Document
N O T I C E
THIS DOCUMENT HAS BEEN REPRODUCED FROM
MICROFICHE. ALTHOUGH IT IS RECOGNIZED THAT
CERTAIN PORTIONS ARE ILLEGIBLE, IT IS BEING RELEASED
IN THE INTEREST OF MAKING AVAILABLE AS MUCH
INFORMATION AS POSSIBLE
NASA Technical Memorandum 81343 l (NASA—TM-8134J) CCMPAVISCN OF ACOUSTIC DATA N32-1608.3 FROM A 10:1 MH CONIC NOZZLE AS BEASURED IN THE RAE
24 —
FOOT WIND TUNNEL AND THE NASA AMES 40 — BY 80 — 'FOOT WIND TUNNEL (NASA) 2J P Unclas HC A 02/M F A01 G3/07 07988
of Acoustic Data From a
Comp ,
arison
.dg
o2mm Conic Nozzle as Measured
foot Wind Tunnel
-
in the RAE 24
An by-80-
QA.Ame- Q
^nnrlfh® KIO foot
..%w NNW .w, n n n ^r^v^—^. rte %31%o n
" B e"
Wind Tunnel
J. McKie
and
A. Atencio, Jr.
N January 1982 FEB IM RECEIVED .NASA m FACIUry
Accra
DEPT,
SA
A National Aeronaulfco and Space Administration.
_: NASA Technical Memorandum 81043
of Acoustic Data From a
Comparison
102mm Conic Nozzle as Measured
in the RAE 24-foot W ind Tunnel
_1 A IV. - i• r w ^w w _..... w A L--- n ^ X._ i
n - es W-uly-OU-rvvi
n^ au the NIAZIA-WAU-
Wind Tunnel
A. Atoncio, Jr,, Amos Rosearch Canter, Moffett Field, California
J. McKie, Royal Aircraft Establishment, England
NASA
National Aeronwitics and a oe Administration Ames Resoarch Conter Moffett Fjold, California 194035 SYMBOLS D diameter of jet, m f frequency, Hz OASPL overall sound pressure level, dB SPL sound pressure level, dB velocity of jet, m/sec Vjet VT velocity in wind tunnel, m/sec fD Strouhal number Vj et °K degrees Kelvin reference angle to inlet PRECEDING PAGE BLANK NOT FILMED mm CONIC NOZZLE COMPARISON OF ACOUSTIC DATA FROM A 102 MEASURED IN THE RAE 24-FOOT WIND TUNNEL AND THE AS NASA AMES 40- BY 80-FOOT WIND TUNNEL A. Atencio, Jr. and J. McKie* Ames Research Center SUMMARY A cooperative program between the Royal Aircraft Establishment (RAE), England, and the NASA Ames Research Center was initiated to compare acoustic measurements made in the RAE 24-Foot Wind Tunnel and in the Ames 40- by 80-Foot Wind Tunnel. The acoustic measurements were made in both.facilities using the same 102 mm conical nozzle supplied by the RAE. The nozzle was tested by each organization using its respective jet test rig. The mounting hardware and nozzle exit conditions were matched as closely as possible. The data from each wind tunnel were independently analyzed, by the respective organization. The results from these tests show good agreement. However, in both facilities interference with accustic measurement is evident at angles in the forward quadrant.
INTRODUCTION Iii 1976, an agreement between the RAE and NASA Ames Research Center was initiated whereby a 102 mm conic nozzle that had been previously tested by the ItAE in its 24-Foot Wind Tunnel at Farnborough, England, would be tested in the Ames 40- by 80-Foot Wind Tunnel. The purpose of the test was to determine if the two organiza- tions would obtain the same results from the same nozzle after each organization independently tested the nozzle, and then used their respective methods to reduce and analyze the data. The RAE supplied Ames with the 102 mm nozzle and drawings showing their test apparatus and measurement approach. In addition, the RAE supplied Ames with reports from the previous tests. The test was performed in the Ames wind tunnel during the summer of 1977. The 102 mm nozzle was tested both hot and cold, and at wind tunnel velocities that matched previous test data from the RAE. The results and comparisons made with the previous set of data are presented in this report.
FACILITIES The 102 mm conical nozzle was tested in the RAE 24-Foot Wind Tunnel and in the Ames 40- by 80-Foot Wind Tunnel. The difference in the facilities created a unique test environment. The following sections describe the wind tunnel test chambers and give some details on i:he installation of the 102 mm nozzle.
*Royal Aircraft Establishment, England.
t The Royal Aircraft Establishment Test Arrangement The RAE 24-Foot Wind 'Tunnel at Farnborough, England, is an open jet test section ,het exhausts into an acoustically treated chamber of circular cross section. The open in circular col- that is 13.5 in x 13.5 9 m. The downstream face of the chamber is a lector and, immediately following the collector, is the drive fan which exits to a in jet nozzle. Figure 1 shows short rectangular return circuit back to the 7.2 the wind tunnel test chamber and collector. The wind tunnel is capable of speeds up the useful range to 50 m/sec, but background noise levels the drive fan reduce from for acoustic tests to velocities of 30 m/sec or less, with frequencies 500 tiz and lower still dominated the tunnel background. The test nozzle was mounted as shown by in figure 2 (ref. 1). The wing rig is placed vertically in the test section, and the conical adaptor to a cascade r..orner. This test nozzles are mounted by means of a turns the airflow to the tunnel downstream direction. The nozzle flow pipe wps treated to minimize rig noise. Reference 1 descritoes the installation in more detail.
The NASA Ames 40- by 80-Foot Wind Tunnel Test Arrangement The 40- by 80-Foot Wind Tunnel at NASA Ames Research Center; Moffett Field, California, is a closed-return wind tunnel with a constant area test section measur- in to long. The cross section has 6.1 in ing 12.2 in high by 24.4 by 24.4 in circular sides, and a middle section that is 12.2 The wind tunnel is capable of test section speeds up to 200 knots (103.0 m/sec). The floor and part of the curved side walls were treated with a 76 mm (3 in.) acoustic foam mat during the test of the 102 mm nozzle as shown in figure 3. The mat was designed to increase the effective hall radius for acoustic measurements and; for this test, was effective down to 500 Hz. Even though the untreated portions remain a source of reflection and reverberant field noise in the closed test section, the treatment increases hall radius two to three times. The background noise levels measured for this test are shown in figure 4. This noist is created by the drive system, the aerodynamic noise of the test rig, and self noise of the microphone. The 102 mm nozzle was tested in two different configurations in the 40 x 80. The first configuration was the nozzle mounted to the standard rig S-duct. The S-duct is a transition tube from the burner, which sits on the wind tunnel floor and rises to a model exit height of 1.52 m. The S-duct attachment is sketched in figure 5. The second mounting configuration was the shown in figure 6. This rig nozzle mounted to a wing rig, attached to the S-duct e.8 mounting was meant to match the RAC nozzle mounting.
The Ames model jet burner is a kerosene fueled burner that was designed for nozzles of 135.2 mm (6 in.) diameter and for continuous hot operation at 810 K (1000 F). The burner air was supplied by bleeding the air mass flow from a Rolls Royce Viper engine compressor. The compressor was driven by the hot exhaust from a G.E. J-85 Turbojet. The compressor was located outside the wind tunnel in an acous- tically treated enclosure (Fig. 7). The air was ducted through more than 15 in pipe to the model burner. There was a remotely operated dump valve at the entrance to the feed pipe to control the air mass flow from the compressor. There was also a short section of pipe at the burner entrance that was acoustically treated to absorb the compressor noise. Since the burner was operated at 820 K, there was no acoustic treatment in the burner ;ind before the nozzle exit on either the S-duct con- figuration or the wing configuration. The wing configuration was made from steel in order to withstand the high temperature. Since the burner was operated at an off design point, there was a tendency for the burner can to "carbon up." When the burner was cold, the carbon would break loose from the burner can and, on subsequent {{ r i few hot operation, chunks of carbon would be forted out of the nozzle exit for a seconds. Therefore, the screen (used by RAE before the nozzle exit) was not the used and no data were taken until the burner was free of the carbon chunks. The expelled carbon was about the size of small pea and was not considered large a enough to change sound characteristics.
140DEL of The test nozzle supplied by the RAE. was a conic nozzle with an exit diameter 70.
102 mm. The nozzle barrel was 73 cm long and was designed with a half angle of system as a (Fig. 8.) The Ames wing configuration was designed ueing the RAE mount guide. However, the Ames wing was made from steel and was not acoustically treated because it was intended to be run with the hot ,flow. The RAE supplied cascade elbow was used only with the Ames wing configuration to turn the flow streamwise at the nozzle attachment. An Ames designed cascade corner was used for the attachment and turning off the burner S-duct. Th jet exit height above the wind tunnel floor was maintained at 1.52 m for both configurations.
INSTRUMENTATION RAE Instrwnentation The RAE instrumentation for the 102 mm nozzle consisted of a single pitot pres- sure probe upstream of the cascades in the corner of the elbow where the 102 mm nozzle was attached. The pitot probe was calibrated using a reference pressure at the exit plane of the nozzle. The air total temperature was measured in the plenum chamber of the air supply. The RAE microphone instrumentation consisted of a single 6-mm (1/4-in.) Bruel and K3aer (B&K) microphone fitted with a nose cone and aligned with the tunnel flow. The microphone was traversed on a 3-m sideline at nozzle exit centerline height, but acoustic measurements were made at fixed positions along the traverse. The data were not corrected for convection.
40 by 80 instrumentation The 40 by 80 test rig had pressure and temperature instrumentation at the coupling Joint at the end of the S-duct. Two area weighted rakes were used and each had ten probe positions: one was total pressure and the other was total temperature.
The readings from these rakes were input to the 40 by 80 computer system where calculations were made to determine jet exit velocity and pressure ratio. These quantities were continuously displayed on digital panel meters, and update was achieved by the real time section of the 40 by 80 computer system. Hard copies of the data were available following a data point. In addition to the model instrumen- tation, tunnel parameters were obtained using the tunnel instrumentation system. No thrust measurements were made since the jet rig is not designed for metric measurements.
Bruel and Kjaer 6-mm (1/4 in.) microphones with attached nose cones were used for all acoustic measurements, The data were measured by traversing a single microphone along a 2.96-m sideline parallel to the jet axis. The microphone covered angles
upstream toward
from 27° to 166° relative to the jet exhaust plane with zero degreea
an imaginary inlet. The signal from the single microphone was passed through a B&K then through two separate amplifiers where gain settings
signal conditioner, and
were made. The two separate amplifiers were used to assure a good signal-to-noise
was determined using
ratio throughout the traverse. The position of the microphone y rotating potentiometer which gave a continuous reading of voltage level. The the traversing rail, and then to
voltage level was transferred to linear position on
posi-
Angle by a second transformation. The microphone acoustic signals and traverse tion voltage signal were recorded on magnetic tape using an Ampex 14 track recorder.
The recorder was set up with direct record modules to allow recording of the higher frequencies at reasonable tape speed. However, the dynamic range or signal/noise around 31.5 kHz
was
ratio was reduced to about 25 dB and the upper limit on frequency adequate for the 102-mm nozzle. As it was necessary for the
at 30 ips, which was
microphone to always point upstream, the orientation of each microphone nose cone to the jet changed constantly with distance down the traverse. Corrections for the change in microphone directivity, as a function of the orientation, were applied using an "average correction." This was very accurate up to 20 kHz and was good to ±2 dB for higher frequencies with the largest errors occurring at shallow angles to the jet. Other corrections made to the data were for atmospheric attenuation and tape recorder response. The speed of the microphone traversing system was set by a motor controller. The speed was set without wind blowing through the test section.
The traverse time used was approximately 5 min with slight variations depending on wind speed and microphone cable drag.
TEST PROCEDURES The test prr_-redure for the RAE tests is outlined in reference 1. Basically, conditions were sei ,, for the nozzle to maintain a certain total pressure at the regu- lated temperature of 300 K, and wind speed was set at either 0, 15, or 30 m/sec. The microphone was then traversed on the sideline to fixed locations where data were recorded. The positions were 145°, 135°, 120°, 105°, 90°, 75°, and 60° reference to an imaginary inlet.
The test procedure used during the test in the 40 by 80 was to set all flow conditions at the nozzle exit for desired temperature, pressure ratio, and exit velocity. When the conditions were set, the microphone was quickly traversed to upstream so that gain settings could be made. The microphone was then slowly tra- versed downstream from forward of the nozzle exit to the aft stop to take data.
Traverse time was about 5 min. During the traverse, propulsion data were recorded at selected points which were later averaged to get jet nnzzle test conditions. The data taken during the Ames tests included both "cold" and "hot" flow data. The operating without the burner being turned on. The a3.r "cold" data were obtained t7 from the Viper compressor was simply passed through the burner can to the S-duct, to the nozzle. Since the compressor operated with ambient air, the nozzle exit tempera- ture depended on pressure ratio through the compressor. There was no means of inde- pendently controlling temperature within the supply system so that an exact match with the RAE results was not possible. The "cold" flow temperatures ranged from 320 K to 360 K, whereas the RAE cold flow was maintained at 300 K. The "hot" flow temperature was maintained by regulating the fuel flow through the burner. Tempera- ture was maintained within t5 K for most test conditions. Hot operation, however,,
did require that the nozzle be run without the screen gauge that was supplied by the
RAE. The Ames model burner was designed for Larger exit area nozzles than the 102 mm, be fuel rich, causing some carbon buildup inside and off design operation tended to was cooled between runs, the carbon When the burner the burner combustion chamber.
buildup would break up due to difference in expansion. When the burner was the a few sec- restarted, the carbon chunks would be expelled through the nozzle exit for onds until gone. No data were taken under these conditions, Several runs were made for both the S-duct configuration and for the wing con- figuration. A summary of runs and conditions is given in table 1.
DATA REDUCTION The data taken during the RAE test were reduced through a GR 1926 real time third-octave spectrum analyzer from 315 Hz to 80 kHz using an integrating time of 2 sec. Digitized paper tapes output from the analyzer were processed further using a digital computer. The computer calculations included corrections for microphone frequenc y response and directivity, atmospheric attenuation, and summations of the propulsion data were processed using the one-third-octave levels to give OASPL. The total pressure and temperature measurements to calculate jet exit velocity assuming isentropic flow.
The propulsion data taken during the Ames test were reduced using a program on the Ames 40- by 80-computer system. Most of the propulsion data were available imme- diately following each wind tunnel, run. The calculations provided wind tunnel veloc- ity and ambient temperature, jet nozzle velocity, jet pressure ratio, and jet exit temperature.
The acoustic data from the traversing microphone were reduced using an Ames wind tunnel system called the Dynamic Analysis System (DAS). The DAS is a synthesized system composed of several special function components that are coupled to a PDP 11 minicomputer. The acoustic data were reduced through a GR 1925 one-third-octave band analyzer that was triggered by commands from the DAS. The commands were part of a special program written to reduce traversing microphone data. The output from the DAS was put on punched paper tape and was later reduced on an IBM 360 computer through a general acoustic program. The acoustic program provided microphone response corrections, tape recorder response corrections, atmospheric attenuation corrections, and source convection correction for wind on data. The final listings included angular position. and SPL for each one-third-octave band frequency. Convection cor- rections were applied to the nozzle exit rather than to actual source location. Tile correction relates initial sound emission angle (6) and the geometric angle (Y) at a particular microphone position. The relationship is: sin 0 Tan MA - cos 6 where MA = ` TaNEL flow Mach number ► ao = ambient sound sp`2d In addition to the angle change, there is also a small level correction to account for the difference in microphone position as measured and the corrected position. The correction is simply: New Radial Distance 1 -20 Loglo (Measured Radal Distance) At 30 m/sec, the level corrections range from 0.1 dB at 90 0 to 0.75 dB at 160 0 . The o change in angle ranges from .5° at 90 to -2° at 160°.
RESULTS the 40 by 80 and the RAE 24-root Wind Tunnels is Comparison of data from limited. Only cold flow dat with the wing rig were measured by the RAE for the 102-mia nozzle. The Ames data for hot flow had to be compared to a limited set of data from smaller nozzles run hot by the RAE in previous tests. The hot data had to be normalizedto account for scale difference and temperature difference. Ames tests without the wing.
included both a wing rig mounting and a simple S-duct attachment A comparison of these data is shown in figure 9 for zero wind speed and cold flow.
Thcre is essentially no difference in the data with or without the wing. However, since the RAE data are for the wing cot.figuration, only data taken in the 40 by 80 with the wing configuration are compared in this report. Also, the RAE data are pri- marily restricted to angles in the aft quadrant of the jet. The initial comparison These of noise from the two facilities for the 102-nun nozzle is shown in figure 10.
data are for the 90 0 position for nominal zero wind tunnel speed. In addition to the facilities data, a jet noise prediction curve is shown using a prediction method from reference 2. The data comparisons show differences up to 4 dB in the lower frequen- at the cies, almost match at mid frequencies near the peak, and 3-4 dB separation high frequencies.
The Ames data are consistently below the RAE data. Comparison with the pre- data above the prediction for low frequencies, and .almost dicted curves show the RAE exact for mid and high frequency. The Ames data, on the other hand, match the pre- diction at the low and mid frequencies, and are below the prediction for the higher frequencies. The differences are probably attributable to the respective facilities and jet test rigs. The low frequency behavior of the RAE" data may be due to the close proximity of reflecting surfaces to the test nozzle, or possibly some residual noise from the model flow system (ref. 1). The high frequency behavior of the Ames data suggested a tape recorder response problem; in fact, the data above 31 kHz (not shown) appear to be spurious electrical noise.
Since it is recognized that severe problems exist with the RAE data at rrequen- cies below 250 Hz and for the Ames data above 31.5 kHz, the comparisons to follow will be restricted to cover only 250 Hz to 31.5 kHz. When the data were compared for a more aft angle near peak jet noise as shown in figure 11 (135° position as measured from the inlet), the data from the two facilities come closer together, suggesting that facility differences are being masked by the jet noise. However, agreement is exact only for the middle frequencies. The data from the 40 by 80 are still lower for the low frequencies and high frequencies by a margin of 1-3 dB or about 1 dB or so Jess than shown for 90°. The spectrum shapes at the aft position are much more consistent. The results for wind on are shown for these two positions j,n figures 12 and 13. 1' The RAE data are line-of-site data and have not been corrected for convection; therefore, only RAE and Ames differences due to forward wind speed for their respec- tive data will be discussed. Figure 12 shows the change in level due to forward noise level speed for the data taken in the RAE wind tunnel. The high background allows in the lower frequency data for wind on at both 90° and 135 0 ; however, when the rest of the spectra is studied, the 90° data show a consistent reduction due to forward speed for the whole range of frequency. This difference is approximately 2 dB for each one-third-octave band frequency from 500 tiz to 50 kHz.
On the other hand, the 135° data showmore reduction in the lower and mid fre- quencies and almost no change for the high frequencies. The data from the Ames tests, shown in figure 13, show that for 90° at 160 Hz up to approximately 600 liz almost no changes in Level Cake place. In the middle and higherfrequencies, how- ever, a fairly consistent 2-3 dB difference is maintained, which is similar to that the observed for the RAE data. The data for 135° measured in the 40 by 80 shows same trend for the low frequency data as was observed for the RAE data and, also, the lack of level chance for the higher frequencies. An overlap of spectra for the 135° case where errors in position should be small due to the low velocity is shown in figure 14. The data from about 500 llz to 10 kliz are almost exact, but above 10 kHz the Ames data tend to be lower by 2-4 dB.
The zero wind speed data were also plotted as Strouhal number versus a AdB minus parameter formed by taking the difference between the one-third-octave band SPL reference 2. The normalized data are shown in the OASPL of the spectra as used in figures 15 and 16 for the cold flow case and in figure 17 for a comparison of hot flow data. Figure 15 shows the comparison for data at the 90 0 position. The solid line is a prediction for clean jet noise (ref. 2). The data trend is similar to the previous spectra comparisons with the exception that the Ames data and RAE data tend to agree at the higher Strouhal numbers, although neither the RAE nor Ames data match the predicted curve. The comparison with the clean jet prediction suggests that more than jet noise may be present at the lower frequencies for both the RAE and Ames data sets. Figure 16 shows the comparison at 140° (where the RAE and Ames data agree well.
with themselves as well as with the prediction) again suggesting that at the peak noise angle, the effect of the facilities was minimal. Figure 17 shows a comparison of hot flow data plotted in the same manner as figures 15 and 16. The hot flow data from the RAE and Ames tend to agree well one to the other, and deviate from the pre- diction at low and high Strouhal numbers. The data for figure 17 is for an angle of 90°.
The difference between the data sets and the prediction curve is due primarily to the original one-third-octave spectra from which the normalized curves were con- structed. For example, the RAE data in the original one-third-octave comparison at 90° showed much higher levels in the lower frequencies than the prediction. This results in the RA2 data having a higher OASPL than the prediction curve. When the differencing is made to construct the curves of figure 15, the RAE data have less differ nce between OASPL and SPL at the lower Strouhal number corresponding to low frequencies, and a larger difference at the high Strouhal numbers since the original comparisons were similar at the high Frequencies. The Ames data set shows similar bias when differenced, but in the Ames case the OASPL is lower due to the high fre- quency dropoff in the original one-third-octave spectra. The resulting comparison of the normalized data at 90° was somewhat fortuitous since the original comparison at 90 of the one-third-octave spectra was not good.
r- by 80 are compared to the In order to show directivity, the data from the 40 by summing only data from the RAE. OASPL have been calculated from the 40 by 80 data reverberant frequencies from 500 11z, discarding the lower, °wequencies because of the ttz is to sLay consistent with the PAL data where field influence. The cutoff at 500 a similar cutoff was made to avoid background contamination.
The comparison for nominal zero wind speed is shown in figure 18, The data have been straight line connected for clarity, The zero wind speed data show the RAE data above the 40 by 80 data by 1 to 1.5 dU up to 120°, and then both data sets match at the peak noise angles. The difference in level can be partially explained by the different jet exhaust temperatures, Others reasons for this difference may be analyses techniques, data reduction equipment, oid other processing differences between the two facilities, or the reflecting surfaces in the :forward are from both facilities.
The directivity for wind on is shown in figure 19. The RAE. data have been cor- rected from line of sight to convected angle to make the comparison. The data in this case are almost matched below 90 0 , and gave differences up to 1.5 dB in the aft angles with the RAD data being lower than the 40 by 80. An estimate of the OASPL differences, due to temperature difference between the Ames and RAE data, was made a correction of using .figures 11, 12, and 13 from reference 4. The figures give 90° and a zero dB correction -0.4 do to be added to the Ames data at angles less than a difference between the Ames and RAH data of about 1.5 dB.
above 90° which leaves These aft quadrant differ^,nces suggest that a larger change in level occurs in the RAE data due to Forward aVeed than happens in the 40 by 80. The correction for tem- 0' ^^11 ♦. n / a ure i it n .p n t> . s gy J bhC. e^ .L Clon ba n .'.1^T 9ii , mp r ov es t/G LFl {. LiL C: tlZpcsi CONCLUSIONS AND RECOMMENDATIONS The test of tilt. RAE 102-mm conical nozzle in the 40- by 80-root Wind Tunnel, and comparison of data with previous RAE results for the same nozzle tested in the RAE 24-Foot Wind Tunnel have shown good agreement between the data sets, especially at the peak noise angles. There are, however, some discrepancies between the data sets in the forward quadrant noise measurements :hat may be due to facility contamination, or data acquisition and reduction techniques. The following items are noted: v ithout wind speed show that the RAE data 1. The spectra comparisons for 90° up to about 800 liz. This higher level does are higher than expected at frequencies not show the usual peaks and troughs associated with reflected noise, although reflection may be responsible. The RAE spectrum is also rather flat through the mid-frequency region before falling off. The RAE spectrum, however, does come close to prediction in the mid- and high-frequency area. The 40 by 80 data for the same condition show fairlygood agreement with the prediction in the lower and mid-- frequencies, although some reflection is evident up to about; 500 Hz. At the higher frequencies, the 40 by 80 data tend to fall more rapidly than the prediction.
2. For the peak noise region, the spectra from the two facilities are nearly the same with little of the effects shown at 90° being present.
3. The forward speed effect on the jet noise appears similar for both facili- ties. There appears to be a slightly larger change due to forward speed for the RAE data, however, when comparisons are made using OASPL directivity. This may be due cd li r to slight differences in how the OASPL were constructed from the raw data sets. The quality of the data sets restricted the construction of OASM to frequencies above 500 Hz. Some residual reflection data above $00 Hz may tend to increase the value of OASPL calculated on this basis.
The tests were not as conclusive as expected, but did show that the facilities, although different in many respects, produced about the same results. The joint program may have been better addressed if the only variable was the facility in which the tests took place and not the differences in the jet rigs, the measurement technique, and data reduction, REFERENCES 1. Cocking, B. J., and Brycet W, D.; "An Investigation of the Noise of Cold Air Jets Under Simulated Flight Conditions," NGTE Report No, R 334, September 1974, Jaeck, C. L.: "Empirical Jet Noise Predictions for Single and Dual Flow ,lets 2.
With and Without Suppressor Nozzles," Volume I, Document No. D6-42929-1, The Boeing Company, 1976.
3. Hoglund, Lief E.: "Flight Effects for Some, Model RAS Nozzles as Measured in the 40_ by 80-Foot Wind Tunnel ) " NASA Contractor Report (unpublished) .
Cocking, E. J.: "The Effect of Temperature on Subsonic Jet Noise," Reports and 4.
Memoranda No. 3771, tiny 1974, NGTE.
1,0
r
TABLE 1.- NOZZLE TEST SCHEDULE
V jet (nom) , VT (note) , PR Temperature m/S m/s 102-mm conical nozzle without wing 1.6 300 0 Cold .
1.6 300 15 Cold 1.6 300 30 Cold 1.73 0 Hot 1.73 480 30 1.73 480 60 1,73 90 2.02 550 0 2.02 550 30 2.02 550 2.02 550 90 2.45 600 0 2.45 600 2.45 6C 2.45 600 102-mm conical nozzle with wing 1.6 300 0 Cold 1.6 300 15 Cold 1.6 300 30 Cold ,f ,y rf a r 77 ^ ORILONAL PAGE BLACK AND WHITE PHOTOGRAPH Figure 1. KAE 24-Foot Wind Tunnel.
r t 730 mm 200 m"t R CASCADE CORNE mm END BLANK WOODEN NOSE r,,«•'" GAUZE WOODEN ' FAIRING I I I I I REFERENCE I i I PITOT PRESSURE I I lil"T COLD AIR SUPPLY Figure 2.- RAE Jet test rig, s ll^ ORIGINAL PAGE BLACK AND WHITE n"^TOGPJV" figure i.- 40- by 80-Foot Wind 'funnel test section.
WIND TUNNEL NOISE 102 mm CONICAL NOZZLE 0 - 90°, 1.45 m SL, STANDARD TEST RIG E ffiEE
V •6 90
V j = 8
V T -62
u: V T R ..I
a
^'^ VT- 16
60 h
50 L -
10 2 103 104 105 FREQUENCY, Hz Figure 4.= Background noise in the 40- by 80-Foot Wind Tunnel..
AIRFOIL STREAMLINING
VJET"f C
TRAVERSING
VTUNNEL
MICROPHONE BU,itv.R 76 mm FOAM Figure 5.- S-duct jet test rig !- t^.
ORIGINAL PAGE SLACK AND WHITE PHOTOGRAPH Fi6uru 6.- Wing rig mount in the 40- by 80-Foot Wind Tunnel.
k FINAL PAC,L bLACK AND WHITE PHOTOGRAPH the treated enclosure.
in Figure I.- Viper compressor 254 m 1 Q2 mm 730 mm I- -nun conic nozzle.
} Figure 8.- RAE 102 O WITH WING RIG Q WITHOUT WING (S-DUCT) 102 snm CONICAL NOZZLE
Vi M 290 m/sac
40 X 80 DATA 2,96 m SIDELINE
VT" Ca
co V .1N 100
PREDICTED
. • . •
d(0 90
co SPECTRUM
> ^ 8g
Q„ c 80 U u: O W ,- .104 oc 70 0 2 105 FREQUENCY, Hz Figure 9.- Rig comparison -- effect of wing rig on 40- by 80-Foot wind Tunnel, data.
0 DATA FROM RAE p DATA FROM 40 X 80 ..._,JET NOISE PREDICTION m 'o
a100 102 mm CONICAL NOZZLE
o q QOU ^ 0 Q a 90 p
uI a
G '
c 80
^^A
ca u: O W L M OC I 102 103 1 104 105 FREQUENCY, Hz Figure 10.- Comparison of data for 90° approximately zero wind speed.
O DATA FROM RAE * DATA FROM 40 X 80 m 102 mm CONICAL NOZZLE • y
ifl
^ Z s CO N
w O
g 80 LL: LU M X r 102 103 104 105 FREQUENCY, Hz Figure 11.- Comparison of data for 135° approximately zero wind speed.
"'--"— VT ^^ 3 m/sec 102 mm CONICAL NOZZLE O VTR ^^ 30 m/sec DATA FROM RAE 24 ft 3.0 m SIDELINE
Vi E^! 300 m /sec
WING RIG
ODa'd00^
o °O pOc)oe 0 c.O` d, deg
n^ So v60 70 E c 70 M N _ wo 0 01 0040000000000 C) ° U.
Uj cc 80 M 70l a Z 100
W
— ^00000000000d
a'
U
M 80 0135
70' ' ' '
10 2 103 104 105
Ft'1EQUENCY, Hz
Figure 12.—Comparison of data with tunnel on and tunnel off — RAE 24-Foot Wind Tunnel data.
a
102 mm CONICAL NOZZLE
.---VT-0
DATA FROM 40 X 80
O VT ^ 29 m/sec
2.96 m SIDELINE
WING RIG
Vi n -^ 290 m/sec
.
EOO
80 0, dog
a 0(
b +% 60
,0
' 90
^bO) C) pU(Op O cc ` ^T O at^, d
o^^
N p 90 Z r Q > 100
^Op^O
v /O : (p O C7tJ ^^ r0 OaO%
/00 O
O aaC^ 70 - L 10 2 103 104 FREQUENCY, Hz Figure 13.- Comparison of data with tunnel on and tunnel off - 40- by 80-Foot Wind Tunnel data.
102 mm CONICAL NOZZLE
135' POSITION
WIND SPEED 30 m/sec
COLD FLOW DATA
O RAE 24 ft WIND TUNNEL
co AMES 40X80 WIND TUNNEL
E
C
Q90
Rn NN Q 80
Q W
r^ cc 70
102 10 3 104 100
FREQUENCY, Hz
Figure 14.— Comparison of data at 135° - wind speed -30 m/sec.
NORMALIZED SUBSONIC CLEAN JET NOISE, T = 294 K
AMES 40 X 80 O 2,96m, 102 mm CONICAL NOZZLE, V j = 292 m/sec, T - 360 K
0 RAC 24 ft O 3.00 m, 102 mm CONICAL NOZZLE, = 292 m/sec, T 304 K
Vi
ZERO WIND SPEED
a
t) 90
p-10
e
O •
z -20
•^0
m 0s•
W -30
H
U
M -40 f I
°° -50
4 .01
.1 1.0 10.0 100.0
fD/V
m Figure 15.- Normalized jet noise spectra at 0 90 0 cold jet.
-----NORMALIZED SUBSONIC CLEAN JET NOISE SPECTRA, T
w 294`K
0 +1
AMES 40 X 80 0 2,96m, 102 mm CONICAL NOZZLE, Vj 292 m/sec, T 360 K
w
0 3.00 m,
r RAE 102 mm CONICAL NOZZLE, Vj = 292 m/sec, T = 304'K
, 0 - 140°
. j —10
Q
O
J 0
'L -20
C
e^
w -3©
a
o -ao
r
..50 .01 .1 1.0 1010 100,0 fD/V
a
Figure 16.- Norinulized j et noise spectra at 0 140° cold jet, NORIVIALiZE-Ut S UBSONIC CLEAN JET NOISE SPECTRA,
Vj ,-- 457 m/sec, T - 795 K
AMES 40 X 80 ^102 mm CONICAL NOZZLE, Vj = 489 m/sec, T = 820 K
V 102 mm CONICAL NOZZLE, Vj = 550 m/sec, T = 820 K
RAE ®
65 min CONICAL NOZZLE, Vj - 355 m/sea, T = 880 K
m 0
M
a
c n -10
Q
O
v i -^tt
i
Z Q w -30 Q o -40 Q -50
.0
Fi a V 0 RAE 3 ,0 m SIDELINE, i 295 m/soc, Tj L 340 K L1 AMES 2.96 m SIDELINE, Vi a 292 m/sec, Tj L , 304 K WIND TUNNEL VELOCIT y ot0 m 'v a O 1 1 1 1 1 1 .1 1 I 1 ..1 90, 1'i.
20 30 40 50 60 70 80 90 MI . 110 120 130 140 150 160 O-REFERENCE *: 1 ^ C TO INLET Figure 18,- OASPL directivity comparison -- wind speed approximately zero.
® RAE 3.0 m SIDELINE, Vi = 296 m/sec, Tj ^5 340 K 0 AMES 2.96 m SIDELINE, V j = 293 m/sec, Tj ^ 304 K C o 102 In J
a
a
O
1 1 1 1 1 1 1 1 1 1 1 1 1 94 1 T 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 O•REFERENCE ANGLE TO INLET ` Figure 19.- OASPL directivity comparison — wind speed »30 m/sec.