Document
NASA CR-132604
RESEARCH
REPORT
ACOUSTICAL CHARACTERISTICS OF THE NASA - LANGLEY FULL-SCALE WIND TUNNEL TEST SECTION By A. L. Abrahamson P. K. Kasper R. S. Pappa Prepared for: ADMINISTRATION NATIONAL AERONAUTICS AND SPACE LANGLEY RESEARCH CENTER HAMPTON, VIRGINIA 23665 CONTRACT NUMBER NAS1-12841 - WYLE LB ARATORIES REPOBIiMRWRH-74____ ((N ASA-CR-132604) ACOUSTICAL CHARACTERISTICS N75-18258 OF THE NASA LANGLEY FULL SCALE WIND TUNNEL TEST SECTION (Wyle Labs., Inc., Hampton, Unclas Va.) 44 p HC $3.75 CSCL 14B \ .... .. ~M iircif 975 . . . . . . .
12436 G3/09
WyE LAE
SCIENTIFIC SERVICES & SYSTEMS GROUP 3200 MAGRUDER BLVD., HAMPTON, VIRGINIA ABSTRACT The full-scale wind tunnel at NASA-Langley Research Center was designed for low-speed aerodynamic testing of aircraft.
Sound absorbing treatment has been added to the ceiling and walls of the tunnel test section to create a more anechoic condition for taking acoustical measurements during aero- dynamic tests.
This report provides the results of an experimental investigation of the present acoustical characteristics of the tunnel test section.
The experimental program included measurements of ambient noise levels existing during various tunnel operating conditions, investi- gation of the sound field produced by an omnidirectional source, and determination of sound field decay rates for impulsive noise excitation.
A comparison of the current results with previous measurements has shown that the added sound treatment has improved the acoustical con- dition of the tunnel test section. An analysis of the data indicated, however, that sound reflections from the tunnel ground-board platform could create difficulties in the interpretation of actual test results.
Although not available for this test series a sound-absorbing ground-board platform has since been fabricated and is expected to minimize this problem.
iii TABLE OF CONTENTS Page 1.0 INTRODUCTION . . . . . . . . . . . .. . . . . . . . . 1 2.0 OUTLINE OF THE CURRENT STUDY . . . . . . . . . . . 2 3.0 TEST PROCEDURE AND ANALYSIS OF RESULTS . . . . . . 2 Ambient Noise Level . . . . . . . . . . . . . . . . . . . 2 3.1 3.2.1 Background . . . . . . . . . . . . . . . . . . . . . . 4 Sound Absorbing 3.2. 2 Measurements Before Addition of M aterial . . . . . . . . . . . . . . . . . . . . . . . 6 3. 2. 3 Measurements After Addition of Sound Absorbing . . . . . .. . . . . 6 M aterial . . . . . . . . . . .
3. 3 Measurement of Reverberation Time in Test . . . . . 10 4.0 CONCLUSIONS . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . 12 REFERENCES .
iv LIST OF TABLES Page STables Radius" Distances Measured from S "Modified Hall . . 13 vs. Distance Data . . . . . . . . .. . . ....
SPL . . . . . . . . . . . . 14 II. "Initial" Reverberation Times . . .
in "Sabine Acoustics" Analysis . . . . . . . . . 15 III. Parameters LIST OF FIGURES Page Figures Wind Tunnel Test 1 Plan and Evaluation Views of Full-Scale Positions for Ambient Noise Section Showing Microphone . . . . . . . . . . . .. 16 M easurem ents . . . . . . . . . .
Test Section . . . . . . . . . . . . 17 2. Ambient Noise Level in Function of Air Octave-Band SPL in Test Section as a 3a.
. . . ... . . . 18 Position No. 2 . . . . . . . . . .
Speed: Function of Air Octave-Band SPL in Test Section as a 3b.
. . . 19 . . . . . . . . . . . . . . .
Speed: Position No. 3 in Test Section as a Function of Air 3c. Octave-Band SPL . . . . . . . . . . . . 20 Position No. 4 . . . . . .
Speed: Level . . . . . . . 21 Normalized Octave-Band Ambient Noise 4.
Comparison of Mean Normalized Octave-Band Ambient 5.
Measurements Made Prior to and After Sound Noise of Test Section . . . . . . . . . . . .
Absorbent Treatment Narrow-Band Spectra of Ambient Level with Tunnel 6.
Rotational Fan Noise Running, Showing Predominant . . . . . . . . 23 Peaks . . . .. . . . . . . . . . . . . .
Point SPL vs. Distance from an Omnidirectional 7.
Total Room Absorption, Source as a Function of Field . . . . . ........ . 24 Assuming a Diffuse Reverberant V LIST OF FIGURES (continued) Page Figures 8. Source and Microphone Positions for Sound Field . 25 Measurements . . . . . . . . . . . . . . . . . . . . .
9a. Measured 63 Hz Octave-Band Sound Pressure Levels in Directions from Acoustic Source Located at Indicated Test Platform . . . . . . . . . . . . . . . . . 26 Center of Pressure Levels in 9b. Measured 125 Hz Octave-Band Sound from Acoustic Source Located at Indicated Directions Platform . . . . . . . . . . . .
Center of Test 9c. Measured 250 Hz Octave-Band Sound Pressure Levels in Indicated Directions from Acoustic Source Located at Center of Test Platform . . . . . . . . . . .
. . . . . . . 28 9d. Measured 500 Hz Octave-Band Sound Pressure Levels in Indicated Directions from Acoustic Source Located at Center of Test Platform . . . . . . . . . . . . . . . . . . 29 9e. Measured 1000 Hz Octave-Band Sound Pressure Levels in Indicated Directions from Acoustic Source Located at Center of Test Platform, with Comparison to Previous Measurements . . . . . . . . . . . . . . . . . . . . . . 30 9f. Measured 2000 Hz Octave-Band Sound Pressure Levels in Indicated Directions from Acoustic Source Located at Center of Test Platform . . . .
. . . . . . . . . . . . . . 31 9g. Measured 4000 Hz Octave-Band Sound Pressure Levels in Indicated Directions from Acoustic Source Located at 1 Center of Test Platform . . . . . . . . . . . . . . . .
. . 32 9h. Measured 8000 Hz Octave-Band Sound Pressure Levels in Indicated Directions from Acoustic Source Located at Center of Test Platform . . . . . . . . . . . . . . . . . . 33 10.
Comparison of Measured Octave-Band Sound Levels with Theoretical Values Considering the Effect of Platform Reflected Sound, as a Function of Distance from Acoustic Source. . . . . . . . . . . .... . .. ...... ..... .34 vi (continued) LIST OF FIGURES Page Figures 11. Comparison of Average Octave-Band "Modified Hall Radius" Distances with "Hall Radius" Values Measured . . 35 Prior to Sound-Absorbent Treatment of Test Section . . .
12. Typical Octave-Band Sound Decay Curves, Showing Measured Reverberation Times . . . . . . . . . . . . . . 36 13. Comparison of Average Octave-Band Reverberation Times with Measurements Made Prior to Sound-Absorbent Treatment of Reflected Component Magnitudes on 14 Effect to Sources Measured Sound Pressure Levels Due with Differing Directional Characteristics . . . . . . . . . . 38 vii ACOUSTICAL CHARACTERISTICS OF THE NASA-LANGLEY FULL-SCALE WIND TUNNEL TEST SECTION 1.0 INTRODUCTION With the emergence of acoustic noise profiles during takeoff and landing as important aircraft design criteria, wind tunnels, originally intended for aero- dynamic testing, have been required to perform a new category of functions in aeroacoustic research and development (Reference 1). Since most existing wind tunnel facilities were designed for the purpose of studying aerodynamic properties and not for measurements of aeroacoustic fields, they usually require modification before meaningful acoustic tests may be performed.
The investigation described in this report was the third of a series aimed at defining and improving the acoustic environment of the NASA-Langley full- scale wind tunnel, prior to conducting aeroacoustic tests.
The first study (Reference 2) consisted of an experimental investigation of the acoustic characteristics of the full-scale wind tunnel. The results provided evidence of the acoustic characteristics of the test section, and provided a quantitative assessment of the distance from an omnidirectional noise source within which acoustic measurements could be taken. It was indicated that this range could be increased through judicious placement of sound-absorbing material.
The second study (Reference 3) led from this conclusion to investigate, by means of an acoustic model, the most cost-effective placement of sound- absorbing material. Based on recommendations contained in this study, sound- absorbing material was installed on the roof of the test section, and on the ,east aid west walls o the.,test:section above test platform height.
2.0 OUTLINE OF THE CURRENT STUDY The current study followed naturally on the earlier work and arose out of the need to: a. Assess the effect of the sound-absorbent material on the reflected acoustic field in the test section, guidelines for future acoustic measurements in the test b. Formulate section.
obtained in this study provided evidence of the non- The experimental results comparison in the test section. In addition, diffuse acoustic environment source of local reflection model indicated a predominant with a theoretical platform The acoustic treatment for the apparently from the test platform.
fabrication and Reference 3, was still in the process of recommended in of this, not been installed for the current tests. In view consequently had at a later date.
it is suggested that the tests be repeated OF RESULTS 3.0 TEST PROCEDURE AND ANALYSIS with the above objectives, and consisted of The test procedure was designed types of acoustic measurements: three noise in various modes of tunnel operation, a. Ambient source, b. Sound field of an omnidirectional sounds.
c. Decay rate for impulsive to that followed in the first study (Reference 2) to This procedure was similar results obtained prior to installation of the sound- facilitate comparison with the absorbent material (Reference 3).
3. 1 Ambient Noise Level wind tun- top and side views of the test section of the full-scale Sketches of the in the in Figure 1. Octave-band ambient noise levels with no airflow nel appear four microphone positions indicated in the tunnel were measured at each of the the noise source was identified as a compressor in figure. An intermittant ambient levels at microphone position 4 were south end of the building. The For subsequent measured both with and without this compressor operating.
2 displays octave-band frequency tests the compressor was switched off. Figure measurements. For all microphone analyses of these four sets of ambient noise approxi- the octave-band levels with the compressor turned off fell positions, a 5-dB spread.
mately within the ambient level increased in all octave bands with With the tunnel running, generated by measurements at increasing airspeed. The family of curves at posi- is shown in Figure 3a, 3b, and 3c for microphones five tunnel speeds reveals a sound 3, and 4 respectively. Examination of the curves tions 2, 18 dB for a in each octave band of approximately pressure level increase doubling of airspeed.
the sixth procedure proposed in Reference 2 to demonstrate Following the clearly, the ambient octave-band levels power dependence on velocity more log u (where u is the airspeed in mph) were normalized by subtracting 60 1 0 levels for each from each value. The ranges of normalized octave-band presented in the data at 6 airspeeds
microphone position, icalculated from
Figure 4. The data spread of normal- Figures 3a through 3c, are shown in 3 dB for all nine each position is small and virtually constant at ized levels for are so small, it is reasonable to estimate the octave bands. Since the spreads than those chosen for the testlby adding octave-band levels at other airspeeds 4.
60 log u to the normalized levels of Figure levels in the test section before and after To compare the normalized ambient material, the mean of these normalized installation of the sound absorbent 4 was calculated for each octave noise levels measured at positions 2, 3, and compared in Figure 5 with similarly obtained band. These mean levels are in Reference 2 from measurements at three com- normalized levels reported positions.
parable microphone of the sound-absorbent material has resulted in a It may seem that addition This is noise level in the test section over all airspeeds.
lower ambient of approximately 2 dB at 31. 5 Hz shown as a decrease in the normalized level to a 7-dB drop at 8 kHz.
rising was conducted for the recorded acoustic A narrow-band frequency analysis speeds. Examples presented in Figure 6 show that data at several tunnel 100 Hz. From peaks are present in the spectrum below about large sharp is proportional to tunnel speed, their shift along the frequency scale, which rotational-noise components.
these are evidently propellor of any future acoustic Care should therefore be applied when evaluating results frequencies below 100 Hz. Similar narrow-band tests in the full-scale tunnel at due to in specific cases would assist in separating spectral peaks analyses aircraft noise.
tunnel propellors from those due to test 3.2 The Sound Field of an Omnidirectional Noise Source Located Above the Center of Test Platform 3.2. 1 Background a sound source within a room is In general, the acoustic field established by direct sound from the source and the multiply-reflected or composed of the of the direct sound decreases with increasing reverberant sound. The intensity distance from the source while, in general, the spatial distribution of the reverberant sound is a function not only of the geometry and sound-absorbing of all interior room surfaces, but also of the directional and spectral properties it is often characteristics of the source. For simplified analytical purposes, assumed that the reverberant sound level is constant throughout the room. The validity of this assumption is approached by "well-behaved" semireverberant uniform spatial distribution of sound-absorbing surfaces.
rooms having nearly level of the direct sound from a nondistr:ibuted source can The sound pressure be expressed as (Reference 4): SPL = PWL + 10 logl0 - 0.5dB (Eq. 1) where SPL = direct sound pressure level (dB re 20 x 10-6 N/mn ) PWL = acoustic source strength (dB re 10 2 watt) Q(e, ) = directivity factor of source (dimensionless) 0, i = azimuth and elevation of measurement position r = distance to acoustic center of source (ft) Thus, the direct sound pressure level will decrease by 6 dB for each doubling of the measurement distance from the source in any radial direction, indepen- dent of the source strength and its directivity.
If the acoustic energy of the reverberant field is uniformly distributed through- out the entire room, the field is said to be diffuse. Under such ideal conditions, a function of the acoustic power the reverberant sound level in a room is only sound source and of the total room absorption, and is given by output of the (Reference 4): (Eq. 2) a + 16.5 SPL = PWL - 10 log 1 0 r SPL = reverberant sound pressure level where 2 ) r (dB re 20 x 10-6 N/m PWL = acoustic source strength (dB re 10-12 watt) (sabins) a = total room absorption decibel sum The total sound pressure level-at a distance r from a source is the and the reverberant sound level (Equation of the direct sound level (Equation 1) 2). Its value is:
SPL = PWL + 10 logl ( Q +
+10.5 (Eq. 3)'
0 2 - N/m2 20 x l0 in dB re SPL for -12 watt PWL, dB re 10-12 r, ft a, sabins For an omnidirectional source, Q = 1 for all 0 and i. The difference SPL - as a function of the total PWL from Equation 3, for Q = 1, is plotted in Figure 7 room absorption, a.
will At some radial distance from the source, say r*, the direct sound level This distance, which can be considered as equal the reverberant sound level.
transition point between the direct and the reverberant sound fields, has the been termed the "hall radius", by several recent investigators (e. g.( Ref- in parentheses of erence 2). Its value, obtained by equating the two terms by: Equation 3, is given = 0. 141 a 1/, (Eq. 4) r* where a = total room absorption graphically from Figure 7. Note The hall radius, r*, may also be determined that each curve asymptotically approaches a specific reverberant sound level for a specified amount of total absorption. If the total absorption in the room is known, the intersection of the corresponding reverberant level horizontal asymptote and the -6 dB/doubling of distance line of the direct sound field will - locate the value for r* . Because the direct and reverberant sound fields are equal in intensity at this intersection point, the total sound pressure level at the hall radius distance will be 3 dB greater than that expected at the same radial distance in the absence of reflections.
3. 2. 2 Measurements Before Addition of Sound Absorbing Material The measurement and analysis in this section were reported in Reference 2, and represent a simplified description for the sound field in the test section full-scale wind tunnel. The sound field of a broadband of the NASA-Langley "omnidirectional" source suspended above the center of the test platform was measured in the vertical, and in horizontal directions perpendicular to the walls of the test section. Octave-band sound pressure levels for each direction were plotted as a function of distance from the source. The magni- tude of the sound field was found practically independent of direction for all octave bands. A mean "hall radius" was then calculated for each octave 11).
band from best-fit lines through data points (Figure 3. 2. 3 Measurements After Addition of Sound Absorbing Material The acoustics of large absorbent rooms are not generally representable by the simple assumptions made in deriving Equation 3 of section 3. 2. 1. The clas- sical "Sabine Assumption" of a diffuse reverberant field is particularly inapplicable when the room absorption is large and concentrated in localized areas. In this case, the reverberant sound field, rather than being diffuse, is dominated by first and second reflections. These result in phase rein- forcement and cancellation and give rise to the presence of spatial maxima and minima in the established sound field.
In contrast to the earlier experiment, the current set of measurements of the sound field about an omnidirectional source disclosed strong directional non- noise spectrum as input Measurements were made using a pink uniformities.
measurement relative to the test platform and to the source which was located shown in Figure 8.
positions sound field with distance, irregulari- To examine the nonuniform decay of the of the source were removed by relating measure- ties in the radiation pattern in the same direction at direction to measurements made ments in a particular relative sound pressure of 5 feet from the source. These values of a distance 9h for octave-band center frequencies are presented in Figures 9a through level free-field respectively. For reference, the theoretical 63 Hz through 8 kHz, through the 5-foot per doubling of distance is superimposed decrease of 6 dB data point in each plot.
farther than in other direc- in the west direction extended Since measurements For convenience, they are these were chosen for closer examination.
tions, be seen that several maxima and grouped together in Figure 10, where it may be a smooth exponential decay with distance might minima occur where expected.
a mathematical explanation of these perturbations, To attempt an analytical The basic and translated into a computer program.
model was constructed sound could be ignored made in this model was that all reflected assumption This assumption is justified for a first- except that from the test platform.
source com- due to the relative proximity of the platform to the order model the high reflection reflecting surfaces, and also due to pared with other to other reflecting surfaces. Since coefficient of the test platform compared the current tests, material was installed on the platform during no absorbent reflecting surface.
assumed that the platform was a perfect it was into a large number of sinusoidal Each octave band of noise was decomposed spectral distri- whose magnitudes were weighted by a pink noise components frequency component for the which were then added vectorially at each bution, sums were combined to form an direct and reflected waves. The resultant shown in the westerly direction for each octave band as interference field in The sharp drop (Similar work is reported in References 5 and 6).
Figure 10.
of the sudden absence of the distance of 42 feet is representative at a horizontal Comparisons in Figure 10, of the mea.- reflected wave as the platform ends.
from the first-order reflection model sured sound field and that calculated combined with the above, show similar trends. This similarity, described source compared with other proximity of the test platform and the relative sufficient to conclude that reflections from reflecting surfaces, is considered the test platform are a major contributing factor to the deviation from exponen- tial decay of sound pressure level with distance from the source Clearly, a complete explanation of all observed phenomena in the sound field requires a substantially more complex model incorporating additional first and higher-order reflections, adjustments for source directivity, and substi- tution of actual impedances at reflecting surfaces. For example, if the phase change on reflection at the platform were -90 'at 125 Hz due to structural resonance, there would be improved agreement with measured data as shown in Figure 10.
A direct comparison of measured data with that from the previous experiment from Reference 2 is shown in Figure 9e. Only the 1000-Hz octave-band data from Reference 2 was available for the comparison; however, the figure shows that levels of reflected sound are generally lower than in the previous experi- ment.
Due to the nonuniform decay of SPLwith distance, combined with directional dependence, it was not possible to derive hall radius values in the same man- ner as in Reference 2. Instead, a parameter analogous to the hall radius, but purely empirical in character, was used. This parameter is measured under the assumption that near the source only the direct field is significant.
The direct field at larger distances is then calculated from an- inverse square law decay (6 dB for doubling of distance) superimposed on a measurement of sound pressure level taken near the source.
The distance at which measured values deviate from the inverse square law decay by 3 dB is taken to be the "modified hall radius". At this point, the reflected sound field equals the direct sound field in power. Values of the mod- ified hall radius derived in this manner are presented in Table I. The average modified hall radius over all measurement directions for each octave band is plotted in Figure 11, and is compared with the hall radius data reported in Reference 2.
Drawing any significant conclusions from this comparison is difficult due to the presence of dominant first-order reflections.
In this regard, however, it is clear that the full benefit of the sound-absorbent material on the walls and ceiling of the test section is unlikely to be derived unless adequate sound-absor- bent material is also added to the test platform.
3. 3 Measurement of Reverberation Time in Test Section To measure the sound decay rate in the test section, the room was impulsively excited by a gun blast at numerous locations. Source positions near the room corners were chosen in order to excite the highest number of normal modes of the room. Microphones were located at various positions throughout the test section, all at least 40 feet away from the source and at least 10 feet from any wall. From tape-recorded decay signals, octave-band decay charts were obtained with a B & K graphic level recorder at a writing speed of 200 mm/sec and a paper speed of 30 mm/sec. Example decay charts are shown in Figure 12.
The decay curves for octave bands centered at 125 Hz, and above, showed dis- tinctive double-slope character.
This is a familiar characteristic of sound decay in rooms having one set of walls more absorbent than the others. In this case, the large duct openings in thenorth and south walls act as highly absorbent surfaces. The initial slope of the decay curve is representative of the maximum energy absorption rate of the room, and thus, is indicative of the total amount of acoustic absorption within the room (Reference 7).
Fourteen sets of source and microphone locations were used for the test. The average "initial" reverberation times for each octave band, calculated from the initial slope of the decay curves, are presented in Table II. The standard deviation for each data group is also shown. The temperature and relative humidity in the room during the tests were 45*F and 55%, respectively.
These results are compared with those obtainied before the' addition of sound- absorbent material in Figure 13. The figure shows that reverberation times are apparently reduced in the midfrequency range after addition of the sound- absorbent material. A completely unambiguous comparison is not possible, however, since in rooms of this size, air absorption significantly affects reverberation times, and no record of relative humidity or temperature were reported for the previous set of tests.
To clarify this comparison, an attempt was made to analytically evaluate the effect of different combinations of temperature and relative humidity on rever- beration times. This effort failed in its intention but provided further important evidence of the inapplicability of "Sabine acoustics" to the test section of the full-scale wind tunnel.
Reverberation time in a large room, under the Sabine assumption of a diffuse field, is given by (Reference 8): 049V 0.
T T = aR d---4m- where T = Reverberation time (seconds) V = Volume of room (cu. ft.) (',7 x 105 cu. ft. for test section of full-scale tunnel) a = Absorption of interior enclosure surfaces (sabins) -l m = Air absorption parameter (ft.- ) For the current test, at a temperature of 45 F and 55% relative humidity, "4m" has the values shown in Table III at octave-band center frequencies.
Solving for "a" in the above expression, using these values of "4m". and the experimental reverberation times given in Table II, yields somewhat unex- pected results. Above 2000 Hz, values of "a" are negative, , indicating a net energy increase on reflection. Clearly, the basic premnises mifst be faulty and an analysiskied uponthe Sabine assumption is unreliable.
At lower frequencies (octave bands centered on 31. 5, 63, and 125 Hz) rever- beration times measuredliduring the current test are longer than in the previous test. This is perhaps due to a different interpretation of the decay character- istic, since the presence of a double slope at these frequencies is open to question (see, for example, the upper curve of Figure 12).
4.0 CONCLUSIONS As originally stated in section 2.0, this study had two objectives: a. To assess the effect of the addition of sound-absorbing material to the test section, b. To formulate guidelines for future acoustic measurements in the test section.
With regard to the first objective, it is probable that the anechoic character of the test section of the full-scale wind tunnel has increased due to installation of sound-absorbent material on the side walls and ceiling of the test section.
This is indicated by a decrease in ambient noise level during tunnel operation and the decreased reverberation times in the midfrequency range, the mag- nitude of which are unlikely to result from differences in relative humidity between tests.
During projected aeroacoustic tests, however, noise sources will be mounted above the test platform. In tests of this nature,it is likely that reflections from the platform will mask any significant improvement gained by the instal- lation of the sound-absorbent material. It is therefore suggested that the sound- absorbent material recommended in Reference 3 be installed on the platform surface and that qualification tests with an omnidirectional source be repeated.
In the case of the second objective defined above, it is difficult at this stage to formulate definitive experimental guidelines. In Figure 14, it may be seen that a directional source may cause reflected component magnitudes significantly different from those of an omnidirectional source. In this example, noise directed preferentially towards the ceiling may cause a reflected field of comparable magnitude to the direct field at critical points below the noise source.
This effect would not be observed in the case of an omnidirectional noise source.
It is suggested, therefore, that further tests be performed with directional sources, as there is sufficient evidence to indicate that the room is nonuniform in its absorbent characteristics.
REFERENCES 1. Bender, J., et al, "Aeroacoustic Research in Wind Tunnels: A Status Report, " NASA CR-114575, Pennsylvania State University, February 1973.
2. Ver, I. L., Malme, C. I., and Meyer, E. B., "Acoustical Evaluation of the NASA Langley Full-Scale Wind Tunnel," NASA CR-111868, January 1971.
3. Ver, I. L., "Acoustical Modeling of the Test Section of the NASA Langley Research Center's Full-Scale Wind Tunnel, " BBN Report No. 2280, November 1971.
4. Rettinger, M., "Acoustic Design and Noise Control," Chemical Publishing Co., New York, 1973.
5. Franken, P. A., "A Theoretical Analysis of the Field of a Random Noise Source Above an Infinite Plane, " National Advisory Committee for Aeronautics, Technical Note 3557, 1955.
6. Howes, W. L., "Ground Reflection of Jet Noise," NASA Lewis Research Center, Technical Report R-35, 1959.
7. Embleton, T. F. W., "Absorption Coefficients of Surfaces Calculated from Decaying Sound Fields, " Journal of the Acoustical Society of America, Vol. 50, 1971, p. 801.
8. Beranek, L. L., (editor), "Noise and Vibration Control," McGraw-Hill Book Company, New York, 1971.
TABLE I "MODIFIED HALL RADIUS" DISTANCES MEASURED FROM SPL VS. DISTANCE DATA Direction Modified Hall Radius, ft for Octave-Band, Hz of - from 125 250 500 1000 2000 4000 8000 Source 63 North 17 21 >25 22 >25 >25 >25 >25 South 12 20 >25 20 16 >25 >25 >25 East 13 >20 >20 >20 12 14 >20 >20 West 17 22 26 26 28 20 23 24 Vertical 15 22 19 25 >35 >35 31 >35 Average 14.8 >21.0 >23.0. >22.6 >23.2 >23. 8 >24.8 >25.8 " > " signifies that measured octave-band sound pressure levels were always less than 3 dBab.aove the theoretical free-field -6 dB/doubling distance line (passing through measured SPL at 5 ft) for all positions less than maximum measurement distance'.
TABLE II "INITIAL" REVERBERATION TIMES Octave-Band Average Standard Center Frequency, Reverberation Time*, Deviation', Hz sec sec 31.5 2.26 0.53 63 2.37 0.45 125 2.01 0.21 1.84 0.32 2.04 0.53 1000 2.05 0.40 2000 1.86 0.32 1.35 0. 12 8000 0.94 0. 09 -14 Source and microphone sets TABLE III Parameters in "Sabine Acoustics" Analysis Absorption Absorption of Interior Octave-Band Volume Coefficient 4m (ft,_-) Surfacesa(Sabiris) Center Frequency 1.53 x 104 x I0 3.62 31.5 - 5 1.44 x 104 7.25 x 10 - 4 1.61 x x 10 1.565 x 10 1.70 x 10 250 2.89 - 4 1. 18 x 104 5.96 x 10 - 3 x 103 5.70 x 10 1.46 - 3 x 104 -1.1 x 10 4.29 - 27 x 104 -6.
x 10 2 1.27 x 105 -2.12 x 10-2 3.64 ...... _ _Plan View I . . ...
i East (C ontrol Room Side)J [, -63 outh
0North - J 42
tream (Upstream) .. Test Platform (Door Side) West , .. S Section View -:,
30'
P-sitions Microphone
O
zi Sound Absorbent Material] Figure 1. Plan and Elevation Views of Full-Scale Wind Tunnel Test Section Showing Microphone Positions for Ambient Noise Measurements.
16- February 15 - -- Microphone Position No. 4 on.
Compressor .-. . Microphone Position No. 4 Compressor off.
70 - February 16, 1974 ----- Microphone Position No. 1 No. 2 Microphone Position 60 "_--.
Position No. 3 -> -.---------- Microphone - 40- Cl) - 30- I I I I I I I 31. 5 63 125 250 500 1000 2000 4000 8'000 Octave-Band Center Frequency, Hz Figure 2. Ambient Noise Level in Test Section.
110- 100- ~65.
a 80- 56.
6!
70-
;4 33
0 60
Functon oAirnSeled: PostinpNoh 2 50- 0\ 40- Microphone Position 2 30-
I I I I
I I I
31. 5!
63 \ 125 \ 250 500 \ 1000 \ 2000 4000 8000!
Octave-Band Center Frequency, Hz, Figure 3a.
Octave-Band SPL in Test Section as a Function of Air Speed: Position No. 2.
110- Tunnel Speed (mph) 90 -
75. Z
65. 41 S56.6 46. 31 a) 70 - (U 60- O .50 - 40- Microphone Position 3Q-
IIII
I
I
I
31.5 63 \ 125 250 500 1000 1 20001 4000 Octave-Band Center Frequency, Hz Figure 3b. Octave-Band SPL in Test. Section as a Function of Air Speed: Position No. 3.
110- Tunnel Speed (mph) 75.
8 65.4 PQ 56.
46. 31 70 - (A U) U) O
50-
40 - Position Microphone 30 --
I I I I I i I
31. 5 63 125 250 500
1000 2000 I 4000 8000
Octave-Band Center Frequency, Hz Figure 3c. Octave-Band SPL in Test Section as a Function of Air Speed: Position No. 4.
= Normalized Octave-Band SPL Octave Band SPL - 60 logl0u, S-10 where u = tunnel speed (mph) - 20 S-30 - -40 - W Microphone Position 2
E
0. Microphone Position 3 Position 4 -50 - Microphone
I I
31.5 63 125 250 500 1000 2000 4000 8000 Octave-Band Center. Frequency, Hz Figure 4. Normalized Octave -Band Ambient Noise Level.
o ,.0
O
*0 SPL (OCT) = SPL(OCT) - 60 log U 1 0 whe re: S SPL (OCT) = Octave-Band Ambient Noise Level U = Air Speed in MPH 125 500\ 1000 20 4000 1 8000 31.5 63 Octave-Band Center Frequency, Hz
* With sound absorbent treatment
O Prior to sound absorbent treatment of walls
microphone positions).
and ceiling of test section (Reference E I (Average of 3 Figure 5. Comparison of Mean Normalized Octave-Band Ambient Noise Measurements Sound-Absorbent Treatment of Test Section.
Made Prior to and After Tunnel Speed (mph)
10 dB
75.2 56.6 an) l I - Microphone Positionlz 0 - . 20 ---- -- - 40 -- - 60 80 -- 100 120 - 140 160 Frequency, Hz Figure 6.- Narrow-Band Spectra of Ambient Level with Tunnel Running, Showing Predominant Rotational Fan Noise Peaks -12 4 + 105'
SPL-PWL 10 logl (
_ 4 rr a - 6 2 -16 SPL= Sound pressure level re 20 x 10 N/m -162 PWL = Source power level re 10-12 watt -20 I III Totalabsorption, a (Sabins) -24 )
20,000 LU
30, 000 -28
60,000
-36 5 10 15 20 25 30 40 50 70 Distance from Source r, feet Figure 7. SPL vs. Distance from an Omnidirectional Point Source as a Function of Total Room Absorption, Assuming a Diffuse Reverberant Field.
Plan View= A Source Suspended 10 Feet Above CenterlOf Platform +Microphone Positions Level With Sour ce East
60' North T- - South 63'
20 10 1 0 20 :20 50o West 52_9'1 52_8 -Section View VerticalA .30
- 3-31' 1
A Source At Center.
Of Platform _
+ Microphone
Positions Directly Above Source Figure 8.1 Source -- andl-Icrphone Positions for Sound Field Measurements.
No rth East 0:
2 12 *
0 W -10- -154- U)-20
3l* *
0- 0 0 - -* * * I I Ii I I I I I 5 2- 15 20 **D
Distance from Source, ftf
| Vert ic-al1
0-0 >* 0 -10-* f: -15 o -ZC 15 20 30 40 50 5 ft 5 10 *Relative to SPL at Distance from Source, ft Figure 9a Measured 63 Hz Octave-Band Sound Pressure Levels in Indicated Directions from Acoustic Source Located at Center of Test Platform.
I % i-Nor th I East_ 0 .
* -5 0-1 Q,
I I I
I I I
I I I | |
> 4 -5 0 0 5 (10 015 20] fj~ -15
Distance from Source, ft 1
(1 1 5 z 1 1 1 1
- 2 I
ft Distance from Sour-ce, Vetical ...
1=~-0 - 5 10I I 15 I l I 50 2Relative to SPL at 5 ft] 0. 40 0 SDistancefrom Source, ft 10 152 SPL atO 5 f P-Relative to Ditance ff rom' Source, ft 125 Hz Octave-Band Sound Pressure Levels Figure 9b. [Measured Sound Pressure Levels Figure 9b.j Measured 125 Hz Octave-Band ..... lin Indicated Directions from Acoustic Source Located I Located' __in Indicated Directions from Acoustic Source Center of Test Platform.
Sat ..
...
0 2 0 II ..
at Center of Test Platform.
East K North I O - 0
10-
. - 0
-15 I I I I I I 0 I I I v
-0 -
-- 4 S-15 a0 I - I -20 15 \20\25 5 10 Distance from Source, ftl Vertical
.- 1 -
Distance from Source, ft Octave-Band Sound Pressure Levels Figure 9c. Measured 250 Hz in Indicated' Directions from Acoustic Source Located at Center of Test Platform.
0 - - -15 .- 20 5) - 0- SI--10
I I 1 I I I I I I I
5 10 151 20 25
\Distance from Source, ft_
rVertical 4- I) -15 - 30 50 Relative toSPLat 5ft10 1520 ft Distance from Source, Figure 9d. Measured 500 Hz Octave-Band Sound Pressure Levels in Indicated Directions from' Acoustic Source Located at Center of Test Platform.
ast SNorth 1E 0 -[- -A 15 - E-15 -c-10 * Suth.
> 0 0 o -C A -
*
from Source, fft Distance Present Measurements Measurements.
From ReferenceF21 Q -53 *Relative to SP Sound Source "on ground" Ao o Sound Source "elevated" A Source "on ground" 1-10 ASound from Acoustic Source Lortcated in Indicated Directions k 15- 30 40 501 10 15 20 to SPL at t -Relative ft NDistance from Source, Figure 9e. Measured 1000 Hz Octave-Band Sound Pressure Levels in Indicated Directions from Acoustic Source Located at Center of Test Platform, with Comparison to Previous Measurements.
S-oti East -5 0 - *5.
5a S 10 a 5 20 -20 5 110 151120 23 5 Distance from Source, ft EO -10-- 3-1 elReative Ito SPL at 5 f t 20 -------- - - - --- 0 L2_1 C31 4~J 501 Distance from Source, ft Figure 9f. Measured 2000 Hz Octave-Band Sound Pressure Levels in Indicated Directions from Acoustic Source Located at Center of Test Platform.
Sivorth
IEastjI a) -15
South We
0ae~a -20 (00 11511 20125 ! 10
I)5
ft --
from Source,
Distance
l
Terv tica
,a, easred400 H OcaveBan SundPresur L'Ul
e Fgur~g
inInictd irctos ro Aostc ouc Lc0e i~~~~CC -5Cne o etPltom
L5- L L 151 20 3040 150
-4(CC) Distance from Source, ft Measured 4000 Hz Octave-Band Sound Pressure Levels pgre g.
in Indicated
D Lrections
from Acoustic Source Located at Center of Test Platform.
32 _ S-Northl Ea st -a -10 5 1 1-20 Sout h We Srct o S-1 -0 EAI0II I, -0 1 20125
w 151 10 15
Distance from Source, fDirections from Acoustic Source Located of Test at Center Platformertical -15 a[ t 33- 50] ZO 30 40 i f 1 15 ea t S a ft Distance from Source, Figure 9h. Measured 8000 Hz Octave-Band Sound Pressure Levels in Indicated Directions from Acoustic Source Located at Center of Test Platform.
3 3 W st iwest , .
N.o
,~
-5
0 9
0 01
1 5 0-,, -_,10-\ 6Hz 125Hz O
I I I I I n I I
I I I I
4) '.-= -1H] * *< " I" I I I_. I .I U-,E -A r5 an 0a -3 indicatede eflection coefficient = I , with latform r -44 with Sound Levels of Measured Octay e-Band 10.. Comparison igure from Acoustic Source.
a Function of Distance Reflected Sound, as
34i
_ 40.
a 20 U ,• cn
31. 5 3 125 250 I 50.0 l 20000 ,40 0. 1 8000
Octave-Band Center Frequency, Hz, O "Hall Radius" prior to sound-absorbent l "Modified Hall Radius" with sound-absorbent treatment of walls and ceiling of test treatment. (Determined graphically from section (Reference 2).
Octave-Band SPL vs. Distance curves.)
Figure . Comparison of Average Octave-Band "Modified Hall Raius" -- Distances.ith "Hall', Radius'!.Values Measured Prior to Sound-Absorbent Treatment"of Test Section.
T R 2. 3 sec 1000 Hz* 10dB \T 1. 5 sec RI= 1.5TR= 5.6 sec sec l e - s l *Sound decay had distinct "doub ope" character in 125-, 250-, 500-, 1000-, 2000-, 4000-Hz octave bands.
Figure 12.] Typical Octave-Band Sound Decay Curves, Showing Measured Reverberation Times.
36- c 4 U C.
1 j63 g 1 4 100,
50 06 5 1.2.00 1000
2000 1 400 7 -8o
Octave-Band Center Frequency, Hz i OPrevious measurements - prior to sound 0 Present measurements - with soundi I abs or bentitreatment of walls and ceiling absorbent treatment' of test section (Reference2) Figure 13. C omparison of Average Octave-Band Reverberation Times with Measure- ments Made Prior to Sound-Absorbent Treatment of Test Section.
Aircraft Noise Source SOmnidirectional Noise Source
16=• 1
30'[
Figure 14. Effect of Reflected Component Magnitudes on Measured Sound Pressure Levels Due to Sources with Differing Directional Characteristics.