Constant Temperature Anemometry: A Tool to Study Turbulent Boundary Layer Flow
Source: Xiaofeng Liu, Jose Roberto Moreto, and Jaime Dorado, Department of Aerospace Engineering, San Diego State University, San Diego, California
A…
1. Hot-wire system dynamic response determination
The purpose of this procedure is to understand how fast the anemometer system can respond to flow signal changes. This capability is gauged by measuring the frequency response when the signal turns on and off by applying a square wave.
- Secure the hot wire probe of the CTA system inside of a wind tunnel using a support shaft.
- Set-up a DC power supply, signal generator and oscilloscope and connect them as shown in Figure 2(a). The signal generator supplies a square wave input to the Wheatstone bridge, and the output waveform is visualized on the oscilloscope.
- Turn on the hot-wire power supply, the oscilloscope and the signal generator.
- Set-up the signal generator to output a square wave with 150 mV amplitude and 10 kHz frequency.
- Observe the output signal in the oscilloscope to make sure that the output wave form frequency and amplitude are correct.
- Close the test section and plug in the serial port. Turn on the wind tunnel and set the airspeed to 40 mph.
- Once the airflow stabilizes, measure the width of the signal overshoot, τ, from the oscilloscope. See Figure 3 for the definition of τ.
- Use the measured value of τ to obtain the cut-off frequency of the hot-wire system using the equation: fcut = 1/1.5τ.
- Turn off the wind tunnel.
2. Hot-wire calibration
The purpose of this procedure is to establish the correlation between the airspeed and the electric potential of the Wheatstone bridge. This allows the flow velocity to be measured.
- Adjust the hot-wire probe to the vertical position so that it is a far enough distance away from the flat plate, which in this case is the floor of the wind tunnel, so that it is located in the free stream region.
- Start the wind tunnel control software.
- Open the virtual instrument software, and set the sampling frequency to 10 kHz and the number of samples to 100,000. These parameters are determined by the flow characteristics of the flow field to be measured and may vary depending on the knowledge of convergence requirement of the targeted statistics.
- Set the wind tunnel speed to 0 mph, and record the voltage on the Wheatstone bridge.
- Increase the wind tunnel air speed by increments of 3 mph up to 15 mph. Allow the flow to stabilize at each air speed before measuring the voltage.
- Increase the wind tunnel air speed by increments of 5 mph up to 60 mph, and measure the voltage at each increment.
- When all measurements are complete, reduce the air flow to 30 mph, and then turn off the wind tunnel.
Figure 3. Schematic for width of signal overshoot, τ as observed on an oscilloscope during a square wave test.
3. Boundary layer survey
- Using the same set-up as in the previous experimental section, slowly lower the hot-wire probe until it touches the test section floor, which will act as a flat plate.
- Turn on the wind tunnel and set the airspeed to 40 mph, the sampling frequency to 10 kHz and the number of samples to 100,000 like previously.
- Record the voltage reading at the lowest vertical setting, which is next to the flat plate and in the boundary layer.
- Move the probe vertically, increasing the height in steps of 0.05 mm up to the height of 0.50 mm, and record the voltage reading at vertical position.
- Increase the probe height in steps of 0.10 mm up to the height of 1.50 mm, and record the voltage reading at vertical position.
- Increase the probe height in steps of 0.25 mm up to the final height of 4.00 mm, and record the voltage reading at vertical position.
- When all measurements are complete, reduce the air speed to 20 mph, and then turn off the wind tunnel, CTA, power supply, oscilloscope, and function generator.
A boundary layer is a thin flow region immediately adjacent to the surface of a solid body in a flow field. The region of flow outside of the boundary layer, called the free stream region has a constant velocity. However, within the boundary layer there is a velocity gradient due to friction at the surface. The boundary layer typically undergoes several stages.
First the laminar boundary state, followed by the transition state and finally, the turbulent boundary layer state, which involves irregular flow and fluctuations, like mixing or eddying. The boundary layer is the basis for the calculation of skin friction drag on aircraft.
Skin friction drag is created within the boundary layer and is due to the viscous shear stress exerted on the surface. Skin friction drag is proportional to fluid dynamic viscosity, mu, and the local stream wise velocity shear strain rate, which is the gradient of the streamwise velocity in the normal direction. So it becomes significant for large areas, such as an airplane wing. Additionally, skin friction drag is higher in turbulent flow, since the fluid particles interact with the surface at high momentum.
One way to measure turbulent boundary layer properties is using hot wire anemometry, which is based on two principles related to the cooling effect of flow on a heated wire. According to the first principle, when a fluid flows over a hot surface, the convective heat coefficient changes, which results in changes in the surface temperature.
The second principle is Joule's law, which states that an electrical conductors heat dissipation, Q, is proportional to the square of the electric current, I, applied to the conductor. We can use the two principles to determine the velocity of fluid flow surrounding a heated metallic wire probe, by measuring the electrical potential E, that has to be applied to maintain a constant temperature of the wire.
A commonly used hot wire technique is Constant Temperature Anemometry or CTA. CTA consists of a very thin metallic wire, called the probe, which is connected to the arm of a Wheatstone bridge. The Wheatstone bridge controls the electrical potential and adjusts it as needed in order to maintain a constant temperature across the wire. Any cooling is caused by fluid flow around the wire. Thus, the change in the potential is a function of the heat transfer coefficient and by extension is a function of velocity.
In this experiment, we will demonstrate the use of a Constant Temperature Anemometry setup to measure the turbulent boundary layer over a flat plate.
First, we will learn how the Constant Temperature Anemometer, or CTA, system responds to flow signal changes using a wind tunnel. To begin, secure the hot wire probe of the CTA system inside of the wind tunnel using a support shaft.
Then, set up a DC power supply, signal generator, and oscilloscope. The components are connected as shown. To begin, turn on the hot wire power supply, the signal generator and the oscilloscope. Set the signal generator to supply a square wave input to the Wheatstone bridge with a 150 mV amplitude and a 10 kHz frequency.
Observe the output signal in the oscilloscope to make sure that the frequency and amplitude are correct. Now close the test section, plug in the serial cable, turn on the wind tunnel and set the wind speed to 40 mph. Once the airflow stabilizes, measure the width of the signal overshoot, tau, observed on the oscilloscope. Use the measured value of tau to calculate the cut-off frequency for the hot wire system using this equation. Then turn off the wind tunnel airflow.
Next we will establish the correlation between wind speed and the electrical potential of the Wheatstone bridge. To begin, raise the CTA probe in the vertical direction so that it is in the free stream region. Start the wind tunnel control software and then start the virtual instrument software. Set the sampling rate to 10 kHz and the number of samples to 100,000.
Now, with the wind tunnel airspeed set to 0 mph, record the voltage on the Wheatstone bridge. Then, increase the wind tunnel airspeed at increments of 3 mph up to 15 mph, measuring the voltage at each increment. Be sure to allow the air flow to stabilize before recording the voltage measurement.
Next, increase the wind tunnel air speed up to 60 mph in 5-mph increments, measuring the voltage at each increment. When all measurements are complete, reduce the airspeed to 30 mph and then turn off the wind tunnel airflow.
Using the same setup as before, lower the CTA probe slowly until it touches the test section floor, which will act as the flat plate. Set the airflow to 40 mph. Keep the sampling frequency at 10 kilohertz and the number of samples at 100,000. Record the voltage at the lowest vertical setting, which is next to the flat plate and in the boundary layer.
Now, move the probe vertically in steps of 0.05 mm up to a height of 0. 5 mm, recording the voltage reading at each position. Then, increase the probe height in increments of 0.1 mm up to a height of 1. 5 mm. And then in steps of 0.25 mm up to a final height of 4 mm, while recording the voltage at each increment.
When all of the measurements have been made, reduce the wind speed to 20 mph and then turn off the airflow. Then shut off the power supply, signal generator, and oscilloscope.
The first step in analyzing the data is to use the data acquired during the calibration step of the experiment, to determine the correlation between the hot wire voltage and air speed. There are several different methods to do this, which involve fitting the data to known heat transfer relationships, and it's covered in detail in the appendix of this video.
Once the mathematical relationship has been determined, use the voltage measurements to calculate velocity at each vertical height. After adjusting the nominal height to account for any artifacts from an overbent probe, plot the velocity profile u(y), which can then be used to determine the boundary layer displacement thickness.
This value represents the distance that the plate would have to be moved vertically in order to obtain the same flow rate that occurs between the surface and the fluid. We can also calculate the momentum thickness, defined as shown, which is the distance the plate would have to be moved vertically in order to have the same momentum that exists between the fluid and plate.
From these two parameters, we can calculate the shape factor, H. The shape factor is used to determine the nature of the flow, where a shape factor of about 1.3 indicates fully turbulent flow and about 2.6 for laminar flow. Between these values is transitional flow. In the case of this experiment, the shape factor was calculated as 1.9, indicating transitional flow.
In summary, we learned about boundary layer flow development, and then used a Constant Temperature Anemometry setup to analyze the turbulent boundary layer over a flat plate and observe low behavior.
View the full transcript and gain access to JoVE Science Education videos
Q1: What is a boundary layer and why does it matter in aeronautical engineering?
A boundary layer is a thin flow region immediately adjacent to a solid surface where viscous effects dominate and friction retards the flow. Outside this region, the free stream has constant velocity. The boundary layer is critical for calculating skin friction drag on aircraft, which results from viscous shear stress on the surface and becomes significant over large areas like airplane wings.
Q2: How does turbulent flow differ from laminar flow in a boundary layer?
Laminar boundary layer flow is smooth and orderly, while turbulent flow involves irregular motion with mixing and eddying. Turbulent flow creates higher skin friction drag because fluid particles interact with the surface at high momentum, enhancing momentum transfer. The boundary layer naturally progresses from laminar to transitional to turbulent states as it develops.
Q3: What principles underlie constant temperature anemometry for measuring flow?
Constant temperature anemometry relies on two principles: first, convective heat coefficient changes when fluid flows over a heated surface, altering surface temperature; second, Joule's law states electrical heat dissipation is proportional to the square of applied current. By measuring the electrical potential needed to maintain constant wire temperature, velocity can be determined since cooling is caused by fluid flow.
Q4: How does a Wheatstone bridge control the hot wire probe in CTA?
The Wheatstone bridge controls and adjusts electrical potential to maintain constant temperature across the thin metallic wire probe. Any cooling caused by fluid flow requires increased electrical potential to restore the wire to its set temperature. This change in potential directly correlates to the heat transfer coefficient and, by extension, to fluid velocity.
Q5: What do boundary layer displacement thickness and momentum thickness represent?
Displacement thickness represents the distance a plate must move vertically to obtain the same flow rate occurring between the surface and fluid. Momentum thickness is the distance the plate must move to have the same momentum existing between fluid and plate. These parameters help characterize boundary layer development and are used to calculate the shape factor, which indicates whether flow is laminar, transitional, or turbulent.
Q6: How is the shape factor used to classify boundary layer flow conditions?
The shape factor, calculated from displacement and momentum thickness, classifies flow regimes. A shape factor around 1.3 indicates fully turbulent flow, approximately 2.6 indicates laminar flow, and values between these indicate transitional flow. This parameter helps engineers understand the nature of boundary layer development and predict skin friction characteristics.
Q7: Why is skin friction drag more significant in turbulent boundary layers?
Skin friction drag is proportional to fluid viscosity and the local velocity gradient at the surface. In turbulent flow, macro-turbulent motion enhances momentum transfer by bringing high-momentum fluid particles down to the surface, increasing shear stress. This effect becomes particularly important over large surfaces like airplane wings, making turbulent boundary layer analysis essential for aircraft design and related studies like airfoil behavior pressure distribution.