The relentless pursuit of higher engine power density, reduced size, and controlled emissions has cemented turbocharging as a pivotal technology in modern automotive engineering. At the heart of this technology lies the turbocharger turbine wheel, a component subjected to extreme thermal and mechanical stresses. To improve transient response and efficiency, these turbines are designed with complex geometries featuring thin, highly-curved blades attached to a substantially thicker hub. This drastic variation in cross-section presents a significant challenge during manufacturing, particularly via the investment casting process, where it predisposes the component to the formation of hot tears—one of the most detrimental and reject-causing defects in precision casting.
This article presents a detailed, first-person perspective investigation into the prediction and mitigation of hot tearing in automotive turbine wheels cast from IN713C nickel-based superalloy using the investment casting method. We employ advanced numerical simulation as our primary tool to dissect the solidification behavior, stress evolution, and ultimately, the propensity for hot crack formation. The core of our analysis focuses on elucidating the relationship between critical process parameters—namely pouring temperature and mold preheat temperature—and the resulting hot tearing tendency. Our goal is to establish a robust framework for optimizing the investment casting process to produce sound, reliable turbine components.
1. The Critical Challenge of Hot Tearing in Investment Casting
Hot tearing, also known as hot cracking, is a defect that occurs in the final stages of solidification, when the alloy is in a semi-solid state characterized by a coherent network of solid dendrites surrounded by thin, inter-dendritic liquid films. The mechanical strength of this mushy zone is extremely low. If the thermally-induced tensile stresses (from constrained contraction) or strains developed during cooling exceed the fragile strength of this coherent solid network, rupture occurs, leading to a crack. In the context of investment casting complex parts like turbine wheels, several factors converge to exacerbate this risk:
- Geometric Constraints: The stark difference in section thickness between the thin airfoil blades (often sub-1mm) and the thick hub creates severe differential cooling rates. The blades solidify and begin to contract significantly earlier and faster than the hub.
- Mechanical Constraint: The rigid ceramic shell used in investment casting provides substantial resistance to the natural thermal contraction of the casting.
- Alloy Characteristics: Nickel-based superalloys like IN713C have a wide solidification temperature range, which prolongs the time the material spends in the vulnerable mushy zone, increasing the window for hot tear initiation.
Predicting the exact location and severity of hot tears through trial-and-error is prohibitively expensive and time-consuming. Therefore, numerical simulation has become an indispensable tool for virtual prototyping and process optimization in investment casting.
2. Material: IN713C Superalloy and Its Casting Characteristics
The material under investigation is IN713C, a precipitation-strengthened nickel-based superalloy widely used for gas turbine components due to its excellent high-temperature strength, oxidation resistance, and fatigue properties. Its typical as-cast microstructure consists of a γ matrix with a primary strengthening phase of γ’ (Ni3(Al, Ti)), alongside minor carbide and boride phases. For simulation accuracy, precise thermophysical and mechanical properties are paramount. These were derived from Scheil-Gulliver solidification calculations and validated literature data, covering the transformation from liquid to solid.
Key Thermophysical and Mechanical Properties for Simulation:
| Property | Value / Characteristic | Notes for Simulation |
|---|---|---|
| Liquidus Temperature (TL) | ~1345 °C | Critical for defining pouring range. |
| Solidus Temperature (TS) | ~1196 °C | Defines end of solidification. |
| Fraction Solid vs. Temperature | Curve derived from Scheil model | Essential for modeling mushy zone behavior. |
| Young’s Modulus (E) | Temperature-dependent function | Decreases significantly near solidus. |
| Yield Strength (σy) | Temperature-dependent function | Approaches zero in the coherency temperature range. |
| Thermal Expansion Coefficient (α) | Temperature-dependent function | Drives thermal contraction strain. |
3. Numerical Modeling Framework for Stress and Hot Tearing
To simulate the evolution of stress and strain during solidification and cooling, a thermo-elasto-plastic material model was employed. This model realistically captures the transition from a purely viscous liquid to a plastic/elastic solid. In the solid state, the material behaves elastically until the von Mises stress reaches the temperature-dependent yield stress, after which plastic deformation occurs.
The total strain increment ${d\varepsilon}$ is decomposed into elastic ${d\varepsilon^e}$, plastic ${d\varepsilon^p}$, and thermal ${d\varepsilon^T}$ components:
$$ {d\varepsilon} = {d\varepsilon^e} + {d\varepsilon^p} + {d\varepsilon^T} $$
The stress-strain relationship in the elastic regime is governed by Hooke’s law:
$$ {d\sigma} = [D^e] {d\varepsilon^e} $$
where $[D^e]$ is the elastic stiffness matrix. For the elasto-plastic regime, the constitutive relation is:
$$ {d\sigma} = [D^{ep}] ( {d\varepsilon} – {d\varepsilon^p} – {d\varepsilon^T} ) $$
where $[D^{ep}]$ is the elasto-plastic stiffness matrix. A linear hardening rule is often assumed for simplicity in casting simulations:
$$ \sigma = \sigma_0 + H \varepsilon_{pl} $$
where $\sigma_0$ is the initial yield stress, $H$ is the plastic modulus, and $\varepsilon_{pl}$ is the accumulated plastic strain.
Hot Tearing Criterion (HTI): Predicting the location of hot tears requires a dedicated damage criterion. We utilized the accumulated plastic strain-based Hot Tearing Index (HTI) model. This index integrates the plastic strain rate over the vulnerable period when the material has low strength—typically from the coherency temperature (where a solid network forms, ~0.7-0.9 fraction solid) to the solidus temperature.
$$ HTI = \int_{t_{coh}}^{t_s} \sqrt{\frac{2}{3} \dot{\varepsilon}^p : \dot{\varepsilon}^p} \, d\tau $$
where $t_{coh}$ is the time at coherency, $t_s$ is the time at the solidus temperature, and $\dot{\varepsilon}^p$ is the plastic strain rate tensor. A higher HTI value at a location indicates a greater propensity for hot tearing.
4. Simulation Setup: Geometry and Process Parameters
The subject was a radial turbine wheel approximately 60 mm in height and 86 mm in maximum diameter, featuring 10 highly curved blades with a nominal thickness of 0.7 mm, connected to a central hub with a section thickness of about 28 mm. A cluster was designed with three wheels attached to a central down-sprue via a pour cup and three ingates. The wheel axis was angled at 130° to the sprue to facilitate filling.
The core of this study was to systematically evaluate the impact of investment casting process parameters. The following matrix was defined for the simulations:
| Process Parameter | Levels Investigated |
|---|---|
| Pouring Temperature (Tpour) | 1400 °C, 1450 °C, 1500 °C, 1550 °C |
| Mold (Shell) Preheat Temperature (Tmold) | 800 °C, 850 °C, 900 °C |
| Interfacial Heat Transfer Coefficient (HTC) | 900 W/(m²·K) (Constant) |
The IN713C alloy was modeled as a thermo-elasto-plastic material, while the ceramic shell was treated as a rigid body. The simulations sequentially solved for fluid flow during filling, heat transfer during solidification, and stress/strain development during cooling.
5. Results and Discussion: From Solidification Sequence to Hot Spot Identification
5.1 Filling and Solidification Patterns
The simulation of the filling phase confirmed that due to the generous gating design and small casting size, the mold cavity filled very rapidly, typically within about one second, ensuring a smooth fill without major surface defects. The critical analysis began with the solidification sequence. The temperature field at an intermediate time clearly showed a steep thermal gradient. The thin blade sections, especially at the tips and trailing edges, cooled and solidified first, while the massive hub remained at a significantly higher temperature for a much longer duration.
This differential solidification is the primary driver of thermal stress. As the blades solidify and attempt to contract, they are constrained by the still-molten or much hotter, less contractive hub region. This restraint leads to the development of tensile stresses in the blades during the later stages of solidification—precisely when their mechanical strength is at its lowest.
5.2 Stress Evolution and Prediction of Hot Tearing Sites
The stress simulation results were unambiguous. The highest tensile stresses were concentrated at the thin, constrained edges of the turbine blades. The corresponding HTI map showed a perfect correlation, with peak HTI values located at these same blade-edge regions. This simulation prediction was validated against actual castings from initial trial runs, where visible hot tears were found to occur consistently at the blade tips and thin sections, confirming the accuracy of the modeling approach for this investment casting application.
5.3 Quantitative Analysis of Parameter Influence
To quantitatively assess the impact of process parameters, a specific node at a high-risk blade edge location (Point A) was monitored throughout all simulation runs. The evolution of stress, temperature, fraction solid, and the computed HTI was tracked.
Effect of Pouring Temperature (at constant Tmold = 800°C):
The results revealed a non-monotonic relationship between pouring temperature and hot tearing tendency.
| Pouring Temp. | Peak Thermal Stress at A | Time in Vulnerable Zone | Resulting HTI |
|---|---|---|---|
| 1400 °C | High (~47 MPa) | Shortest | Low |
| 1450 °C | Medium-High | Longer | Highest |
| 1500 °C | Medium (~25 MPa) | Long | Medium |
| 1550 °C | Low | Longest | Low |
Interpretation: At lower pouring temperatures (1400°C), high thermal stress dominates, but the vulnerable solidification period is short, limiting strain accumulation. As pouring temperature increases to 1450°C, the stress remains significant, but the prolonged mushy zone duration allows for greater strain accumulation, leading to the peak HTI. At even higher temperatures (1500-1550°C), although the vulnerable time is longest, the thermal stress is dramatically reduced due to a slower cooling rate and reduced thermal gradients. The lower stress overrides the longer time, resulting in a decreased HTI. This highlights the competing mechanisms of stress level versus vulnerability window duration.
Effect of Mold Preheat Temperature (at constant Tpour = 1500°C):
In contrast, increasing the mold preheat temperature showed a consistently beneficial effect on reducing hot tearing.
| Mold Temp. | Peak Thermal Stress at A | Time in Vulnerable Zone | Resulting HTI |
|---|---|---|---|
| 800 °C | High | Medium | High |
| 850 °C | Medium | Longer | Medium |
| 900 °C | Low | Longest | Lowest |
Interpretation: A hotter shell reduces the cooling rate and thermal gradient between the thin and thick sections of the casting more effectively. This directly lowers the magnitude of thermally induced tensile stresses. While the solidification time and vulnerable period increase, the reduction in stress is the dominant factor, leading to a steady decrease in the computed HTI. This makes mold preheat temperature a more straightforward and effective parameter to control for hot tear mitigation in this investment casting scenario.
5.4 Interaction Effects and Process Optimization
A comprehensive view of the interaction between pouring and mold temperature is crucial. The analysis shows that when one parameter is already at a favorable level (e.g., high mold temperature), the sensitivity of HTI to changes in the other parameter (e.g., pouring temperature) diminishes. Conversely, when conditions are severe (low mold temperature), the process is highly sensitive to changes in pouring temperature.
Based on the full matrix of simulation results, an optimized process window was identified. While the lowest HTI was achieved at the highest temperatures (Tpour=1550°C, Tmold=900°C), excessively high pouring temperatures can promote other issues like coarse grain structure, increased metal-shell reaction, and shrinkage porosity. Therefore, a balanced optimum is recommended:
- Optimized Pouring Temperature: 1500 °C
- Optimized Mold Preheat Temperature: 900 °C
This combination maintains a sufficiently high temperature to minimize thermal stress and hot tearing risk while avoiding the potential drawbacks associated with extreme superheat in the investment casting of IN713C.
6. Conclusions and Perspectives
This systematic numerical investigation successfully demonstrated the power of simulation in addressing the complex problem of hot tearing in the investment casting of intricate components. For the IN713C automotive turbine wheel:
- The primary location for hot tearing was accurately predicted to be the thin edges of the turbine blades, a result of severe differential cooling and constraint, which was confirmed by experimental castings.
- The hot tearing tendency, quantified by the HTI, exhibits a complex, non-linear relationship with pouring temperature, initially increasing before decreasing due to the competing effects of stress level and duration in the mushy zone.
- Increasing the mold preheat temperature consistently reduces the hot tearing tendency by lowering thermal gradients and stresses, making it a highly effective control parameter.
- A synergistic optimization of parameters is key. The recommended setpoint (Tpour = 1500°C, Tmold = 900°C) provides a robust solution that minimizes hot tearing risk while maintaining overall casting quality.
The methodology outlined here—combining accurate thermophysical data, a thermo-elasto-plastic stress model, and a strain-based hot tearing criterion—provides a reliable framework for virtual process development. It enables foundries to proactively design gating systems and define process parameters for investment casting of complex, high-integrity components, reducing costly physical trials, shortening lead times, and improving yield. Future work could integrate microstructural models to predict grain size and segregation, further refining the prediction of the coherency point and mushy zone strength, thus enhancing the fidelity of hot tearing simulation in investment casting.