Key Design Considerations for Gating Systems in Investment Cast Stainless Steel Products

Expert Perspectives on Precision Casting

heweifeng

7/8/20253 min read

Systematic Research on Investment Casting Process for Stainless Steel Components

Practical Optimization of Full-Process Based on 15 Years of Production Experience

Author Profile

With 15 years of experience in investment casting, the author has evolved from a novice to a professional engineer mastering core process design. This paper, based on 327 validated process parameters, presents actionable optimization strategies for stainless steel precision casting.

Chapter 1: Gating System Design Principles and Practice

1.1 Sequential Solidification Principle

Core Principle: Gates must be positioned in the last solidifying region to ensure directional solidification

Thick Section Priority: For components with wall thickness variation >3:1, adopt multi-gate design (add 1 auxiliary gate per 5mm thickness increase)

Hot Spot Compensation: Use ProCAST software to predict hot spots and strategically place risers (riser diameter ≥1.8× hot spot dimension)

Case Study: A flange component achieved a 12%→2.3% shrinkage defect reduction with triple-gate design

1.2 Cold End Effect Application

Use Case: Non-ideal geometry components (e.g., thin-walled irregular parts)

Reverse Design Method: Relocate gates away from critical areas to utilize thermal gradients

Example: A turbine wheel achieved directional solidification at the blade root by shifting gates 35mm, increasing yield by 19%

1.3 Gate Dispersion Strategy

Technical Guidelines:

Circular Component Design: Use odd-numbered gates (3/5/7 recommended) to reduce roundness deviation by 0.08-0.12mm

Thermal Distribution Optimization: Adjacent gate spacing should exceed 3× wall thickness to avoid localized overheating

Experimental Data: Three-gate vs dual-gate comparison reduced deformation standard deviation by 41%

1.4 Gate Length Thermomechanical Balance

Parameter Optimization Model:

Gate Length (mm) | Feeding Efficiency (%) | Deformation (mm) |

|------------------|------------------------|------------------|

| ≤15 | 82 | 0.35 |

| 16-30 | 91 | 0.22 |

| ≥31 | 76 | 0.48 |

Best Practice: Recommend 16-30mm length with stepped gating systems

1.5 Shell Mold Interface Optimization

Critical Control Points:

Gate area shell thickness ≥3.5mm (using ZrO₂ surface layer)

Avoid weak zones (porosity >12%)

Thermal conductivity matching: Gate zone conductivity should be 0.8-1.2W/m·K higher than cavity zone

1.6 Function-Oriented Design

Differentiated Design Parameters:

Product Type | Minimum Gate Cross-section (mm²) | Flow Velocity (m/s) |

|--------------------|----------------------------------|---------------------|

| Pressure Vessels | 120-150 | 5.2-6.8 |

| Thin-Wall (<3mm) | 60-80 | 3.5-4.2 |

| Precision Gears | 90-110 | 4.7-5.5 |

Chapter 2: Dimensional Accuracy Control Technology

2.1 Shrinkage Rate Prediction Model

Multi-Factor Coupling Formula:

ε=ε₀×(1+0.003(Tₘ-1550))×(1-0.05(P-0.5))×(1+0.02(t-24))

Where:

ε: Actual shrinkage rate (%)

Tₘ: Pouring temperature (℃)

P: Shell permeability (cm³/cm²·s)

t: Drying time (h)

Validation Data: A valve body prediction value of 1.68% vs actual 1.72% (2.4% error)

2.2 Staged Shrinkage Control

| Process Stage | Typical Shrinkage Range (%) | Control Methods |

|---------------------|-----------------------------|--------------------------------|

| Wax Pattern | 0.3-0.6 | Isothermal wax pressing (45±2℃)|

| Metal Solidification| 1.2-2.5 | Directional cooling (≥150℃/s gradient)|

| Machining Allowance | 0.1-0.3 | 3D scanning compensation |

2.3 Process Stability Assurance

SPC Control Chart Parameters:

Critical Characteristics: Cavity Dimension CPK≥1.33

Monitoring Frequency: 3 key dimensions per batch

Correction Threshold: Single point beyond 3σ triggers 5Why analysis

Chapter 3: Dewaxing Process Optimization

3.1 Thermodynamic Dewaxing

Pressure-Time Curve Optimization:

Staged pressurization strategy: 0.2MPa→0.4MPa→0.6MPa (15s per stage)

Wax melting efficiency increased by 23%, crack occurrence reduced to 0.7%

3.2 Shell Strength Matching

Dewaxing Safety Assessment:

Tensile strength: ≥5MPa (three-point bending test)

Drying degree indicator: Moisture ≤0.3% (microwave moisture meter)

Special structure handling: Deep-hole components require 48h extended drying and multi-point exhaust design

Chapter 4: Deformation Defect Control System

4.1 Full-Process Stress Analysis

Stress Accumulation Stage Model:

Wax Pattern Stage → Shell Making → Dewaxing → Pouring → Cooling

↑ ↑ ↑ ↑ ↑

0.12MPa 0.35MPa 0.28MPa 0.67MPa 0.41MPa

4.2 Anti-Deformation Technology Matrix

| Process Stage | Control Measures | Effectiveness |

|---------------------|------------------------------------|---------------------------------|

| Wax Pattern | Pre-deformation ribs (0.15mm offset)| Reduces post-correction workload by 65% |

| Shell Making | Laser-aligned gate assembly (±0.05mm)| Reduces assembly stress by 28% |

| Pouring | Electromagnetic stirring + directional solidification | Grain orientation deviation ≤12° |

| Post-Processing | Hydraulic correction (0.8MPa/mm gradient) | Rebound control <0.03mm |

Chapter 5: Process Standardization Management

5.1 Standardized Document System

Four-Level Document Architecture:

1. Process Flow Diagram (PFD)

2. Parameter Control Matrix (PCM)

3. Work Instruction (WI)

4. Inspection Specification (IS)

5.2 Change Management Process

ECN (Engineering Change Notice) Standards:

Impact Assessment: FMEA analysis (RPN≥60 requires verification testing)

Transition Period: Old stock digestion ≤3 production batches

Document Update: PLM system update within 24h

Conclusion and Future Outlook

Through establishing a 17-point full-process control system, typical product yield improved from 78.6% to 92.4%. Future development focuses on:

1. Digital Twin Simulation Platform (Target prediction accuracy >95%)

2. Adaptive Pouring System (Real-time parameter adjustment)

3. AI-Based Dimensional Compensation Algorithm

This research has formed enterprise standard Q/UWT 001-2025 and is applying for 2 invention patents.