starkly different cooling rates (i.e., “thin” and “thick” sections in terms of solidification time). This pattern enables the early-expanding regions to feed the later-contracting regions, and the subsequent expansion in the heavier sections to self-feed any remaining micro-shrinkage.
4. Process Superiority: Among the trials, the riserless process with discrete external chills (Process 3) proved optimal, delivering a 90% yield and consistent internal soundness. It outperformed the conventional risered approach both economically and qualitatively.
This methodology extends the applicability of riserless casting to a broader range of ductile iron castings geometries. It provides a practical framework for foundry engineers to analyze casting geometry not just as a fixed shape, but as a moldable thermal system where strategic cooling interventions can unlock the full potential of graphite expansion to achieve dense, high-quality components without the cost and waste associated with traditional feeding risers.
