Laminar and Turbulent Flow in Wavy Pipes under Strong Wall Modulations

This study utilizes direct numerical simulations to demonstrate that strong axial wall modulations in pipes induce early flow reversal, subcritical turbulence transitions, and fully rough turbulent regimes, necessitating the use of hydrodynamic concepts like effective hydraulic radius and sandgrain roughness to accurately characterize friction and stability across all flow regimes where classical models like the Moody diagram fail.

The Gist

Imagine you are driving a car down a highway. On a perfectly smooth, straight road, your car moves efficiently, and you can predict exactly how much gas you'll need to get to your destination. This is what engineers call "smooth pipe flow."

But now, imagine that same highway is built on a giant, rolling wave. The road goes up and down in a perfect, rhythmic pattern. Sometimes the road narrows into a tight tunnel; other times it widens into a massive open plaza. This is the world of wavy pipes, and it's the subject of this new research.

The scientists behind this study wanted to understand how water (or any fluid) behaves when it's forced to travel through these bumpy, wave-like tunnels. They used powerful computer simulations to watch the water move, looking at everything from slow, gentle flows to fast, chaotic turbulence.

Here is what they discovered, broken down into simple concepts:

1. The "Traffic Jam" in the Laminar Zone (Slow Flow)

When water flows slowly, it usually moves in smooth, parallel layers (like a calm river). In a smooth pipe, we have a perfect formula to predict how hard it is to push the water through.

The Discovery: In a wavy pipe, even at slow speeds, the water gets confused. As the pipe narrows and then widens, the water doesn't just squeeze through; it actually starts to spin in circles (recirculation) in the widening sections, like a car spinning out on a sharp turn.

• The Analogy: Imagine a crowd of people walking down a hallway that suddenly widens. Instead of just walking forward, some people get confused, stop, and start walking in circles near the walls. This creates a "traffic jam" that makes it much harder to push the crowd forward.
• The Result: The water experiences much more friction (resistance) than we expected. The old formulas failed because they didn't account for these little "eddies" or spinning zones. The researchers had to invent a new way to measure the "effective size" of the pipe to get the math right.

2. The "Tipping Point" (Transition to Turbulence)

In a smooth pipe, water usually stays calm and orderly until it reaches a very high speed (Reynolds number). Only then does it suddenly turn chaotic and turbulent (like a whitewater rapid).

The Discovery: In wavy pipes, the water loses its cool much earlier. The waves themselves act as a constant nudge, shaking the water and making it unstable.

• The Analogy: Think of a tightrope walker. On a calm day, they can walk a long way without falling. But if the tightrope is constantly swaying back and forth (the wavy pipe), the walker will lose balance and fall much sooner, even if they are walking slowly.
• The Result: Turbulence can start at speeds as low as 25 (in their specific units), whereas a smooth pipe would stay calm until 2,000. The waves trigger the chaos early.

3. The "Rocky River" (Turbulent Flow)

Once the water is moving fast and is fully turbulent, the behavior changes again. In a smooth pipe, the friction depends on how fast the water is moving.

The Discovery: In a wavy pipe with strong waves, the water becomes "fully rough." It doesn't matter how fast you go anymore; the friction is dominated by the shape of the walls. The water crashes against the waves, creating massive pressure drag.

• The Analogy: Imagine running through a forest. If the trees are small (smooth pipe), running faster makes you work harder. But if the forest is filled with giant boulders (strong waves), it doesn't matter how fast you run; you are constantly bumping into rocks. The "boulders" (the waves) dictate the difficulty, not your speed.
• The Result: The researchers found that the height of the wave (the amplitude) is the single most important factor. They could use this height to predict exactly how much resistance the water would face, similar to how we measure the roughness of sandpaper.

Why Does This Matter?

You might wonder, "Who cares about wavy pipes?"

Actually, nature is full of them!

• Karst Caves: Water flows through limestone caves that have been eaten away by acid, creating irregular, wavy tunnels.
• Stented Arteries: Medical stents placed in human arteries often have a wavy structure.
• Ventilation Ducts: Many air ducts are corrugated to be flexible.

The Big Takeaway:
The famous "Moody Diagram" (a chart engineers have used for 80 years to calculate pipe friction) assumes pipes are either smooth or have tiny, random bumps like sand. It fails when the pipe has big, structured waves.

This study tells us that for these wavy systems, we can't just use the old rules. We need new "hydrodynamic concepts" (like the effective radius and equivalent roughness) that understand that the shape of the wave itself creates the resistance, not just the surface texture.

In a nutshell: If you are designing a system with wavy pipes (or studying a cave), don't trust the old textbooks. The waves create their own unique physics, causing early chaos and extra drag that only a new set of rules can explain.

Technical Summary

1. Problem Statement

Real-world conduits (e.g., karst caves, stented arteries, ventilation ducts) often exhibit significant geometric variations, including large-scale wall undulations, rather than just small-scale surface roughness. Classical hydraulic models, such as the Hagen-Poiseuille law for laminar flow and the Moody diagram for turbulent flow, rely on assumptions of smooth or uniformly rough pipes with small-scale perturbations. These models fail when:

• Wall amplitude is large: The roughness significantly alters the cross-sectional area and flow topology.
• Flow separation occurs: Large-scale undulations induce flow reversal and recirculation zones even at low Reynolds numbers.
• Transition mechanisms differ: The deterministic geometry of wavy walls may trigger turbulence at Reynolds numbers far below the classical critical threshold ($Re \approx 2300$) for smooth pipes.

The study aims to quantify the hydrodynamic impact of strong, periodic wall modulations on laminar, transitional, and turbulent flow regimes, specifically addressing the limitations of current friction factor correlations and the definition of equivalent roughness.

2. Methodology

The authors employed Direct Numerical Simulations (DNS) to solve the incompressible Navier-Stokes equations without turbulence modeling.

• Geometry: A circular pipe with an axisymmetric sinusoidal wall defined by $R(z) = R_{max} - \frac{k}{2}[1 - \sin(2\pi z/\lambda)]$.
  • $R_{max}$: Maximum radius.
  • $k$: Peak-to-peak wall amplitude (geometric roughness).
  • $\lambda$: Wavelength (fixed at $2R_{max}$).
  • The study varied the relative roughness $R/k$ (inverse of relative roughness) across five cases: $0, 10, 5, 10/3, 2$.
• Numerical Setup:
  • Solver: Nek5000 (spectral element method).
  • Domain: Pipe length $L_z = 14R_{max}$ (7 wavelengths).
  • Boundary Conditions: Periodic in the streamwise direction with a uniform forcing term to maintain constant bulk velocity; no-slip at the wall.
  • Reynolds Numbers: Bulk Reynolds numbers ($Re_b$) ranging from 50 to 5300, covering laminar, transitional, and turbulent regimes.
  • Resolution: Grids were refined near the wall to resolve viscous sublayers and recirculation zones, with wall-unit spacings ($\Delta r^+, \Delta \theta^+, \Delta z^+$) meeting standard DNS criteria.
• Key Metrics:
  • Darcy friction factor ($f$).
  • Mean velocity statistics and Reynolds stresses.
  • Roughness function ($\Delta U^+$) and equivalent sand-grain roughness ($k_s$).

3. Key Contributions

1. Redefinition of Hydraulic Radius in Laminar Flow: The paper demonstrates that classical geometric averages (arithmetic or harmonic mean radius) fail to predict friction in wavy pipes. It introduces an effective hydraulic radius ($R_h$) derived from the actual pressure drop, which collapses laminar data onto the Hagen-Poiseuille line ($f = 64/Re$).
2. Mechanism of Early Transition: The study identifies that strong wall modulations induce flow reversal and recirculation at very low Reynolds numbers ($Re_b \approx 25$ for high amplitude). These recirculation zones act as intrinsic finite-amplitude disturbances, triggering subcritical transitions to turbulence at $Re_b$ between 500 and 1000, well below the smooth-pipe threshold.
3. Scaling of Critical Reynolds Number: A power-law scaling for the critical Reynolds number ($Re_c$) is proposed: $Re_c \sim (k/R_h)^{-1/\gamma}$, with $\gamma$ between $5/4$ and $3/2$. This aligns with finite-amplitude transition scenarios, confirming that the wall geometry itself provides the necessary disturbance to destabilize the flow.
4. Hydrodynamic Roughness in Turbulent Regime: In the fully rough turbulent regime, the flow becomes independent of Reynolds number and is dominated by inertial separation and pressure drag. The study validates that the geometric amplitude ($k$) serves as a robust estimator for the equivalent sand-grain roughness ($k_s$), with a relationship $k_s \approx 1.5k$.

4. Key Results

Laminar Regime ($Re_b \lesssim 2000$)

• Friction Factor: Classical models overestimate the effective area, leading to under-predicted friction. The actual friction factor is 5–80% higher than the Hagen-Poiseuille prediction based on $R_{max}$.
• Flow Structure: As $Re_b$ increases, adverse pressure gradients in diverging sections cause flow separation and the formation of steady recirculation bubbles.
  • For $R/k = 10$, separation occurs at $Re_b \approx 1000$.
  • For $R/k = 2$, separation occurs as early as $Re_b \approx 25$.
• Effective Radius: The effective hydraulic radius $R_h$ decreases much faster than the arithmetic mean as $k$ increases. A slope-corrected harmonic radius model ($R_\sigma$) is proposed to approximate $R_h$ with high accuracy.

Transitional Regime

• Onset of Turbulence: Turbulence is triggered by the interaction between the laminar base flow and the deterministic disturbances caused by the wavy wall (recirculation zones).
• Scaling: The transition boundary follows a subcritical scaling law. The critical Reynolds number decreases as the wall amplitude increases, consistent with bypass transition mechanisms where the disturbance amplitude is fixed by geometry rather than external noise.

Turbulent Regime ($Re_b > 1000$)

• Fully Rough Behavior: At high Reynolds numbers, the flow enters a "fully rough" regime where the friction factor becomes nearly independent of $Re$.
• Velocity Profiles: Mean velocity profiles exhibit a downward shift ($\Delta U^+$) relative to smooth pipes. When scaled by the equivalent sand-grain roughness $k_s$, the profiles collapse onto a universal log-law.
• Equivalent Roughness: The study finds that $k_s \approx 1.5k$ (or $k_s \approx k$ for milder waviness). This confirms that for large-scale periodic modulations, the peak-to-peak amplitude is the dominant length scale controlling drag, rather than statistical roughness parameters like RMS height used for random roughness.
• Outer Layer Similarity: Despite the large-scale geometry, Townsend's outer-layer similarity hypothesis holds when the coordinate is shifted by a virtual origin ($\ell$) and scaled by $k_s$.

5. Significance and Implications

• Limitations of the Moody Diagram: The study conclusively shows that the Moody diagram and standard empirical correlations (like Colebrook) are insufficient for conduits with strong, large-scale wall modulations. These models assume roughness is a small perturbation, whereas here, roughness defines the global flow topology.
• Karst Hydrology and Natural Conduits: The findings are directly applicable to karst aquifers and fractured rocks, where dissolution creates large, wavy channels. The study explains why flow resistance in these systems is often higher than predicted and why turbulence can occur at much lower flow rates than in smooth pipes.
• Engineering Design: For applications involving corrugated pipes (heat exchangers, stents), relying on geometric averages for pressure drop calculations will lead to significant errors. The proposed effective hydraulic radius ($R_h$) for laminar flow and equivalent sand-grain roughness ($k_s \approx 1.5k$) for turbulent flow provide robust, physics-based parameters for design.
• Theoretical Advancement: The work bridges the gap between "roughness" (small-scale texture) and "waviness" (large-scale geometry), demonstrating that in the limit of strong modulation, the distinction blurs, and the geometry itself acts as the primary driver of hydrodynamic resistance and instability.