1、Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-NASA Technical Memorandum X-71763Interim Prediction Method forFan and Compressor Source NoiseMarcus F. H
2、eidmannLewis Research CenterCleveland, OhioNILSIXNational Aeronauticsand Space AdministrationScientific and TechnicalInformation Branch1979Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-Provided by IHSNot for ResaleNo reproduction or networking perm
3、itted without license from IHS-,-,-SUMMARYA prediction method is presented for interim use by NASA for evaluating and as-sessing the noise generated by fans and compressors of turbojet and turbofan aircraftengines. The method was formulated in partial support of the NASA Aircraft NoisePrediction Pro
4、gram (ANOPP). Spectral distributions of the sound pressure levelsreferenced to 20 micronewtons per square meter at a 1-meter radius in a free fieldare predicted as a function of the polar angle about the fan or compressor axis. Thetotal noise levels are obtained by spectraUy summing the predicted le
5、vels of broadband,discrete-tone, and combination-tone noise components.The interim prediction method is a modification of a method previously developedby other investigators. The modifications are based on a partial analysis of data fromfull-scale, single-stage fan tests performed at NASA Lewis Rese
6、arch Center. A dis-cussion of the modifications and a detailed procedure for the prediction are presented.Comparisons of the predicted and measured noise of the NASA fans are presented anddiscussed. Some major problem areas and research requirements are identified.INTRODUCTIONNASA is engaged in a va
7、riety of programs directed toward reducing the noise as-sociated with aircraft operation. A significant source of the noise produced by aircraftis that generated by fans and compressors for aircraft engines. Both contractual andin-house studies are being supported by NASA to explore and develop quie
8、ter fans andcompressors for a variety of engine cycles and types of aircraft. One requirementfrequently arising from such studies is the ability to readily establish and assess thenoise performance of fans and compressors. The source noise of new or existing fan-compressor designs or applications mu
9、st often be evaluated relative to previous expe-rience, the specification of acoustic treatment, and the impact on community noiselevels. What is needed is a specific and comprehensive method of predicting thesource noise of any fan or compressor to a quantitative precision which reflects thestate-o
10、f-the-art. NASAs objective is to develop and continuously update such a pre-diction method. There is an immediate need, however, to specify a prediction methodwhich can be used until comprehensive prediction methods can be fully explored, de-veloped, and verified. The purpose of this report is to sp
11、ecify an interim predictionProvided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-method for fan and compressor source nois_ in terms of its free-field spectral anddirectivity properties.An interim prediction method is specified in response to the immediate
12、 need forpredicting fan-compressor noise as a component of the total aircraft noise for theNASA Aircraft Noise Prediction Program (ANOPP). This NASA Langley ResearchCenter program is being developed jointly by various NASA centers with help from in-dustry representatives. The purpose of ANOPP is to
13、develop a comprehensive methodof predicting noise footprints for single or multiple event air-flights which reflects thestate-of-the-art and is useful in exploring t_clmological advances to economically re-duce community noise levels.This report first discusses the approach used to formulate the int
14、erim predictionmethod and the scope of this study with regard to the applicable literature. The methodis then introduced by a general description of the prediction procedure where the appli,cations and the necessary design and operating parameters of a fan or compressor arealso specified. The noise
15、properties of fans and compressors used in the predictionmethod are then specified within a limited framework of background material. A spe-cific procedure for predicting free-field noise patterns follows. Finally, comparisonsof measured and predicted results are made, and some remarks are presented
16、 aboutproblem areas and future research requirements.BF1, F2, etc.ffbLL CM TMTRSYMBOLSnumber of rotor bladesfunctions as given by appropriate curvesfrequency, Hzblade passage frequency, B_, Hzsound pressure level correction, dB (referenced to 20/_N/m 2 at 1-mradius)characteristic sound pressure leve
17、l, dB (referenced to 20 #N/m 2 at 1-mradiusnormalized peak noise level, dB (referenced to 20 _N/m 2 at 1-m radiusrotor tip or wheel speed Mach numberrotor tip relative inlet Mach numberMTR at fan design point2Provided by IHSNot for ResaleNo reproduction or networking permitted without license from I
18、HS-,-,-oPNLPRPWLRSSSHPSPL(f)TATAT oV5T/e0mass flow rate passing through fan or compressorreference value of m (0.453 kg/see, 1 ibm/see)perceived noise level, dBfan stage total pressure ratiosound power level, dB (referenced to 10 -13 Wrotor-stator spacing factor (see fig. 14(a)rotor shaft horsepower
19、spectrum level as function of frequency, dBtotal temperature at fan face, K (OR)total temperature rise across a fan or compressor stage, AT/T =reference value of AT (0. 555 K, 1 R)number of stator bladescutoff factor (see eq. (3)fan stage adiabatic efficiencydireetivity or polar angle reference to i
20、nlet axiswheel speed, rev/seeAPPROACH AND SCOPEThe approach used herein to formulate an interim prediction method was to modifyan existing method. The prediction method adopted for this purpose was the fan andcompressor portion of a more comprehensive prediction program recently developedby the Boei
21、ng Company under contract with the NASA Ames Research Center. Thisprediction method is explicitly specified in reference 1. The modifications to theBoeing-Ames prediction method made at this time are entirely related to correlationsand interpretations of the acoustic data from the full-scale fan tes
22、ts performed atNASA Lewis (reported in refs. 2 to 4). Some of the prediction curves and equationsspecified in the interim prediction method correspond exactly to those appearing in theBoeing-Ames method. In particular, the extension of single-stage fan prediction tomultistage fans and compressors is
23、 used as specified.Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-The Boeing-Ames method was selected for modification because it is a thoroughlydocumented method and is addressed to the total problem of providing far-field direc-tivity and spectral
24、 properties from multiple sources, features considered essential forthe interim prediction method. The choice of the method was from among a variety ofempirically and theoretically based prediction methods. Most of these other methodsdo not provide complete far-field properties or are restricted to
25、a single noise sourceor noise component. For example, the frequently referenced correlation of Smith andHouse (ref. 5) does not provide far-field directivity. Only two well-documented pre-diction methods in addition to the Boeing-Ames method appeared to satisfy the needs ofthis study. Although these
26、 methods (refs. 6 and 7) were formulated at an earlier date,they prescribe prediction procedures in many ways similar to the Boeing-Ames method.The Boeing-Ames method was adopted because it used a data base acquired from oper-ational aircraft rather than from scale model static tests and because it
27、more nearlysatisfied the immediate needs of NASA. Some additional prediction methods developedby the aeronautics industries which may satisfy the needs of NASA are referred to inthe literature (refs. 8 to 11. Many of the details of these methods, however, have notbeen disclosed.A survey and evaluati
28、on of the reported prediction methods and techniques are be-yond the intended scope of this study. In this regard, it should be pointed out that noreported prediction method has received a comprehensive evaluation against availabledata. In general, these methods have been examined only for the manne
29、r in which theyfit the data base from which they were obtained. Another purpose of the interim pre-diction method is that it is a vehicle by which models and techniques of other predictionmethods can be evaluated. Such evaluations are considered to be best accomplishedwithin the matrix of a multiple
30、 source - multiple component prediction method ratherthan by examining the isolated behavior of a specific source or component.The main objective of this study is to provide a precise and all-encompassing pre-diction method which assimilates all the available technology. The currently used pre-dicti
31、on methods are basically empirical with only a phenomenological application oftheory. A more quantitative application of theoretical analysis appears necessary forprecise and all-encompassing predictions. The theoretical literature on noise genera-tion by turbomachinery is extensive and growing. A r
32、eview of this literature and po-tential applications to the prediction method are not covered in this initial study.One aspect of the theoretical literature may be particularly significant with regardto upgrading the prediction method. This involves the coupling between a noise sourceand its far-fie
33、ld noise pattern. The theoretical coupling between a fan or compressornoise source and the far-field noise pattern involves wave propagation through ductsand duct terminations so that the free-field pattern depends on both the source and theduct properties. From an ideal theoretical standpoint, the
34、source noise should be4Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-characterized by the amplitudes of the propagating modes entering the duct rather thanby the far-field noise patterns as currently used in prediction methods. A major prob-lem in
35、this regard is that the specification of the propagating modes depends on exactdescriptions of the complex unsteady flow patterns within a fan or compressor. Un-fortunately, substantiated techniques which relate free-field problems to modal ampli-tudes for operational fans do not now exist. However,
36、 their development could signifi-cantly affect the method for predicting the source noise or the approach to any upgrad-ing effort as now envisioned.GENERAL DESCRIPTION OF PREDICTION METHODAn interim procedure for predicting the source noise for fan and compressorstages was formulated based on the B
37、oeing-Ames method described in reference 1.The procedure predicts the one-third octave band spectral intensities of the free-fieldnoise pattern. The noise is assumed to be propagating from a “virtual source“ at apoint defined by the plane and axis of a fan compressor. The intensities are given asthe
38、 sound pressure levels (in dB referenced to 20 ttN/m 2) at a 1-meter radius and aredevoid of any atmospheric attenuation or ground effects. The predicted free-field radi-ation varies with the polar angle about the fan or compressor axis and is assumed to besymmetric about this axis. The prediction m
39、ethod is applicable to turbojet compres-sors and to single- and two-stage turbofans with and without inlet guide vanes. Thefans or compressors are assumed to be representative of good aerodynamic designwith the operating point near the normal operating line and not above the design point.It is assum
40、ed that the inlet and discharge duct configurations are such that they neithercause nor attenuate the generated noise - that is, short ducts with no blow-in doors,aerodynamically poor obstructions, or acoustic treatment.The predicted free-field radiation patterns consist of a composite of the follow
41、ingseparately predicted noise components:(1) Noise emitted from the fan or compressor inlet duct(a) Broadband noise(b) Discrete-tone noise(c) Combination-tone noise(2) Noise emitted from the fan discharge duct(a) Broadband noise(b) Discrete-tone noiseThe procedure involves predicting the spectrum sh
42、ape, spectrum level, and free-field directivity for each of the noise components. The component spectra at any polarangle are then combined on an energy basis to form a single spectrum. The procedure5Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-is
43、 specifically directed toward a single-stage fan. For two-stage fans, each stage istreated as an independent source and the sound energy produced by each stage is com-bined. No correction is made for blade row attenuation. In the case of a turbojet en-gine, the noise from the first compressor stage
44、is assumed to be representative of thefar-field noise. These multistage fan and compressor procedures are those of theBoeing-Ames method and are not reexamined in this study.The spectrum shape of each component is basically fixed, and the spectral predic-tions primarily involve positioning the spect
45、rum shape within the frequency domain.This positioning is accomplished in all cases by referencing the spectral frequenciesto the blade passage tone frequency. The blade passage tone frequency, therefore, isone of the basic parameters required in the prediction procedure.Predictions of the spectrum
46、levels of each of the components depend on the designand performance parameters of the fans and compressors. A spectrum level is pre-dicted for a reference polar angle and the complete far-field pattern is obtained fromspecified directivity corrections. Each component, therefore, is characterized by
47、 afixed spectral distribution which varies in level with the polar angle. Since the direc-tivity corrections differ for each noise component, they give the result of a varyingspectra as well as level with polar angle when they are combined.Four parameters must be specified to predict the basic spect
48、rum levels: the massflow rate and total temperature rise associated with a fan or compressor stage and thedesign and operating point values of the rotor tip relative inlet Mach number. Thesebasic levels are then corrected or adjusted. One correction depends on the presenceor absence of inlet guide v
49、anes. Such vanes are always present for the second stageof a conventional two-stage fan. A correction is also applied for rotor-stator spacingeffects. For a rotor with both inlet and exit stators the correction depends solely onthe most closely spaced stator.Two additional corrections are applied to the spectrum levels or shapes.
