Introduction: The Quest for Aerodynamic Perfection

What if you could combine the lightweight agility of a Mazda Miata MX-5 with the aerodynamic sophistication of a McLaren Speedtail? That's the vision behind Spiata—a mini-supercar design project that merges iconic automotive design with cutting-edge computational fluid dynamics (CFD) analysis.

For the Thermofluids module in Year 2 at Imperial College London, we were tasked with creating a vehicle using CAD software and validating its aerodynamic performance through CFD simulation. The goal? Calculate the drag coefficient and compare it with real-world wind tunnel data of similar vehicles.

Design Philosophy

Spiata takes inspiration from two automotive legends:

• Mazda Miata MX-5: The world's best-selling roadster, known for lightweight design and driving purity
• McLaren Speedtail: The most aerodynamic car ever produced by McLaren

The challenge was to create a mini-supercar that captures the essence of both—nimble like a Miata, yet slippery through the air like a Speedtail.

Design Process: From Concept to CAD

Phase 1: Conceptualization

Before touching CAD software, I studied the aerodynamic features of both inspirational vehicles:

From the Miata MX-5:

• Low-slung profile
• Compact dimensions
• Simple, organic curves
• Minimal frontal area

From the McLaren Speedtail:

• Teardrop-shaped body (optimal for drag reduction)
• Smooth underbody
• Streamlined rear with integrated airflow management
• Covered rear wheels (reducing turbulence)

The goal: Combine these principles into a cohesive design that prioritizes aerodynamic efficiency.

Phase 2: Surface Modeling in Fusion 360

I used Fusion 360 for the CAD design, employing advanced surface modeling techniques—a powerful method for creating complex, organic automotive shapes.

Surface Modeling vs. Solid Modeling

Why surface modeling?

Traditional solid modeling (extrude, revolve, sweep) works well for mechanical parts but struggles with the flowing curves of automotive design. Surface modeling allows:

• Freeform organic shapes: Create compound curves found in car bodies
• Precise control: Manipulate individual surfaces for exact aerodynamic profiles
• Complex curvature: Achieve the smooth transitions critical for airflow

Key Modeling Techniques Used

1. Loft Surfaces

Created the main body by lofting between profile curves:

Front profile → Mid-body sections → Rear taper

This technique ensures smooth transitions and continuous curvature—essential for laminar airflow.

2. Sculpting Tools

Used Fusion 360's sculpt mode to refine:

• Hood contours
• Roof line
• Side character lines
• Rear deck tapering

3. Patch and Trim

• Filled surface gaps with patch surfaces
• Trimmed intersecting surfaces for clean body panels
• Ensured G2 continuity (curvature continuity) at surface junctions

Design Features for Aerodynamics

Front End:

• Low nose to reduce frontal area
• Smooth integration of headlight volumes
• Gentle slope to direct air over the hood

Side Profile:

• Streamlined cabin with minimal tumblehome (inward taper)
• Flush door handles (modeled as surface depressions)
• Covered rear wheel arches (reducing drag by 3-5%)

Rear Design:

• Extended tail inspired by Speedtail's teardrop shape
• Integrated spoiler (subtle, not aggressive)
• Smooth underbody transition to prevent flow separation

Phase 3: Model Preparation for CFD

Before CFD analysis, the CAD model required specific preparation:

1. Watertight Geometry

• Ensured all surfaces formed a closed volume
• No gaps or overlapping faces
• Critical for CFD mesh generation

2. Simplified Details

• Removed minor features (door handles, badges)
• Focused on aerodynamically significant geometry
• Reduced computational load

3. Export to Solidworks

• Converted Fusion 360 model to STEP format
• Imported into Solidworks Flow Simulation
• Verified scale and units

CFD Analysis: Computational Fluid Dynamics

Setting Up the Simulation

Software: Solidworks Flow Simulation

Simulation Parameters:

• Fluid: Air at standard atmospheric conditions (1.225 kg/m³)
• Velocity: 30 m/s (~67 mph / 108 km/h)
• Reynolds Number: ~4.5 million (turbulent flow regime)
• Turbulence Model: k-ε (standard for external aerodynamics)
• Boundary Conditions:
  • Velocity inlet
  • Pressure outlet
  • No-slip wall on car surface
  • Symmetry plane (half-car simulation for efficiency)

Meshing Strategy

Created a computational mesh with:

• Base mesh: 1.2 million cells
• Refinement zones:
  • Around car body (boundary layer capture)
  • Wake region (turbulence resolution)
  • Near wheels (complex flow features)
• Boundary layer: 10 layers with inflation (y+ < 1)

Results and Analysis

Flow Visualization

1. Side Flow Trajectories

The streamline visualization revealed:

• Smooth airflow over the hood and roof
• Minimal flow separation until the rear deck
• Well-attached flow along the Speedtail-inspired tail

2. Pressure Distribution

2D Pressure Density Plot showed:

• High pressure (red/yellow) at the nose: Stagnation point
• Low pressure (green/blue) over the roof: Accelerated flow
• Pressure recovery at the rear: Gradual (good for low drag)

Key observation: The gentle pressure gradient at the rear indicates effective flow management—abrupt pressure changes cause separation and drag.

3. Velocity Field

2D Velocity Plot demonstrated:

• Flow acceleration over the roof (Venturi effect)
• Boundary layer growth along the body
• Wake velocity deficit (indicates drag)

4. Isometric Flow Trajectories

The 3D visualization highlighted:

• Symmetric flow pattern (good design balance)
• Vortex formation at rear wheel arches
• Clean wake with minimal turbulence

Drag Coefficient Calculation

From the CFD simulation:

Drag Force (F_d): 185 N

Drag Coefficient Formula:

C_d = F_d / (0.5 × ρ × V² × A)

Where:

• F_d = Drag force (185 N)
• ρ = Air density (1.225 kg/m³)
• V = Velocity (30 m/s)
• A = Frontal area (1.2 m²)

Calculated C_d: 0.28

Performance Benchmarking

Comparing Spiata's C_d = 0.28 with real vehicles:

| Vehicle | Drag Coefficient | |---------|-----------------| | Spiata (CFD) | 0.28 | | McLaren Speedtail | 0.278 | | Mazda Miata MX-5 (ND) | 0.31 | | Porsche 911 | 0.29 | | Tesla Model S | 0.24 | | Average sedan | 0.30-0.35 |

Result: Spiata achieves supercar-level aerodynamics, nearly matching the McLaren Speedtail!

Key Aerodynamic Features That Worked

1. Teardrop Profile

The elongated rear (Speedtail influence) allowed gradual pressure recovery, reducing form drag significantly.

2. Smooth Underbody

Flat underbody with minimal protrusions prevented turbulent airflow beneath the car—a major drag contributor.

3. Covered Rear Wheels

Reduced wheel-induced turbulence. Open wheels can contribute 10-15% of total drag.

4. Optimized Frontal Area

Small frontal area (1.2 m²) combined with streamlined shape minimized drag force.

Technical Deep Dive: CFD Theory

Understanding Drag Components

Total drag consists of:

1. Form Drag (Pressure Drag)

• Caused by flow separation
• Creates wake region with low pressure
• Dominant at high speeds

2. Skin Friction Drag

• From boundary layer shear
• Proportional to surface area
• Reduced with smooth surfaces

3. Induced Drag

• From lift generation (minimal for ground vehicles)
• Relevant at very high speeds

Spiata's Drag Breakdown (estimated from CFD):

• Form drag: ~65%
• Skin friction: ~30%
• Other (wheels, mirrors): ~5%

Reynolds Number and Turbulence

Re = (ρ × V × L) / μ

For Spiata:

• L (characteristic length) ≈ 3.2 m
• μ (dynamic viscosity) = 1.81 × 10⁻⁵ Pa·s
• Re ≈ 6.5 million (fully turbulent)

This high Reynolds number confirms:

• Turbulent boundary layer over most of the car
• k-ε turbulence model is appropriate
• Results are representative of real-world conditions

Mesh Independence Study

To ensure result accuracy, I performed a mesh sensitivity analysis:

| Mesh Density | Cell Count | C_d | Δ C_d | |--------------|-----------|-----|-------| | Coarse | 600k | 0.292 | - | | Medium | 1.2M | 0.280 | -4.1% | | Fine | 2.4M | 0.278 | -0.7% |

The change from medium to fine mesh is <1%, confirming mesh independence at 1.2M cells.

Challenges and Learnings

Challenge 1: Surface Continuity

Problem: Initial CAD model had G1 continuity (tangent) but not G2 (curvature)

Impact: CFD showed unexpected flow separation at surface transitions

Solution:

• Rebuilt critical surfaces using spline curves
• Ensured G2 continuity using Fusion 360's curvature analysis
• Re-ran CFD: C_d improved from 0.32 to 0.28

Challenge 2: Mesh Quality

Problem: Poor mesh near wheel arches caused convergence issues

Impact: Simulation failed to converge after 500 iterations

Solution:

• Increased local refinement around wheels
• Added boundary layer inflation
• Adjusted mesh growth rate from 1.5 to 1.2
• Achieved convergence in 280 iterations

Challenge 3: Computational Resources

Problem: Full-car CFD simulation estimated 18+ hours on available hardware

Solution:

• Used symmetry plane (half-car model)
• Reduced domain size while maintaining 3× car length wake
• Simulation time: 4.5 hours (75% reduction)

Results and Impact

Academic Performance

• Drag coefficient: 0.28 (excellent for a compact vehicle)
• Validation: Within 5% of similar production vehicles
• Grade achievement: High marks in Thermofluids module

Skills Developed

CAD Proficiency: ✅ Advanced surface modeling techniques ✅ Automotive design principles ✅ G-continuity management

CFD Expertise: ✅ Simulation setup and boundary conditions ✅ Mesh generation and refinement ✅ Results interpretation and validation

Engineering Process: ✅ Iterative design optimization ✅ Performance benchmarking ✅ Technical documentation

Design Insights

Key learnings about automotive aerodynamics:

1. Small changes, big impact: 5° change in rear taper angle → 8% drag reduction
2. Underbody matters: Smooth underbody contributed 15% drag improvement
3. Flow visualization is crucial: CFD reveals non-intuitive flow behavior
4. Inspiration + innovation: Combining proven designs with original ideas

Future Improvements

If continuing this project, next steps would include:

Aerodynamic Refinements

1. Active aero: Deployable rear spoiler for high-speed stability
2. Diffuser design: Rear diffuser for downforce without drag penalty
3. Wheel fairings: Optimize covered rear wheel design

Advanced Analysis

1. Transient CFD: Study unsteady aerodynamics (crosswinds, overtaking)
2. Aero-acoustic: Predict wind noise
3. Thermal management: Integrate cooling airflow

Physical Validation

1. Wind tunnel testing: Validate CFD with scaled model
2. Coast-down testing: Real-world drag measurement
3. 3D printing: Create scaled physical model

Conclusion: Engineering Meets Art

Spiata represents the intersection of engineering analysis and automotive design. By combining iconic design elements from the Mazda Miata and McLaren Speedtail with rigorous CFD analysis, I created a mini-supercar concept that achieves:

✅ C_d = 0.28 - Supercar-level aerodynamics ✅ Validated design process - CFD confirms design decisions ✅ Aesthetic + performance - Beautiful form following aerodynamic function

Key Takeaways

For aspiring automotive engineers:

• Surface modeling is essential for organic shapes
• CFD is a powerful validation tool, not just analysis
• Iterative design leads to optimal solutions
• Real-world benchmarking provides context

For design enthusiasts:

• Aerodynamics shapes automotive beauty
• Every curve serves a purpose
• The best designs balance form and function

This project deepened my understanding of:

• CAD techniques for complex surfaces
• Fluid dynamics in automotive contexts
• Simulation methodology for design validation

Most importantly, it proved that with the right tools and knowledge, anyone can design vehicles that rival production supercars in aerodynamic efficiency.

---

Technical Details:

• Software: Fusion 360 (CAD), Solidworks Flow Simulation (CFD)
• Model: Half-car symmetry, 1.2M cells
• Conditions: 30 m/s, Standard atmosphere, k-ε turbulence
• Result: C_d = 0.28 (validated against benchmarks)

This project was completed as part of the Thermofluids module (Year 2) at Imperial College London.