In order to perform tests in the PRC wind tunnel, the team must first understand the setup, the equipment, and the operation of the tunnel.
The wind tunnel we used for testing is a 5’ x 7’ atmospheric-intake, continuous flow subsonic tunnel located at the Pickle Research Center in Austin, Texas. The test section of the tunnel is approximately 50’ long and the full length of the tunnel is 220’ long. The wind tunnel is comprised of a metal frame and plywood. The test section and all test equipment are housed inside a building; the inlet and diffuser extend outside the building. The roof of the wind tunnel has a ± 3° divergence. The tunnel was designed to have a ±1.5° divergence, but an error in miscommunication caused the error (Westkaemper, 1982).
The wind tunnel is powered by four 200 hp fans that create a suction effect, forcing air to flow through the test section rather than air being blown into the test section. The fans have the capability to reach a maximum tunnel speed of 200 mph. The fans are adjustable to achieve a variety of different tunnel speeds and conditions. Vortices can be simulated by adjusting the power for one or two fans (Pena et. al., 2003).
A large perimeter fence surrounds the inlet to prevent birds, small animals, and debris from being sucked into the wind tunnel. An inlet screen covers the nozzle of the wind tunnel. The screen is used to help reduce turbulence of the air flow entering the inlet. It dissipates large turbulent eddies and instead creates several smaller eddies (Pakhurst, 1982). Figure 9 shows the setup of the wind tunnel and indicates the locations of the inlet screens and fans.
Figure 9: Plan View of the Pickle Research Center Low-Speed Wind Tunnel
The equipment section describes the equipment used to test the UAV. It is the equipment researched and present in the wind tunnel at Pickle Research Center.
The sting is used to mount the aircraft into the test section of the wind tunnel. The sting is also used to adjust the yaw, roll, and pitch of the model, as well as change the angle of attack of the model within the tunnel. The sting is controlled by voltage signals sent from LabVIEW.
The sting force balance is located at the tip of the sting. It holds strain gages that are used to measure the forces, such as lift, drag, and cross-wind, and the pitching, rolling, and yawing moments exerted on the aircraft with a very high degree of accuracy (Westkaemper, 1982). The balance used in the tunnel is a NASA #407 balance and is extremely sensitive. The balance has limits for each force that could not be exceeded.
Ø The limits for the balance are (Westkaemper, 1982):
Ø Normal force = ±200 lb
Ø Axial force = ±100 lb
Ø Side force = ±200 lb
Ø Pitching moment = ±1100 in-lb
Ø Rolling moment = ±200 in-lb
Ø Yawing moment = ±1100 in-lb
If any of the experimental tests exceeded the limits, the balance could be vulnerable to severe damage.
A pitot-static rake is a pressure-measuring device that is connected to a pressure transducer control box, which in turn is connected to the data acquisition board inside of the PC. Currently, a 7’ pitot-static rake is mounted in the test section of the wind tunnel to take readings of the stagnation and static pressure across the wind tunnel. Bernoulli’s equation is used to compute the exact flow velocity at each pitot-static probe. Bernoulli’s equation is defined as:
(1)
where
is the ambient air density and
is the local air speed. Since a barometer can measure the
ambient pressure and a thermometer measures the ambient temperature, T, the
ambient air density can be calculated from the ideal gas equation:
(2)
where R is the universal gas constant 8.3134 J/kgK. Bernoulli’s equation can be used to calculate the velocity profile from the stagnation measurement readings obtained.
The pitot-rake is fitted with tubes that attach to pressure transducers. Half of the tubes are attached to the pitot port on the rake and the other half to the static pressure port on the rake. The tubes are marked with numbers to determine which tubes lead to which ports. The pitot-rake also measures the velocity profile and checks for laminar flow (Pena, 2003). The rake can also traverse vertically in order to take measurements of pressure in the wake of the testing models or with no models present. Figure 10 shows the schematic of the pitot-static rake and the mounting system.
Figure 10: Schematic of pitot-static rake and mounting system (Pena, 2003)
The data obtained from the sting balance and pitot rake is collected on a Dell Optiplex GX 200 PC with a Windows 2000 operating system and a Pentium III processor. LabVIEW, a computer software program by National Instruments, has the ability to define input and output signals and then write them out as computer algorithms. The LabVIEW software helps to filter out unwanted signals and yields voltage readings that are proportional to pressure. LabVIEW copies data taken by the user and inserts it into a text file. This text file can be placed into a graph-generating spreadsheet to organize and plot the data. LabVIEW helps to collect the experimental data into a form readable by the user.
The data acquisition (DAQ) board records pressure readings. The DAQ board used at the wind tunnel is a National Instruments PCI-6034E DAQ board. The board has a 16-bit resolution and can take 200,000 samples per second. The board is installed in the PC and will send the readings to LabVIEW (Bonnecarrere et. al, 2003).
The PRC wind tunnel has three pressure transducers known as scanivalves. The scanivalves are used to measure static and total pressure inputs. Figure 11 is a picture of one of the scanivalves at the wind tunnel.
Figure 11: Picture of a Pressure Transducer (Bonnecarrere 2003)
A “porcupine” is an adapter that is used to connect the tubes from the pitot rake and wall static pressure taps to the back of the scanivalves. Figure 12 is a schematic of the Pickle wind tunnel. As the airflow enters the tunnel, the pitot-static rake takes measurements and sends the readings to the transducers. The transducers then send the readings to the data acquisition card where the information is processed and entered into the LabVIEW software installed on the computer. The transducers are powered by supply boxes.
Figure 12: Pickle Wind Tunnel (Pena, 2003)
A wind speed indicator is available to check the velocity in the test section of the wind tunnel. The indicator is connected to the pressure probe and measures the fluid velocity flow entering the test section. The pressure probe measures the stagnation and static pressures to determine the dynamic pressure. The equation for dynamic pressure is:
(3)
where
(4)
We can
solve for
, the actual velocity, since we know the values for
, 
and
. The flow density of air,, can be solved with the following equation:
(5)
Ten static “taps” were fixed inside the walls and ceiling to take static pressure readings throughout the test section. The taps were also used to check for buoyancy effects. After taking measurements, one tap was not working so we only had nine taps to rely on (Brown, 2004).
A wind tunnel is defined as having one converging section and one diverging section. The order in which they are positioned depends on what speeds you need to create. A supersonic tunnel has a diverging section followed by a converging section. This configuration produces speeds of Mach 1 and over. A subsonic tunnel has a converging section followed by the diverging section. This latter configuration is how the PRC wind tunnel is set up; therefore, it is unable to produce speeds over Mach 1.
We can assume the airflow in the tunnel is incompressible. We can assume this because compressibility effects are negligible when the air speed is Mach 0.3 or less. Our wind tunnel is only capable of reaching Mach 0.25. We can also assume that the air flow is viscous. We know this because velocity gradients are present and a boundary layer occurs on the surface of models. Velocity gradients are differences in velocity along different spatial directions. A boundary layer is the “region close to the surface of a solid body, where the effects of viscosity produce an appreciable loss of total head” (Pankhurst, 1952). Total head, H, is the pressure obtained when the relative velocity is zero. The equation for total head in incompressible flow is:
(6)
where the
variable q is the local velocity, p is static pressure, and
is the fluid density.
Subsonic tunnels tend to produce more turbulence than what occurs naturally in the atmosphere. Turbulent flow is characterized by unsteady eddying motions that are in constant motion with respect to each other. At any point in the flow, the eddies produce fluctuations in the flow velocity and pressure (Pankhurst, 1952). The turbulence is created by the fans, the roughness of the walls, vibrations running through the walls, and sound in general (Pankhurst, 1952). In our case, turbulence can be created by the weather. Since the tunnel is an atmospheric-intake tunnel, the conditions of the weather the days of testing can cause the data to be skewed, especially if it is gusty or rainy.
We can reduce these effects by installing wire gauze, or we can calculate the turbulence factor and take it into account. Since the cost of installing the gauze will prevent us from using it, we will have to take the turbulence factor into account. The turbulence factor is defined mathematically as
where
is the Reynolds number of the wind tunnel and
is the atmospheric free air Reynolds number (Pankhurst,
1952). The turbulence factor can be measured by conducting a pressure and
velocity profile of the empty tunnel. The purpose for the turbulence factor is
so that the Reynold’s number in the tunnel can be agree with the Reynold’s
number of free air.
