表面染料フロー可視化:超音速流のストリークラインパターンを観察する定性的方法

JoVE 과학 교육
Aeronautical Engineering
JoVE 비디오를 활용하시려면 도서관을 통한 기관 구독이 필요합니다.  전체 비디오를 보시려면 로그인하거나 무료 트라이얼을 시작하세요.
JoVE 과학 교육 Aeronautical Engineering
Surface Dye Flow Visualization: A Qualitative Method to Observe Streakline Patterns in Supersonic Flow

4,769 Views

08:12 min

April 30, 2023

개요

Flow visualization around or on a body is an important tool in aerodynamics research. It provides a method to qualitatively and quantitatively study flow structure, and it also helps researchers theorize and verify fluid flow behavior. Flow visualization can be divided into two categories: off-the-surface visualization and surface flow visualization. Off-the-surface flow visualization techniques involve determining the flow characteristics around the body of interest. They include but are not restricted to particle image velocimetry (PIV), Schlieren imaging, and smoke flow visualization. These techniques can provide qualitative as well as quantitative data on the flow around a body. However, these techniques are generally expensive and difficult to set-up. Surface flow visualization techniques, on the other hand, involve coating the body of interest with a dye to study flow on the surface. These techniques, which are more invasive in practice, include dye flow visualization and, more recently, use pressure-sensitive paint, which gives a detailed image of the flow on the body's surface. This allows researchers to visualize different flow features, including laminar bubbles, boundary layer transitions, and flow separation. Dye flow visualization, the technique of interest in the current experiment, provides a qualitative picture of the surface flow and is one of the simplest and most cost-effective surface flow visualization methods, specifically for visualizing gaseous flows on a body.

In this experiment, the surface flow behavior on six bodies are studied in supersonic flow. The streakline patterns are obtained using the dye flow visualization technique, and the flow paths, degree of flow attachment and separation, and location and type of shocks are identified and studied from the flow images.

Principles

In dye flow visualization, fluid particles are marked with a dye to obtain the path that is traced by the particles as flow is introduced. The dye is a semi-viscous mixture of fluorescent dye particles and oil. The fluorescent dye colors the fluid particles and illuminates them when they are excited by a UV light source, and the oil helps maintain the flow patterns on the surface, even after the body is no longer exposed to flow. The dye flow visualization technique provides a very simple, cheap, and quick way to analyze the flow patterns over any surface.

Depending on the method of imaging, dye flow visualization can be used to find the streaklines as a result of the fluid flow. If the image is taken with prolonged exposure, the dye can be used to track the path taken by a single fluid particle as it moves in the flow. In the technique used in the current experiment, all fluid particles passing through a point or area are marked with a dye, and the line joining all the dyed particles after the body has been placed in an active flow is the streakline. Here, a single frame captured at the end of the flow visualization experiment provides enough information to study the general surface flow on the body. Dye visualization via streaklines, in addition to providing details on flow movement along the surface, also helps identify surface flow features. Using dye visualization in supersonic flow can identify flow separation, shock formation, and movement of flow across the body's surface, all of which are features that help to optimize the body aerodynamically. 

Procedure

  1. Observing streaklines in supersonic flow
    1. Mix the fluorescent dye powder and mineral oil in a plastic bowl. Add small amounts of mineral oil to the dye in increments, mixing continuously until a semi-viscous mixture is obtained. The mixture should not be runny.
    2. Mount the sting above the supersonic wind tunnel test chamber and lock it into place. A blow-down supersonic wind tunnel with a 6 in x 4 in test section and an operating Mach number range of 1.5 to 4 was used in this demonstration, as shown in Figure 1. The Mach number is varied by adjusting the block setting (changing the area ratio of the test section).
    3. Screw the 2D wedge model onto the sting mount and fix the direction of the wedge such that the wedge surface is facing the transparent sidewalls of the wind tunnel test section. All models are shown in Figure 2.
    4. Use a paintbrush to apply a sufficient amount of the dye mixture onto the model. Ensure that the dye does not drip off of the model. See Figure 3 for reference.
    5. Adjust the block setting for the desired free-stream Mach number.
    6. Close and secure the wind tunnel panels.
    7. Run the wind tunnel for 6 seconds.
    8. After the run is complete, shine a UV light onto the model to illuminate the dye. Capture the streakline image with a camera.
    9. Adjust the angle of attack or Mach number according to the test matrix listed in Table 1 for model and repeat steps 1.4 – 1.9.
    10. Repeat steps 1.3 – 1.9 for all models listed in Table 1.
    11. When all models have been tested, shut down the wind tunnel, and dismantle the set-up.


Figure 1. Blow-down supersonic wind tunnel.


Figure 2. Wind tunnel models (left to right) 2D wedge, 3D wedge, cone, blunt nose body, sphere, and missile.

Table 1. Test Matrix.

Model Angle of Attack (q) or Mach Number (M) setting
2D 10° Wedge θ = 0, 12, and -12°
3D 10° Wedge θ = 0, 12, and -12°
Cone θ = 0, 13, and -13°
Blunt Nose Body θ = 0, 11, and -11°
Missile θ = 0 and 11°
Sphere M = 2, 2.5, and 3


Figure 3. Representative image of fluorescent dye painted on the 2D wedge.

Visualizing flow around an autobody is critical to understanding and quantifying flow structure as well as for theorizing fluid flow behavior. One type of flow visualization is called surface flow visualization which uses a dyed fluid to observe the path traced by fluid flow around an object.

Dye flow visualization involves coating the body of interest with a dye to observe flow patterns along the body surface. The dye is a semi-viscous mixture of fluorescent dye particles and oil. The highly viscous nature of the oil helps maintain the flow patterns on the body surface. While the fluorescent dye lets us visualize those patterns under a UV light.

If the image is taken with prolonged exposure, the dye can be used to track the path taken by a single fluid particle as it moves in the flow. As dye marked fluid particles pass through a point or area, we can observe the line joining all of the dyed particles. This is called the streakline.

In supersonic flow, these streaklines can be used to identify the point of flow separation, shock formation and movement of flow across the surface.

Now let's take a closer look at flow over the sphere. Attached flow appears as smooth streaklines and the direction of the streaklines tells us the direction of flow on the surface. Flow separation can be identified as the region where the dye clumps up and appears brighter. This is because dye beyond the point of flow separation is undisturbed.

In supersonic flow, we can also observe the formation of shock waves on the surface of the body like on the fins of a missile shown by a thin bright curve. We can also use this technique to identify deformities on a surface as evidenced by regions where the streaklines are disturbed.

In this lab, we will demonstrate the dye flow visualization technique using several different bodies exposed to supersonic flow.

For this experiment, we'll use a blow down supersonic wind tunnel with an operating Mach number range of 1. 5 to 4. This wind tunnel has a 6 in x 4 in test section. The Mach number is varied by adjusting the block section. In other words, by changing the area ratio of the test section. We will test and observe the streaklines around several different models: a 2D wedge, a 3D wedge, a cone, a blunt nose body, a sphere and a missile.

To begin the experiment, mix fluorescent dye powder and mineral oil in a plastic bowl. Add small amounts of mineral oil to the dye in increments mixing continuously until the mixture is semi-viscous and not thin and runny.

Now, mount the sting above the wind tunnel test chamber and lock it into place. Then, screw the 2D wedge model onto the sting mount. Fix the direction of the wedge so that the wedge surface is facing the transparent sidewalls of the test section.

Use a paint brush to apply a thick layer of dye to the surface of the model ensuring that there is not so much that it drips off. Then adjust the block setting to reach the desired free stream mach number. Adjust the angle of attack alpha to 0° using a digital level.

Now, close and secure the test section door and run the wind tunnel for 6 s. Shine a UV light on the model during the run to illuminate the dye. This allows us to observe the evolution of the streakline patterns.

Once the run is complete, capture an image of the final flow patterns. Next, adjust the angle of attack to 12°. Paint the model with dye as before and run the wind tunnel for 6 s. Illuminate the streaklines with the UV light and capture the image with a camera.

Repeat these steps for the 2D wedge model at -12°. Execute the test and capture streakline images for all of the models according to the test matrix shown here. When all of the tests have been completed on each model, shut down the wind tunnel and disassemble the setup.

Now let's take a look at the results starting with the streaklines over the 2D wedge. At 0°, the streakline pattern shows uniform flow throughout the body except in the region where there is a surface deformity in the center causing flow to separate. When the wedge is angled to 12°, the flow along the surface is deflected upwards while the flow is deflected downward at the -12° setting.

Looking at the 3D wedge, we can see that the flow pattern at the center of the model is similar to that observed for the 2D wedge at all angle settings. However, the flow pattern at the top and bottom edges show deflection and the tip vortex effect is observed along their length.

Streakline patterns for the cone show that for all angles of attack, the flow curves around the body. We can also observe that flow separation occurs at the end of the cone as indicated by the region where the dye clumps up.

For the blunt nose model, we observe attached flow throughout the body at an angle of attack of 0°.  At 11 and -11°, the flow curves around the body following the surface contour and separates along the line where the dye coalesces.

While flow patterns in the front of the missile model are similar to that of the blunt nose body, the streaklines on the fins show varied features. At 0°, the streaklines on the top and bottom fins show attached flow at the front of the fin with gradual separation occurring in a cross pattern. We also observe that flow detaches a lot earlier at the root of the fins as compared to the tips.

If we look at the coalesced dye at the leading edge of the central fin, we can see that the streakline patterns indicate a bow shock with the shape of the shock marked by the dye. At an 11° angle of attack, we observe fully attached flow on the bottom fin but separated flow close to the root of the top fin. Similar to the 0° case, the presence of the central fin causes a bow shock at the fin's leading edge.

Finally, for the sphere, we varied mach number as opposed to angle of attack as the flow patterns remain the same regardless of deflection angle. We can see that as the mach number increases, the point of separation moves toward the aft of the body showing decreasing flow separation. This is due to the fact that higher velocity flows have more momentum which helps the flow overcome the adverse pressure gradient over the sphere. This leads to a higher degree of flow attachment with increased mach number.

In summary, we learned how streaklines can be used to identify the point of flow separation, shock formation and movement of flow across a surface. We then exposed several bodies to supersonic flow in a wind tunnel and observed the streaklines that formed on each surface at varying angles of attack.

Results

The streakline flow patterns for the six models and conditions listed in Table 1 are shown below. For the 2D wedge, a uniform flow pattern is observed over the body, as shown in Figure 4, except in the region where there is a surface deformity, which causes the flow to separate. When angled at 12°, the flow along the surface is deflected upwards. This effect is mirrored when the model is angled at -12°. In general, all cases show attached flow across the entire surface, except at and behind the region of the surface deformity.


Figure 4. Streakline flow patterns over the 2D wedge (left to right) for Ɵ = 0°, 12°, and -12°.

Observations from Figure 5 show that while the flow patterns at the center of the 3D wedge are similar to that which was observed for the 2D wedge at all three angle settings, the flow patterns near the top and bottom edges show flow deflection. This could be attributed to the tip vortices at the edges of the wedge. While tip effects exist for the 2D wedge, the larger distance between the wedge center and the edge negates the effect of the tip on the central wedge flow. Additionally, the streaklines show no flow separation.


Figure 5. Streakline flow patterns over the 3D wedge (left to right) for Ɵ = 0°, 12°, and -12°.

Streakline flow patterns for the cone, shown in Figure 6, show streamlined, attached flow across the body for all angles of attack with the flow curving in the direction of the deflection. We also observe that flow separation occurs at the end of the cone, as indicated by the region where the dye clumps up.


Figure 6. Streakline flow patterns over the cone (left to right) for Ɵ = 0°, 13°, and -13°.

Figure 7 compares the flow patterns over a blunt edge at three angles of attack. When Ɵ = 0°, we see attached flow over the entire body. At Ɵ = 11 and -11°, the flow curves around the body (following the surface contour) but separates along the line where the dye coalesces.


Figure 7. Streakline flow patterns over the blunt nose body (left to right) for Ɵ = 0°, 11°, and -11°.

While the flow patterns at the front of the missile are similar to those observed on the blunt nose body, the streaklines on the missile fins (Figure 8) show interesting flow features. At Ɵ = 0°, the streaklines on the top and bottom fins show attached flow at the front of the fin with gradual separation occurring in a cross pattern, which originates from the fin tips and roots. We also observe that flow detaches much earlier at the root of the fins as compared to the tips. Another interesting observation is made by studying the coalesced dye at the leading edge of the central fin. The streakline patterns indicate a bow shock with the shape of the shock marked by the dye. When the missile is angled at 11°, we observe fully attached flow on the bottom fin but separated flow close to the root of the top fin. Similar to the 0° case, the presence of the central fin causes a bow shock at the fins leading-edge.


Figure 8. Streakline flow patterns over the missile (left to right) for Ɵ = 0° and 11°.

For the sphere, as the Mach number was varied, the flow patterns around the sphere remained the same, regardless of the deflection angle. Observations from Figure 9 show that as the Mach number increases, the region of separation (indicated by the area where the dye is not disturbed) decreases. This is because higher velocity flows have more momentum, which in turn allows the flow to overcome the adverse pressure gradient over the sphere. This causes a higher degree of flow attachment with increasing Mach number.


Figure 9. Streakline flow patterns over the sphere (left to right) M = 2, 2.5, and 3.

Applications and Summary

Streakline flow patterns over six bodies in supersonic flow were studied using surface dye flow visualization. Flow patterns over the 2D and 3D wedges showed that tip effects play a dominant role in determining surface flow structure. Flow over the cone was shown to be fully attached for a deflection range of ±13°. The blunt nose model was the first body to show a clear separation line when deflected at an angle of 11°, a pattern that was also observed in the initial section of the missile. The flow patterns on the missile fins indicate interesting features, such as flow separation and shock formation. We also deduced the type of shock (bow-shock) that formed at the leading edge of the fin. Finally, varying the Mach number for flow over a sphere showed that the point of flow separation moves aft on the sphere with increasing flow velocity. Overall, the experiment demonstrated the simplicity and effectiveness of streakline dye flow visualization, a technique used by aerospace engineers in rapid-design processes to obtain more streamlined and efficient aero-vehicles.

내레이션 대본

Visualizing flow around an autobody is critical to understanding and quantifying flow structure as well as for theorizing fluid flow behavior. One type of flow visualization is called surface flow visualization which uses a dyed fluid to observe the path traced by fluid flow around an object.

Dye flow visualization involves coating the body of interest with a dye to observe flow patterns along the body surface. The dye is a semi-viscous mixture of fluorescent dye particles and oil. The highly viscous nature of the oil helps maintain the flow patterns on the body surface. While the fluorescent dye lets us visualize those patterns under a UV light.

If the image is taken with prolonged exposure, the dye can be used to track the path taken by a single fluid particle as it moves in the flow. As dye marked fluid particles pass through a point or area, we can observe the line joining all of the dyed particles. This is called the streakline.

In supersonic flow, these streaklines can be used to identify the point of flow separation, shock formation and movement of flow across the surface.

Now let’s take a closer look at flow over the sphere. Attached flow appears as smooth streaklines and the direction of the streaklines tells us the direction of flow on the surface. Flow separation can be identified as the region where the dye clumps up and appears brighter. This is because dye beyond the point of flow separation is undisturbed.

In supersonic flow, we can also observe the formation of shock waves on the surface of the body like on the fins of a missile shown by a thin bright curve. We can also use this technique to identify deformities on a surface as evidenced by regions where the streaklines are disturbed.

In this lab, we will demonstrate the dye flow visualization technique using several different bodies exposed to supersonic flow.

For this experiment, we’ll use a blow down supersonic wind tunnel with an operating Mach number range of 1. 5 to 4. This wind tunnel has a 6 in x 4 in test section. The Mach number is varied by adjusting the block section. In other words, by changing the area ratio of the test section. We will test and observe the streaklines around several different models: a 2D wedge, a 3D wedge, a cone, a blunt nose body, a sphere and a missile.

To begin the experiment, mix fluorescent dye powder and mineral oil in a plastic bowl. Add small amounts of mineral oil to the dye in increments mixing continuously until the mixture is semi-viscous and not thin and runny.

Now, mount the sting above the wind tunnel test chamber and lock it into place. Then, screw the 2D wedge model onto the sting mount. Fix the direction of the wedge so that the wedge surface is facing the transparent sidewalls of the test section.

Use a paint brush to apply a thick layer of dye to the surface of the model ensuring that there is not so much that it drips off. Then adjust the block setting to reach the desired free stream mach number. Adjust the angle of attack alpha to 0° using a digital level.

Now, close and secure the test section door and run the wind tunnel for 6 s. Shine a UV light on the model during the run to illuminate the dye. This allows us to observe the evolution of the streakline patterns.

Once the run is complete, capture an image of the final flow patterns. Next, adjust the angle of attack to 12°. Paint the model with dye as before and run the wind tunnel for 6 s. Illuminate the streaklines with the UV light and capture the image with a camera.

Repeat these steps for the 2D wedge model at -12°. Execute the test and capture streakline images for all of the models according to the test matrix shown here. When all of the tests have been completed on each model, shut down the wind tunnel and disassemble the setup.

Now let’s take a look at the results starting with the streaklines over the 2D wedge. At 0°, the streakline pattern shows uniform flow throughout the body except in the region where there is a surface deformity in the center causing flow to separate. When the wedge is angled to 12°, the flow along the surface is deflected upwards while the flow is deflected downward at the -12° setting.

Looking at the 3D wedge, we can see that the flow pattern at the center of the model is similar to that observed for the 2D wedge at all angle settings. However, the flow pattern at the top and bottom edges show deflection and the tip vortex effect is observed along their length.

Streakline patterns for the cone show that for all angles of attack, the flow curves around the body. We can also observe that flow separation occurs at the end of the cone as indicated by the region where the dye clumps up.

For the blunt nose model, we observe attached flow throughout the body at an angle of attack of 0°.  At 11 and -11°, the flow curves around the body following the surface contour and separates along the line where the dye coalesces.

While flow patterns in the front of the missile model are similar to that of the blunt nose body, the streaklines on the fins show varied features. At 0°, the streaklines on the top and bottom fins show attached flow at the front of the fin with gradual separation occurring in a cross pattern. We also observe that flow detaches a lot earlier at the root of the fins as compared to the tips.

If we look at the coalesced dye at the leading edge of the central fin, we can see that the streakline patterns indicate a bow shock with the shape of the shock marked by the dye. At an 11° angle of attack, we observe fully attached flow on the bottom fin but separated flow close to the root of the top fin. Similar to the 0° case, the presence of the central fin causes a bow shock at the fin’s leading edge.

Finally, for the sphere, we varied mach number as opposed to angle of attack as the flow patterns remain the same regardless of deflection angle. We can see that as the mach number increases, the point of separation moves toward the aft of the body showing decreasing flow separation. This is due to the fact that higher velocity flows have more momentum which helps the flow overcome the adverse pressure gradient over the sphere. This leads to a higher degree of flow attachment with increased mach number.

In summary, we learned how streaklines can be used to identify the point of flow separation, shock formation and movement of flow across a surface. We then exposed several bodies to supersonic flow in a wind tunnel and observed the streaklines that formed on each surface at varying angles of attack.