Optimization of Higher-Order Harmonic Surface Tessellations for Additively Manufactured Air-to-Air Heat Exchangers

This study demonstrates that an optimized higher-order harmonic surface tessellation, developed through an analytical and numerical framework, outperforms conventional gyroid TPMS structures in turbulent flow regimes by achieving a superior balance of high thermal effectiveness and lower pressure drop, with secondary surface wave frequency identified as the critical design parameter.

The Gist

Imagine you are trying to cool down a hot cup of coffee by blowing air over it. You want the air to grab as much heat as possible from the coffee, but you also don't want to use a giant, noisy fan that burns electricity just to push the air through. This is the daily struggle of engineers designing air-to-air heat exchangers—devices that swap heat between two streams of air (like in your home's ventilation system or an airplane's cabin) without letting the air mix.

For a long time, engineers have been trying to make these devices more efficient. The current "gold standard" for the internal shape of these devices is something called a Gyroid. Think of a Gyroid like a complex, 3D sponge or a twisted, endless maze. It has a huge amount of surface area packed into a tiny space, which is great for grabbing heat. However, because the air has to twist and turn through this maze, it gets stuck and fights against the walls, creating a lot of drag (like trying to run through waist-deep water). This requires powerful fans to push the air through, which uses a lot of energy.

The New Idea: The "Harmonic Wave"

The researchers in this paper asked a simple question: Can we make a surface that grabs heat almost as well as the complex Gyroid maze, but lets the air flow through much more easily?

They came up with a new design based on higher-order harmonic waves.

• The Analogy: Imagine the Gyroid is a dense, twisting forest where you have to weave through every tree. The new design is more like a field of gentle, rolling hills with small, rhythmic ripples on the surface. It's still wavy and textured, but the path is straighter and less chaotic.

They used Additive Manufacturing (3D printing) to build these shapes. Since 3D printing can make almost any shape, they could create these precise, wavy patterns that would be impossible to make with traditional metal stamping.

How They Tested It

The team used powerful computer simulations (like a virtual wind tunnel) to test their new "wavy" design against the classic "Gyroid" maze. They looked at two main things:

1. Effectiveness: How much heat did the air actually pick up?
2. Pressure Drop: How hard did the fan have to work to push the air through?

The Surprising Results

Here is what they found, broken down by how the air was moving:

1. When the air is moving slowly (Laminar Flow):

• The Gyroid Maze: It grabbed heat very well, but it was a nightmare for the fan. The air had to fight so hard to get through the twists that the pressure drop was massive.
• The New Wave: It grabbed slightly less heat than the Gyroid, but the air flowed through it like a breeze.
• The Winner: The new wave design was the clear winner. It offered the best "bang for your buck." It was about 1.7 to 2.2 times better overall because it saved so much energy on the fan while still doing a great job at heat transfer.

2. When the air is moving fast (Turbulent Flow):

• The Gyroid Maze: At high speeds, the Gyroid actually started to perform better at grabbing heat, but it still demanded a lot of energy from the fan.
• The New Wave: At very high speeds, the new wave design actually started to catch up and even surpass the Gyroid in overall efficiency. It managed to grab more heat and kept the pressure drop lower than the Gyroid.
• The Winner: The new wave design held its own and became the superior choice for high-speed applications because it didn't require a massive fan to overcome resistance.

The "Secret Sauce": Frequency vs. Amplitude

The researchers also played with the "knobs" of their design:

• Amplitude (Height of the wave): Making the waves taller did increase heat transfer, but it also made the air struggle more.
• Frequency (How many waves fit in the space): They found that having more, smaller waves (higher frequency) was the magic trick. It increased heat transfer significantly without making the air struggle as much as just making the waves taller.

The Bottom Line

Think of this research as finding a better way to design the "lungs" of a building or a machine.

• Old Way (Gyroid): Like a complex, twisting maze. Great at catching things, but exhausting to run through.
• New Way (Harmonic Waves): Like a well-designed, gently rippling river. It catches things almost as well, but you can move through it with much less effort.

Why does this matter?
This new design means we can build heat exchangers that are:

• Smaller and lighter (great for airplanes and electronics).
• More energy-efficient (less electricity needed for fans).
• Cheaper to run over time.

By using 3D printing to create these specific, mathematically optimized waves, engineers can finally get the best of both worlds: high performance without the high energy cost. It's a step toward smarter, greener, and more efficient technology for everything from your home's air conditioner to the cooling systems of future electric vehicles.

Technical Summary

1. Problem Statement

Air-to-air heat exchangers (AAHX) are critical for energy recovery and thermal management but face a persistent trade-off: conventional designs often suffer from low effectiveness, while advanced nature-inspired geometries (like Triply Periodic Minimal Surfaces or TPMS, specifically the Gyroid structure) enhance heat transfer at the cost of excessive pressure drops and pumping power.

• Limitations of Current Designs: While Gyroid lattices offer high surface-area-to-volume ratios, their complex, tortuous flow paths induce high form drag and pressure losses, particularly in turbulent flow regimes.
• Operational Gaps: Many existing studies rely on idealized steady-state or laminar assumptions, failing to capture the complexities of real-world turbulent flows, flow maldistribution, and the dynamic balance between heat transfer enhancement and flow resistance.
• Goal: The study aims to develop a novel, additively manufacturable heat exchanger geometry that balances high thermal effectiveness with manageable pressure drops across a wide range of operating conditions (laminar to turbulent).

2. Methodology

The researchers employed a multi-faceted approach combining analytical modeling, numerical simulation, and optimization algorithms.

• Geometry Definition:
  • Proposed Design: A second-order harmonic surface tessellation. The geometry is defined analytically by the equation:
    $$z = A \sin(n\pi x) + B \cos(m\pi y) \cos(m\pi x) e^{-k\sqrt{x^2+y^2}} + \epsilon$$
    Where $A$ and $B$ are amplitude coefficients, $n$ and $m$ are spatial frequencies, and $k$ controls localized undulations (dimples/ridges) to induce vortices.
  • Benchmark Design: A standard Gyroid TPMS structure obtained from the COMSOL CAD library, serving as the reference for comparison.
• Simulation Setup:
  • Software: COMSOL Multiphysics coupled with MATLAB.
  • Physics: Conjugate Heat Transfer (CHT) using the Shear Stress Transport (SST) $k-\omega$ Reynolds-Averaged Navier-Stokes (RANS) model for turbulent flow and laminar equations for low Reynolds numbers.
  • Domain: A unit cell approach with periodic boundary conditions. Air properties were modeled as temperature-dependent polynomials; fins were modeled as aluminum.
  • Mesh: High-resolution tetrahedral meshes with boundary layer refinement to capture wall effects ($y^+$ validation).
• Optimization Framework:
  • Objective: Maximize effectiveness ($\varepsilon$) while constraining pressure drop ($\Delta P$) and hydraulic diameter ($D_h$).
  • Algorithm: A surrogate optimization solver (surrogateopt in MATLAB) coupled with COMSOL via LiveLink.
  • Constraints: $\Delta P \leq 5.07$ Pa, $D_h \in [2.0, 2.2]$ cm, and manufacturing feasibility limits on amplitude and frequency parameters.

3. Key Contributions

1. Novel Geometry: Introduction of a higher-order harmonic tessellation as a manufacturable alternative to complex TPMS structures, offering tunable surface features (amplitude and frequency) to control flow dynamics.
2. Parametric Sensitivity Analysis: Identification that secondary surface wave frequency ($m$) is a more critical control parameter than amplitude ($B$) for optimizing thermal-hydraulic performance. Increasing frequency yields significant effectiveness gains with marginal pressure drop penalties compared to amplitude increases.
3. Comprehensive Performance Metrics: Development of specific correlations for Nusselt number ($Nu$) and Fanning friction factor ($f$) for both the harmonic and Gyroid structures across laminar ($Re \leq 2000$) and turbulent ($Re \geq 3000$) regimes.
4. Comparative Benchmarking: A rigorous comparison against the Gyroid structure using the London Area Goodness Factor (AGF, $j/f$) and Thermal Enhancement Factor (TEF), revealing distinct performance regimes where each design excels.

4. Key Results

• Optimization Outcome:

  • The optimized harmonic structure achieved an effectiveness of $\varepsilon \approx 13.94$ with a pressure drop of 0.42 Pa (well below the 5.07 Pa limit) at $Re=1000$.
  • Sensitivity analysis showed that increasing the secondary wave frequency ($m$) from 0 to 8 increased effectiveness by ~59% with only a ~31% increase in pressure drop, whereas increasing amplitude ($B$) by 5x increased effectiveness by ~70% but spiked pressure drop by ~170%.

• Performance vs. Reynolds Number:

  • Laminar Regime ($Re \leq 2000$): The Gyroid structure showed higher raw effectiveness and Nusselt numbers (approx. 5.2x higher $Nu$). However, it suffered from significantly higher flow resistance (approx. 8.4x higher friction factor). Consequently, the Harmonic structure had a superior AGF (1.72x higher), indicating better overall efficiency per unit of pumping power.
  • Transition/Turbulent Regime ($Re \geq 7000$): The performance crossover occurred at $Re \approx 7000$. Beyond this point, the Harmonic structure outperformed the Gyroid in both effectiveness and pressure drop. The Gyroid's tortuous flow path caused excessive pressure losses that outweighed its heat transfer benefits in high-speed flows.
  • Pressure Drop: The Gyroid structure consistently exhibited higher pressure drops, averaging 10.5x higher in laminar flow and 2.75x higher in turbulent flow compared to the harmonic design.

• Flow Physics:

  • The harmonic structure induces controlled local vortices near the walls, thinning the thermal boundary layer and enhancing mixing without creating the severe stagnation zones and flow acceleration/deceleration seen in the Gyroid structure.

5. Significance and Implications

• Design Paradigm Shift: The study demonstrates that complex TPMS geometries are not universally superior. For air-to-air applications, especially those operating in turbulent regimes or where pumping power is a constraint, simpler, optimized harmonic surfaces can offer a better balance of performance and manufacturability.
• Additive Manufacturing (AM) Viability: The proposed harmonic geometry is highly compatible with AM techniques (e.g., SLS, SLA), offering a scalable solution for compact, lightweight heat exchangers in building ventilation, aerospace, and electronics cooling.
• Operational Guidance: The findings provide a clear decision matrix for engineers:
  • Use Gyroid/TPMS for low-flow, laminar applications where maximum heat transfer density is the sole priority and pumping power is less constrained.
  • Use Optimized Harmonic Tessellations for high-flow, turbulent applications or systems requiring low pressure drops and high overall thermal-hydraulic efficiency (AGF).

In conclusion, this research validates a "balanced pathway" for next-generation heat exchangers, proving that tailored surface modifications can achieve high performance without the prohibitive pressure penalties associated with traditional TPMS lattices.