After settling on an array design, the next step in preparation for an aero-acoustic phased array test is to construct and install the array panel. There are a (perhaps surprising) number of details to be considered in this process.
Test Section Side View
C::::::::i~::::::::::::::::;::i::~~::::::;=====::J1 Ceiling
/ \
Small Array
Center " 3 dB Contours
" - Gear Source
~ '
. - Slat Source, ~
Flow
C::::===========::JI
FloorFig. 3.38. Side view of NASA Langley Low Turbulence Pressure Tunnel with phased ar-ray and airframe model installed. Note that the phase center of the small phased arar-ray (Figure 3.36) is not collinear with the landing gear and slat. Three dB down contours for noise sources at the landing gear and slat do not overlap making resolution of the two independent sources possible (compare with Figure 3.37)
Panel Strength Requirements
The mounting structure for the array sensors will typically replace an existing panel, or perhaps several existing panels, in the wind tunnel. An example of one such installation is shown in Figure 3.39. Strength requirements for re-placement panels may be found in wind tunnel design specifications and ar-ray panels must adhere to these requirements. When practical, it may be de-sirable to modify or remove tunnel infrastructure to remove impediments to array sensor installation. Any such modifications will likely increase the com-plexity of the array panel design, manufacture, and installation process.
Simultaneous Measurement Considerations
Aeroacoustic phased array measurements can be combined with other types of aerodynamic measurements such as flow visualization using mini tufts (Storms et al.1998) or Pressure measurements using pressure sensitive paint 12.
The reference to simultaneous measurements doesn't necessarily mean that various types of data are acquired at the same moment in time. They may be acquired back-to-back but are simultaneous in the sense that they are
ac-12 Boeing aero acoustic phased array and pressure sensitive paint teams have conducted si-multaneous phased array and pressure sensitive paint measurements at both the NASA Langley 14- by 22-Foot Subsonic Tunnel and the NASA Langley Low Turbulence Pressure Tunnel.
Fig. 3.39. Array panel installed in wall of the NASA Langley 14- by 22-Foot Subsonic Tun-nel (Gentry et al. 1990). PaTun-nel installation required removal of windows, slot covers and
"permanent" tunnel wall panels. The only remaining encumbrances were three 2-in. wide support beams running the width of the array panel. Photo courtesy of NASA Langley Research Center
quired during the same test condition. The point of discussion is whether the ability to make multiple types of measurements in the tunnel needs to be sup-ported at the same time. Such a requirement may place additional constraints on array panel material selection. For example, the array panel may need to be constructed from a transparent material to provide camera and/or flash access for a flow visualization measurement technique. Depending on the size of the access required, it might be sufficient to use a non-transparent material and design a small window into the panel for visual access. In many cases, the ar-ray panel will replace an existing panel that has one or more rows of static pressure ports running through it. These pressure ports need to be provided for in the replacement array panel. Of course, all of these considerations must have been addressed during the array design process so that there are no con-flicting requirements between array sensor positions and locations required for other simultaneous measurement needs.
Sensor Mounting
Sensor mounting issues are paramount to array panel design and construc-tion. With potentially hundreds of array sensors, the need to troubleshoot mal-functioning elements may be commonplace. The sensors should be easily ac-cessible. Sensors that work well for aero acoustic phased array measurements in low speed wind tunnels 13 are typically both expensive and of limited ro-bustness. For in-flow use, the screens or grid caps that protect the fragile sens-ing elements must be removed. To reduce the risk of damage, it is best if the sensors can be installed after the array panel is installed in the wind tunnel. If circumstances require sensor installation prior to panel installation in the tunnel, then provisions should be made for how the panel will be handled dur-ing installation to reduce the risk of sensor damage.
Sensors must be mounted as flush as possible to limit flow noise at the sen-sor. A recess should be avoided since the cavity created by such introduces the risk of discrete tones being created due to a vortex formation at the cavity edge. Protrusion should be avoided, because the closer the microphone di-aphragm is to the wall, the deeper it will be buried in the turbulent boundary layer resulting in lower velocity and less turbulence 14.
Care must also be taken to provide electrical isolation of the sensors from the mounting structure as well as from one another. This can be accomplished in one of two ways: mounting each sensor in a non -conducting material that is then inserted into the array panel; or selecting a material for the array panel that is non-conductive.
Calibration Requirements
Ideally, the array will be calibrated in-situ. In-situ calibrations reduce array panel handling and thus reduce the risk of sensor damage. However, there are circumstances where in-situ calibration is not feasible. For example, the prox-imity of the model to the array may preclude suitable placement of a calibra-tion source 15. In such a case, it will be necessary to remove the array panel from the tunnel and perform the array calibration in the immediate vicinity where a suitable calibration source placement with respect to the array can be achieved. The implication on array panel design is that the panel must be rel-atively easy to remove from and reinstall in the test section wall. Mounting of the array panel for use during calibration will also need to be considered as part of the panel design process.
13 This topic is covered in detail later in this chapter.
14 Personal correspondence with Dr. Gerald C. Lauchle.
15 The topic of array calibration is covered in detail later in this chapter.
Traverse Requirements
A fixed array provides an aperture at a single location that can image sources in 3-dimensional space. As such, the phased array is a powerful tool for map-ping and determining the strength of noise sources. It is also relatively com-mon in aero acoustic testing to want to know the directivity of each noise source so as to be able to determine how the sources contribute to the overall noise at various emission angles. This might be particularly important for ex-ample in trying to predict component impacts on overall noise from an air-plane flyover during an approach or takeoff condition. In tests where directiv-ity is important, it may be desirable to traverse the array. An alternative to tra-versing is to position an array at each directivity angle of interest, but this can be cost-prohibitive due to enormous sensor-count requirements.
In free-jet wind tunnels such as the Boeing Low Speed Aeroacoustic Facil-ity (Allen and Reed 1992) or in large closed circuit wind tunnels such as the NASA Ames Research Center 40- by 80-Foot Wind Tunnel (Schmitz et al.
1994), a traversing array may be set up on existing traverse mechanisms typi-cally used for single in-flow microphones. In these scenarios, the array may be positioned at an infinite number of locations allowing for detailed directivity studies for each noise source of interest. With an automated traverse system, the array may be positioned at as many locations as desired during a test con-dition!6 with array data collected at each location. The layout for an in-flow traversable array in the NASA Ames 40- by 80-Foot Wind Tunnel is shown in Figure 3.40. A traversable array installed in the Boeing Low Speed Aeroa-coustics Facility, a free-jet wind tunnel, is shown in Figure 3.41.
In closed-circuit wind tunnels where the walls are relatively close to the model such that an in-flow traverse would disturb the aerodynamic flow field and/or be in too close of proximity to the model, provisions need to be made to manually traverse the array. Where a large unencumbered region of a tun-nel wall is available, a manual traverse capability may be implemented by con-structing an array panel and a number of other fractional-width panels (e.g., half, quarter, and eighth widths). The panels may then be rearranged to estab-lish different array positions corresponding to directivity angles of interest.
Figure 3.42 illustrates the concept for a manually traversable array!7. This method of traversing is of course much less efficient than that of an automated traverse since the tunnel must be shut down each time the array is to be "tra-versed:' Still, it provides a reasonably straightforward way to acquire directiv-ity information without incurring extensive instrumentation cost.
16 A "test condition" here refers to a given model configuration and tunnel set point so that the only thing changing during the condition would be the position of the array. Every-thing else is assumed to be static.
17 Manually traversable arrays using fractional width panels have been deployed by Boeing at the NASA Ames 7- by 10-Foot Wind Tunnel and the NASA Langley Basic Aerodynam-ics Research Tunnel.
Fig. 3.40. TWo views of a traversable in-flow aero acoustic phased array in the NASA Ames 40- by SO-Foot Wind Tunnel. A larger fixed-position in-flow array is shown behind the tra-verse track. The fixed array may be used when the smaller traversable array is at either end of the traverse track. Rendition courtesy of NASA Ames Research Center
Fig. 3.41. Traversable array in the Boeing Low Speed Aeroacoustic Facility. The array panel sits on a cart that rides on a track outside the sheer layer of the free jet. Photo courtesy of Boeing Aerodynamics/Noise/Propulsion Laboratory
Interchangeable Panels
Quarter Panel Array Panel
~
Flow
Interchangeable Panels
Half Panel
Fig. 3.42. Concept for a manually traversable array in a closed-circuit wind tunnel. Array and fractional panels may be interchanged to "traverse" the array panel. For the pictured configuration, the array can be traversed to eleven different positions
Fig. 3.43. Two configurations for a repositionable array at the NASA Glenn Research Cen-ter Aero-Acoustic Propulsion Laboratory. The array is positioned below the jet at 90 degrees (top) and 120 degrees (bottom) relative to a selected location in the jet plume. A linear ar-ray can be seen to the left of the jet. Photo courtesy of Boeing Aerodynamics/Noise/Propul-sion Laboratory
Another consideration is that of a "repositionable array:' This is not really a traversing array in that it is not attached to a rail or track where the array slides along some axis. A repositionable array might just be mounted on a cart that can be wheeled around and/or tilted to various angles to provide multiple source directivity views. An example from a test at the NASA Glenn Research Center Aero-Acoustic Propulsion Laboratoryl8 (Bridges 1999) is shown in Figure 3.43. Figure 3.44 shows a traversable and repositionable array installa-tion in the Boeing Low Speed Aeroacoustic Facility that was used for an in-stalled jet noise test (Blackner and Bhat 1998).
Nested Arrays
Nested arrays do not necessarily impose any additional requirements on array construction. In the case of a single array plate, the nested array would just consist of some additional sensor locations. However, depending on the sen-sor mount requirements, the nested array may need to be a modular part that inserts into the larger array. An example of an array where this was necessary is shown in Figure 3.45. The microphone-preamplifier strain relief require-ments necessitated a double wall array panel construction. The front wall (tunnel wall replacement) held the 1/4-inch microphone and 1/4-inch micro-phone to 1/2-inch preamplifier adapter assembly while the back wall held the preamplifier. This configuration made access for installing the microphone-preamplifier assemblies difficult, especially for the tightly packed micro-phones at the center of the array panel. The solution, shown in Figure 3.46, was to build a detachable center section that could be instrumented on a desktop and then inserted into the larger panel. This solution had the added benefit in that it allowed for independent use of the small array for other small scale (high frequency) phased array testing. When a modular approach is used, it is imperative that the design and construction for assembly of the parts insures a known and repeatable relative positioning of those parts. The repeatability must insure that the sensor positions relative to one another are known within the tolerance specified in the array design process. Tolerance requirements are discussed in a following section.
Cable Strain Relief
A plan for cable strain relief should be developed as part of the array panel design process. This is particularly important when the array will be moved after instrumentation has been installed. With all the cables connected, in-adequate strain relief can cause sensor system failure and contribute to sub-stantial test delays for troubleshooting of instrumentation problems.
18 http://www.grc.nasa.gov/WWW/Aapl/
Fig. 3.44. A traversable and repositionable phased array installation in the Boeing Low Speed Aeroacoustic Facility. The cart that supports the array is mounted on traverse tracks enabling computer controlled traversing of the array during a test condition. The angle of the array can be repositioned manually from zero (horizontal) to 90 degrees. Photo cour-tesy of Boeing Aerodynamics/Noise/Propulsion Laboratory
When an array panel is installed in a fixed position and then instrumented, the strain relief does not need to be provided as an integral part of the array panel. Typically, some part of the tunnel infrastructure may be used for cable strain relief. An array at the NASA Ames Research Center 12-Foot Pressure Wind Tunnel shown in Figure 3.47 is an example of an installation where the existing tunnel infrastructure was used for cable strain relief. Figure 3.48 shows an installation at the NASA Langley Basic Aerodynamic Research Tun-nel where strain relief was designed into the array paTun-nel due to the require-ment to move the instrurequire-mented array for calibration purposes.
Fig. 3.45. A nested array installation at the NASA Ames Research Center 7- by IO-Foot Wind Tunnel. For this installation, the inner array was constructed as a physically separate piece to provide access for instrumentation. The small window to the left of the inner array was used for camera and flash access to support mini-tuft flow visualization. Photo cour-tesy of NASA Ames Research Center
Heating to Avoid Potential Condensation Problems
Condenser microphones are susceptible to failure due to moisture. In applica-tions where the microphones are exposed to significant temperature drops and then warming, if the microphone cartridge remains below ambient tem-perature, moisture may form on the cartridge. If this moisture collects be-tween the microphone diaphragm and backplate that form the capacitive por-tion of the sensing circuit, bias voltage leakage may occur, temporarily ren-dering the sensor useless. Situations where array sensors are particularly prone to condensation include pressure tunnels and tunnels where the test
Fig. 3.46. Desktop instrumentation for the nested array shown in Figure 3.45. Photo cour-tesy of Boeing Aerodynamics/Noise/Propulsion Laboratory
Fig. 3.47. Array installation at the NASA Ames 12-Foot Pressure Wind Tunnel. Cable strain relief is accomplished by securing cables to the existing tunnel infrastructure above and to the right of the array panel. Photo courtesy of NASA Ames Research Center
Fig. 3.48. Array installation at the NASA Langley Basic Aerodynamic Research Center. The array is mounted in a calibration fixture separate from the tunnel. Since the array had to be moved back and forth from the tunnel to the calibration fixture, strain relief for the cables is built into the array panel. Photo courtesy of Boeing Aerodynamics/Noise/Propulsion Laboratory
arena is exposed to outside air. In some pressure tunnels the test section can be vented very rapidly and the temperature can drop as much as 50 of (27.8 °C) in a matter of seconds. When the test section is then exposed to the much warmer ambient air (from outside the test section), a layer of moisture forms on everything inside the test section. Upon re-pressurization, if the moisture has not had enough time to evaporate, it may be forced into the microphone through the back vent causing permanent damage to the microphone. In non-pressurized facilities where the test arena is exposed to outside air, moisture problems tend to occur overnight. When temperatures drop overnight, the mi-crophones are cooled, the relative humidity rises, and internal condensation may create electrical noise. The microphones will be inoperable until warmed to the point where the water has evaporated (Wong and Embleton 1995). Even then, impurities (salts) left on the surface may render the microphone unus-able in high humidity 19.
To prevent condensation, it is necessary to keep the instrumentation slightly above ambient temperature. A simple and low-cost method is to use a string of
19 Personal communication with Donald W. Boston, March 2001.
lights in close proximity to the array sensors. For the NASA Ames 7- by 10-Foot Wind Tunnel, an array panel was built into one side of an enclosed box. The ra-diant heat from a string of lights was used to heat the instrumentation, keep-ing it slightly above ambient temperature. In other applications, heat lamps have been used without an enclosure. The heat lamps are typically only re-quired when the tunnel is not operating. Newer miniaturized instrumentation has allowed for production of smaller preamplifiers that in some cases provide enough self-heat to effectively combat condensation.
Geometric Survey Considerations
After the array panel is installed in the wind tunnel, it will be necessary to de-termine where the array is and how it is oriented in relation to other items of interest in the tunnel such as the model under test or a calibration source.
Geometric survey techniques that may be used in this process will be covered in a later section. The focus here will be to point out a couple of items that should be considered during the array design/build effort that may be useful in performing geometric surveys.
A mount for a laser at a known location in the panel relative to the known sensor locations and set so that the laser beam will project perpendicular to the plane of the array provides one useful means of assisting the experimen-talist in determining the array position relative to other items of interest. The hole location in the array panel (through which the laser beam will project), laser mount, and laser mount means of attachment should be included in the array panel design effort. For in-flow applications, provisions will need to be made for the laser hole to be covered so as not to disturb the airflow over the array panel and/or cause an unwanted noise source.
A mount for a laser at a known location in the panel relative to the known sensor locations and set so that the laser beam will project perpendicular to the plane of the array provides one useful means of assisting the experimen-talist in determining the array position relative to other items of interest. The hole location in the array panel (through which the laser beam will project), laser mount, and laser mount means of attachment should be included in the array panel design effort. For in-flow applications, provisions will need to be made for the laser hole to be covered so as not to disturb the airflow over the array panel and/or cause an unwanted noise source.