Turbulent spots in hypersonic transitional planar and axisymmetric boundary layers

This study experimentally characterizes turbulent spots in hypersonic transitional boundary layers over flat plates and axisymmetric cones at Mach 5.85, revealing that while spot leading edges convect at 90% of the freestream speed for both geometries, planar spots exhibit slower trailing edges, faster streamwise growth, and generation rates between 1 and 3 million spots per meter per second compared to their axisymmetric counterparts.

The Gist

Imagine you are driving a car on a highway. Usually, the air flowing over your car is smooth and orderly, like a line of soldiers marching in perfect formation. This is called laminar flow. But sometimes, a single soldier gets confused, starts running wildly, and pulls a few others into the chaos. Suddenly, you have a small, swirling mess of traffic. In the world of physics, this chaotic mess is called a turbulent spot.

As the car (or in this case, a hypersonic vehicle) speeds up, these "traffic jams" of air start to form more often. They grow, they move, and eventually, they merge until the entire flow behind the vehicle becomes a chaotic, turbulent mess. This process is called transition.

This paper is like a detective story where scientists tried to catch these "traffic jams" (turbulent spots) in the act, specifically at hypersonic speeds (about 5.85 times the speed of sound!). They wanted to see if the shape of the vehicle matters: does a flat road (a flat plate) behave differently than a cone-shaped road (an axisymmetric cone)?

Here is the breakdown of their investigation using simple analogies:

1. The Setup: The Hypersonic Wind Tunnel

The researchers used a special machine called a Shock Tunnel (HST4). Think of this as a giant, high-speed wind tunnel that can simulate the air conditions a spacecraft would face when re-entering the atmosphere. They fired air at their test models at Mach 5.85.

They tested two shapes:

• The Flat Plate: Like a ruler lying flat.
• The Cone: Like a sharpened pencil tip.

Both were placed in the same "wind" so the only difference was their shape.

2. The Detective Work: "Thermal Cameras"

How do you see invisible air turbulence? You can't see it, but you can feel its heat. Turbulent air is hotter than smooth air.

The scientists glued tiny, super-sensitive heat sensors (thin-film sensors) onto the surface of their models. These sensors acted like thermal cameras.

• Smooth air: The sensor reads a steady, cool temperature.
• Turbulent spot passing by: The sensor suddenly spikes in temperature, then cools back down.

By watching these temperature spikes, they could tell exactly when a "turbulent spot" arrived and when it left.

3. The Findings: The Race Between Shapes

The scientists tracked these spots like race cars to see how fast they moved and how big they got. Here is what they discovered:

A. The Front Runner (Leading Edge)

Imagine the turbulent spot as a wave. The front of the wave is the "leading edge."

• The Discovery: Whether the spot was on the flat plate or the cone, the front of the wave moved at the same speed.
• The Analogy: It's like two runners starting a race. The front runner on the flat track and the front runner on the curved track are both running at 90% of the speed of the wind. They are equally fast at the front.

B. The Slowpoke (Trailing Edge)

Now look at the back of the wave, the "trailing edge."

• The Discovery: The back of the wave on the flat plate moved slower than the back of the wave on the cone.
• The Analogy: Imagine the turbulent spot is a stretchy rubber band being pulled forward.
  • On the cone, the rubber band stretches out evenly.
  • On the flat plate, the front pulls fast, but the back drags its feet. Because the front is fast and the back is slow, the rubber band stretches out much longer on the flat plate.

C. The Growth Rate

Because the front moves fast and the back moves slow on the flat plate, the "turbulent spot" gets longer much faster there than on the cone.

• The Result: The chaotic mess covers more distance on the flat plate in a shorter amount of time.

D. The Crowd Size (Generation Rate)

The scientists also counted how many of these "spots" were born per second.

• The Discovery: The flat plate was a "factory" that produced more spots than the cone.
• The Analogy: If the cone is a small town that occasionally has a traffic jam, the flat plate is a busy city intersection where traffic jams happen constantly.

4. The Big Picture: Why Does This Matter?

The main goal of this research is to help engineers design better hypersonic vehicles (like future space planes).

• The Problem: Turbulent air creates a lot of heat and drag. If a vehicle stays in "smooth air" (laminar) for too long, it might overheat or lose control when it finally transitions to turbulence.
• The Conclusion: The study found that on a flat surface, the transition from smooth to chaotic happens sooner and faster than on a cone.
  • The flat plate gets "messy" (turbulent) quickly because it spawns more spots, and those spots stretch out longer.
  • The cone stays "clean" (laminar) for a longer distance.

Summary

Think of the flat plate and the cone as two different types of roads.

• The Cone is a smooth, winding mountain road where traffic jams (turbulence) are rare and short-lived.
• The Flat Plate is a straight, busy highway where traffic jams happen often, and once they start, they stretch out for miles.

By understanding exactly how these "traffic jams" behave, engineers can design the shapes of future hypersonic vehicles to either delay the chaos (to save fuel) or manage the heat (to prevent melting). This paper gave them the specific data on how the shape of the vehicle changes the rules of the game.

Technical Summary

1. Problem Statement

The transition from laminar to turbulent flow in hypersonic boundary layers occurs through the sporadic formation and growth of isolated turbulence regions known as "turbulent spots." While the physics of these spots is well-understood in subsonic and low-supersonic regimes, there is a significant lack of comprehensive data regarding their characteristics in hypersonic flows (Mach > 5).

Specifically, there is a scarcity of comparative studies analyzing how geometry affects turbulent spot behavior under identical freestream conditions. Most existing literature focuses on either flat plates or cones in isolation. Understanding the differences between planar (flat plate) and axisymmetric (cone) boundary layers is critical for the design of hypersonic vehicles, as the transition location and length directly impact aerodynamic heating and vehicle performance. This study aims to fill this gap by characterizing turbulent spots on both geometries under the same hypersonic freestream environment.

2. Methodology

Experimental Facility:

• Experiments were conducted in the Hypersonic Shock Tunnel 4 (HST4) at the Indian Institute of Science (IISc).
• Flow Conditions: Mach number $M = 5.8 \pm 0.1$; Freestream Reynolds number ($Re_\infty$) ranging from $3.0$ to $6.0 \times 10^6$/m.
• Duration: Useful test time of 3.5 ms.

Test Models:

• Flat Plate: Sharp leading edge, length 800 mm, mounted at an angle of attack of $10^\circ \pm 0.5^\circ$ to match the edge Mach number of the cone.
• Axisymmetric Cone: Sharp tip, semi-angle $12^\circ$, axial length 800 mm.
• Both models were fabricated from duralumin with stainless steel tips/edges and a surface roughness of N10 ISO grade.

Instrumentation:

• Sensors: 20 platinum thin-film sensors were embedded on each model to measure wall heat transfer.
• Data Acquisition: Signals were acquired at 1.0 MSa/s using a NI PXI 6133 DAQ.
• Sensor Spacing: Varied from 10 mm to 70 mm to capture laminar, transitional, and turbulent states.

Data Analysis & Intermittency Calculation:

• Intermittency ($\gamma$): Defined as the fraction of time the flow is turbulent at a specific location.
• Detector Function ($D_i$): The authors proposed a modified detector function. Unlike previous methods (e.g., Clark et al., 1994) that scaled the time derivative of heat transfer against theoretical laminar/turbulent limits, this method scales the derivative against the difference between the local maximum and mean heat transfer values. This eliminates variance caused by different theoretical heat transfer estimation methods.
• Criterion Function ($C_i$): An exponential weighted moving average of the detector function was used to distinguish turbulent from non-turbulent regions.
• Spot Characterization: Leading edge ($U_{le}$) and trailing edge ($U_{te}$) speeds were calculated by tracking the time difference of spot passage across multiple sensors. Streamwise length scales were estimated using these speeds and time stamps.

3. Key Contributions

1. Comparative Analysis: First direct comparison of turbulent spot characteristics (speeds, growth rates, generation rates) between planar and axisymmetric hypersonic boundary layers under identical freestream conditions.
2. Methodological Improvement: Introduction of a modified detector function for intermittency calculation that reduces dependency on theoretical heat transfer models, improving robustness in experimental data processing.
3. Quantification of Spot Generation: Calculation of turbulent spot generation rates ($n$) for both geometries in the hypersonic regime, a parameter previously scarce in open literature for these conditions.

4. Key Results

Intermittency and Transition Onset:

• The spatial distribution of intermittency matched the universal intermittency distributions for both geometries.
• Transition Onset: The transition onset occurred earlier on the flat plate compared to the cone.
• Transition Length: The total length of the transitional boundary layer was shorter on the flat plate than on the cone.

Turbulent Spot Velocities:

• Leading Edge Speed ($U_{le}$): For both geometries, the leading edge of the turbulent spots convected at approximately 90% of the boundary layer edge speed ($U_e$). This speed was comparable for both the flat plate and the cone.
• Trailing Edge Speed ($U_{te}$): The trailing edge of spots on the flat plate moved significantly slower than those on the cone.

Streamwise Growth and Length Scales:

• Due to the slower trailing edge speed, turbulent spots on the flat plate exhibited a higher streamwise growth rate compared to those on the cone.
• Consequently, for the same longitudinal distance, spots on the flat plate occupied a larger area and merged more quickly.

Spot Generation Rates:

• The spot generation rate ($n$) was found to be in the range of $10^6$ to $3 \times 10^6$ spots/m/s.
• At the same unit edge Reynolds number, the flat plate generated a higher number of turbulent spots than the cone.
• As the Reynolds number increased, the generation rates for both geometries converged toward similar values.

5. Significance and Conclusions

The study provides a physical explanation for why transition occurs earlier and over a shorter distance on flat plates compared to cones in hypersonic flows:

1. Higher Receptivity: Planar boundary layers are more receptive to freestream noise, leading to earlier transition onset.
2. Higher Generation Rate: More spots are generated per unit length on the flat plate.
3. Faster Growth: The slower trailing edge speed on the flat plate causes spots to stretch more rapidly in the streamwise direction.

Conclusion: The combination of a higher generation rate and a faster streamwise growth rate on the flat plate leads to the rapid merging of spots, resulting in a fully developed turbulent boundary layer over a shorter distance compared to the axisymmetric cone. These findings are crucial for accurate prediction of aerodynamic heating and skin friction in the design of hypersonic vehicles, where geometry plays a pivotal role in transition management.