When water flows over an acoustic sensor, non-acoustic pressure fluctuations caused by turbulence can decrease the signal-to-noise ratio and make it difficult to sense incoming acoustic waves. The Quiet Water Tunnel Facility at the Naval Undersea Warfare Center in Newport, RI is a unique test facility capable of investigating these pressure fluctuations and evaluating new and existing technologies aimed at reducing flow noise and drag due to skin friction. These technologies include modifications to the surface itself, such as riblets or compliant coatings, or modifications to the flow, such as suction or injection of water into the boundary layer.


Water tunnel components and flow schematic.
The Quiet Water Tunnel Facility was built at the Naval Underwater Sound Laboratory in New London, Connecticut in 1965. When constructed, the facility included only a circular test section for studying fully developed turbulent pipe flow. A rectangular test section was added in 1974 to allow for flat plate wall pressure measurements and the investigation of compliant coatings beneath turbulent boundary layers. The facility was subsequently moved to the Naval Undersea Warfare Center in Newport, Rhode Island in 1995.

The Quiet Water Tunnel depicted in the accompanying diagram is a recirculating flow facility that contains approximately 2000 gallons (7600 liters) of fresh water. Mass flow is controlled by a 745.7 watt induction motor that is coupled with a centrifugal pump. Maximum mass flow is approximately 3,300 gal/min (210 L/s). Static pressure is kept between 20-40 psi (140-210 kPa) during testing to prevent pump cavitation, and water temperature can be maintained from 60-90 °F (15.5-32 °C) with a counterflow heat exchanger.

As seen in the diagram, circular and rectangular test sections are installed in parallel and can be run independently or concurrently with one another. The circular test section consists of an acrylic pipe with an inner diameter of 3.5 inches (89mm) and a wall thickness of 0.5 inch (13mm). Flow enters the pipe from a transition section which is connected to the upper plenum chamber via a rubber hose. Centerline velocities up to 80 ft/s (24.5 m/s) are possible in the circular test section, resulting in Reynolds numbers based on pipe diameter up to 2.4×106. The generated boundary layer is half the pipe diameter, or approximately 1.75 inches (45mm) thick.

The rectangular test section is 83 inches (2.11 meters) long, with a constant interior width of 12 inches (305mm). In order to compensate for the growth of the boundary layers on the walls and to maintain a zero pressure gradient flow, the interior height increases from 4 inches (102mm) at the inlet to 4.41 inches (112mm) at the exit. If needed, the bottom plate of the test section can be reconfigured to establish an adverse or favorable pressure gradient. Free stream velocities up to 20 ft/s (6.2 m/s) are possible in the rectangular test section. A rectangular contraction nozzle upstream of the test section in the middle plenum chamber is used to accelerate the flow into the test section while minimizing free stream vorticity, resulting in a turbulence intensity of approximately 1% in the free stream. Also, the test section has minimal spanwise variation in the boundary layers on the top and bottom walls. The side wall boundary layers have minimal effect on measurements that are taken from the center of the channel on the top or bottom walls.

Custom instrumentation can be easily installed in each test section. Both test sections have a modular design with easily removable and replaceable fixtures. In the circular test section, sections of the acrylic pipe can be removed and replaced with instrumented sections. In the rectangular section, six ports in the top of the test section can be removed and machined in order to accommodate a variety of sensors and test fixtures, including piezoelectric wall pressure sensors, flush mounted hot film wall shear stress sensors, pitot tubes, and static pressure taps. For example, one current port has a pressure sensor array consisting of 48 tightly-spaced piezoelectric sensors flush mounted at the fluid/solid interface, allowing direct wave-number-frequency measurements of turbulent boundary layer wall pressure fluctuations to be made.

« Start Prev 1 2 Next End»