The presence of internal residual stress within sand casting products often poses significant challenges for subsequent machining, assembly, and in-service performance. For critical components like clutch pressure plates, residual stress can directly influence key mechanical properties such as strength, fatigue limit, and rigidity. More critically, it detrimentally affects dimensional accuracy and can lead to premature failure through warping, distortion, or cracking under operational loads, thereby reducing the component’s service life. Therefore, a comprehensive investigation into the origins, influencing factors, and mitigating strategies for residual stress is paramount for enhancing the quality and reliability of such components. This article focuses on a GGV30 vermicular graphite iron (VGI) automobile clutch pressure plate, a typical sand casting product, and conducts a systematic study on the generation and reduction of its casting residual stress to provide a theoretical and technical foundation for guiding practical low-stress production.
The genesis of residual stress in sand casting products is rooted in the non-uniform thermal history during solidification and cooling. When molten metal cools, it undergoes liquid and solid contraction. Differential cooling rates across the component, caused by variations in section thickness or proximity to mold walls, create internal temperature gradients. If the free thermal contraction of one region is constrained by another that is cooling at a different rate, or by the mold itself, internal stresses develop. These are locked into the component as residual stress after the mold is removed. The fundamental principle can be illustrated with a simplified mechanical analogy. Consider a system where three springs of different free lengths (L1, L2, L3) and stiffness (a1, a2, a3) are compressed between two rigid plates to a common length L. In the absence of external force, the system is in equilibrium because each spring exerts an internal force: Q1=a1(L-L1); Q2=a2(L-L2); Q3=a3(L-L3). The sum of these forces is zero (Q1+Q2+Q3=0). These internal forces, Q1, Q2, and Q3, are analogous to the residual stress within a constrained sand casting product.
Residual stresses in sand casting products can be categorized by their scale. Macrostresses (Type I) act over large areas encompassing many grains; their imbalance can cause macroscopic dimensional changes. Microstresses exist between grains (Type II) or within the crystal lattice due to dislocations (Type III). Based on their origin during the casting process, the primary stresses are: Thermal Stress arising from temperature gradients, Phase Transformation Stress resulting from volumetric changes during solid-state phase changes occurring at different times in different sections, and Shrinkage Stress caused by external mechanical restraint from the mold cores. The final residual state is a complex superposition of these.
Accurately measuring residual stress is crucial for both research and quality control of sand casting products. Techniques are broadly classified as non-destructive, semi-destructive, and destructive. Table 1 provides a comparative overview of key methods.
| Testing Method | Measurement Depth | Advantages | Disadvantages |
|---|---|---|---|
| X-ray Diffraction | ~10-20 µm (surface) | Fast, non-destructive, good accuracy for surfaces. | Surface only; expensive; sensitive to surface finish. |
| Neutron Diffraction | Up to several cm | Deep penetration, can map 3D stress fields. | Extremely limited facility access; very high cost. |
| Ultrasonic | <1 mm | Fast, portable, low cost. | Low spatial resolution; accuracy affected by microstructure. |
| Blind Hole Drilling | <2 mm (semi-destructive) | Well-established, relatively inexpensive, practical for engineering. | Semi-destructive; measures near-surface stress; moderate accuracy. |
| Crack Compliance / Contour Method | Through-section (destructive) | High accuracy for internal stress profiles; provides 2D maps. | Fully destructive; complex procedure and analysis. |
For this study, given the need to validate simulation results on actual sand casting products with minimal damage, the blind hole drilling method was employed. The principle involves bonding a strain gauge rosette to the surface, drilling a small hole at its center to relax the local stress, and measuring the resulting strain field. The principal stresses (σ1, σ2) and their direction are calculated from the measured strains (ε1, ε2, ε3) using the formulas:
$$ \sigma_{1,2} = \frac{E}{2} \left( \frac{\epsilon_1 + \epsilon_3}{A} \pm \frac{1}{B} \sqrt{(\epsilon_1 – \epsilon_3)^2 + (\epsilon_1 + \epsilon_3 – 2\epsilon_2)^2} \right) $$
$$ \tan 2\theta = \frac{\epsilon_1 + \epsilon_3 – 2\epsilon_2}{\epsilon_1 – \epsilon_3} $$
Where E is the Young’s modulus, and A and B are strain release coefficients dependent on the material’s Poisson’s ratio (ν), hole diameter (d), and gauge geometry (rm is the mean rosette radius):
$$ A = -\frac{1+\nu}{2E} \frac{d^2}{r_m^2}, \quad B = -\frac{2}{E} \left( \frac{d^2}{r_m^2} – \frac{3(1+\nu)}{4} \frac{d^4}{r_m^4} \right) $$
Finite element simulation has become an indispensable tool for predicting and analyzing residual stress in sand casting products, allowing for virtual process optimization. In this work, the commercial software ProCAST was utilized to simulate the entire casting process of the VGI pressure plate. The workflow encompassed 3D modeling, meshing, assignment of material properties (thermophysical and mechanical), setting of boundary conditions (pouring temperature, mold interface heat transfer), and coupled thermal-stress calculation. The material properties for the specific GGV30 VGI composition, such as thermal conductivity (k), density (ρ), enthalpy (H), and solid fraction (fs), were calculated as functions of temperature, as shown in the representative curves below.
$$ k = f(T), \quad \rho = f(T), \quad H = \int C_p dT, \quad f_s = f(T) $$
The simulation of the original top-gating process revealed intricate thermal and mechanical histories. Filling was generally stable, concluding in approximately 7.35 seconds. However, significant temperature gradients were established immediately post-fill, with the thin sections at the outer/inner rim and small lugs cooling fastest (~125°C difference from the hotter mid-plate regions). The solidification pattern showed a dominant progression from both the outer and inner rims towards the central plate body, while areas around the large lugs (attached to risers) exhibited a small-scale directional solidification from the inner rim towards the riser. Post-solidification cooling analysis further detailed these radial thermal gradients, confirming that temperature differences were most pronounced in radial lines passing through the small lugs.
This non-uniform thermal history directly dictated the deformation and final stress state. Displacement analysis showed that radial contraction occurred from the outer perimeter inwards towards the central hub, with a cumulative effect leading to significant constraint at the inner rim. Thickness-wise, the plate exhibited a squeezing deformation from both faces towards the mid-plane. The resulting residual stress field, as predicted by simulation, was non-uniform. Tensile stresses dominated, with the highest concentrations found at the inner rim opposite the small lugs and at the junctions of large lugs with risers. Radial stress analysis confirmed that inner rim stresses were substantially higher than outer rim stresses, with the plate surface stresses intermediate. The stress gradient was most severe along radials through small lugs.
To validate the numerical model, residual stress measurements were conducted on actual sand casting products (pressure plates) using the blind hole method at seven locations on the plate surface. A comparison with the simulation data extracted from corresponding nodes showed consistent trends in stress variation across the component. Although the simulated values were generally lower in magnitude and of opposite sign (simulated tensile vs. measured compressive in some cases), the discrepancies were within an acceptable range for such complex analyses. The differences can be attributed to idealized boundary conditions (constant heat transfer coefficients), approximations in material property data, and the inherent challenges in exactly matching measurement and node locations. Nevertheless, the correlation validated the use of simulation as a reliable tool for investigating residual stress in these sand casting products.
Building on the validated model, a comprehensive study was undertaken to identify strategies for reducing residual stress in the VGI pressure plate sand casting products. The investigated avenues included natural aging, modification of chemical composition, optimization of process parameters, and redesign of the gating system.
1. Natural Aging: Measuring stress relaxation over time showed that a 10-day natural aging period achieved most of the stress reduction observed after the initially specified 15 days, offering a potential 5-day reduction in the production cycle while maintaining dimensional stability.
2. Chemical Composition: The influence of Carbon Equivalent (CE = %C + %Si/3) and Silicon-to-Carbon ratio (Si/C) was critical. Higher CE within the studied range (4.32% to 4.64%) promoted a healthier graphite morphology (shorter, thicker vermicular graphite), increased ferrite content in the matrix, and enhanced graphite expansion during eutectic solidification. These factors collectively reduced both thermal and transformation stresses. Experimentally, increasing CE from 4.32% to 4.64% reduced residual stress by an average of approximately 24%. Furthermore, at a fixed CE, increasing the Si/C ratio initially decreased stress to a minimum before causing a slight increase, following a parabolic trend. The synergistic effect of high CE (e.g., 4.64%) and a high Si/C ratio (e.g., 0.73) was found to be most effective for minimizing residual stress in these ferrous sand casting products.
3. Process Parameters: Numerical studies on pouring and shakeout temperatures revealed clear trends. A higher pouring temperature (1420°C vs. 1360°C) resulted in a slower cooling rate after solidification, reducing thermal gradients and final stress by about 5 MPa. However, further increases showed diminishing returns. More dramatically, shakeout temperature had a profound impact. Simulation of shakeout at 500°C (within the elastoplastic transition range) versus 200°C revealed that while the stress state at 500°C shakeout was initially lower, the subsequent rapid and uneven cooling in air led to a final stress nearly 12% higher than that for the 200°C shakeout. This underscores the importance of allowing the sand casting product to cool slowly in the mold to below approximately 300°C to minimize locked-in thermal stress.
4. Gating System Design: The original top-gating system was compared with a redesigned bottom-gating system. The bottom-gate design promoted smoother filling, eliminated turbulence-related defects, and, crucially, modified the thermal field. The horizontal runner in the bottom-gate design acted as a thermal mass, reducing the cooling rate of the lower plate section, particularly around the small lugs. This reduced the thermal gradients between the small lug radials and the rest of the plate. Consequently, the simulated residual stress in the critical inner rim region was found to be approximately 30 MPa lower in the bottom-gate design compared to the top-gate design, demonstrating a significant improvement achievable through process design for sand casting products.
In conclusion, this integrated study employing numerical simulation, experimental validation, and multi-factorial analysis provides a clear pathway for controlling residual stress in vermicular graphite iron sand casting products like clutch pressure plates. The key findings indicate that residual stress can be effectively minimized by employing a bottom-gating system, utilizing a higher carbon equivalent (around 4.64%) coupled with an optimized high Si/C ratio (around 0.73), implementing a higher pouring temperature (~1420°C), and ensuring a lower shakeout temperature (below 300°C). Adopting these strategies enables the production of dimensionally stable, high-integrity sand casting products with enhanced performance and longevity.