environmental technical manual on the use of procedures in the

ICAO COMMITTEE ON AVIATION ENVIRONMENTAL PROTECTION

ENVIRONMENTAL TECHNICAL MANUAL ON THE

USE OF PROCEDURES IN THE NOISE

CERTIFICATION OF AIRCRAFT

STEERING GROUP APPROVED REVISION

SGAR/7September 2000

jskaleck

7/15/03 AC 36-4C Appendix 1

jskaleck

AC 36-4C 7/15/03 Appendix 1

i

ICAO ENVIRONMENTAL TECHNICAL MANUAL ON THE USE OFPROCEDURES IN THE NOISE CERTIFICATION OF AIRCRAFT

This Steering Group approved revision includes material which has been approved by the relevant WorkingGroups of the ICAO Committee on Aviation Environmental Protection (CAEP). Its purpose is to make availablenew information to certificating authorities, noise certification applicants and other interested parties as soon asit has been agreed by the working groups, therefore eliminating the delay which would otherwise occur shouldits publication be limited only to post CAEP meetings. Prior to these meetings the then current approvedrevision will be reviewed for submission to CAEP for formal endorsement and subsequent publication by ICAO.

CONTENTS PAGE

NOMENCLATURE iv

SECTION 1 GENERAL 1

1.1 Purpose 11.2 Framework 11.3 Incorporation of equivalent procedures into the noise compliance

demonstration plan1

1.4 Changes to the noise certification levels for derived versions 21.5 Re-certification 3

SECTION 2 EQUIVALENT PROCEDURES FOR SUBSONIC JET AEROPLANES 5

2.1 Flight test procedures 52.1.1 Flight path intercept procedures 52.1.2 Generalised flight test procedures 5

2.1.2.1 Derivation of noise, power, distance data 52.1.2.2 Procedures for the determination of changes in noise

levels7

2.1.3 The determination of lateral noise certification levels 72.1.4 Take-off flyover noise levels with power cut-back 82.1.5 Measurements at non-reference points 82.1.6 Atmospheric Test Conditions 92.1.7 Reference Approach Speed 9

2.2 Analytical procedures 92.2.1 Flyover noise levels with power cut-back 92.2.2 Equivalent procedures based on analytical methods 10

2.3 Static tests and projections to flight noise levels 112.3.1 General 112.3.2 Limitation on the projection of static to flight data 112.3.3 Static engine tests 12

2.3.3.1 General 122.3.3.2 Test site requirements 132.3.3.3 Engine inlet bell mouth 132.3.3.4 Inflow control devices 13

2.3.3.4.3 ICD Calibration 132.3.3.5 Measurement and analysis 142.3.3.6 Microphone locations 142.3.3.7 Acoustic shadowing 152.3.3.8 Engine power test conditions 162.3.3.9 Data system compatibility 162.3.3.10 Data acquisition, analysis and normalisation 16

2.3.4 Projection of static engine data to aeroplane flight conditions 162.3.4.1 General 162.3.4.2 Normalisation to reference conditions 172.3.4.3 Separation into broadband and tone noise 18

jskaleck

jskaleck


ii

2.3.4.4 Separation into contributing noise sources 182.3.4.5 Noise source position effects 192.3.4.6 Engine flight conditions 202.3.4.7 Noise source motion effects 20

2.3.4.7.1 External jet noise 202.3.4.7.2 Noise sources other than jet noise 20

2.3.4.8 Aeroplane configuration effects 212.3.4.9 Airframe noise 212.3.4.10 Aeroplane flight path considerations 222.3.4.11 Total noise spectra 222.3.4.12 EPNL computations 222.3.4.13 Changes to noise levels 22

SECTION 3 EQUIVALENT PROCEDURES FOR PROPELLER DRIVEN AEROPLANES OVER9000 kg

25

3.1 Flight test procedures 253.1.1 Flight path intercept procedures 253.1.2 Generalised flight test procedures 253.1.3 Determination of the lateral noise certification level 253.1.4 Measurements at non-reference points 263.1.5 Reference Approach Speed 26

3.2 Analytical procedures 263.3 Ground static testing procedures 27

3.3.1 General 273.3.2 Guidance on test site characteristics 273.3.3 Static tests on the gas generator 27

SECTION 4 EQUIVALENT PROCEDURES FOR PROPELLER DRIVEN AEROPLANES NOTEXCEEDING 9000 kg

29

4.1 Source noise adjustments 294.1.1 Fixed pitch propellers 294.1.2 Variable pitch propellers 29

4.2 Take-off test and reference procedures 30

SECTION 5 EQUIVALENT PROCEDURES FOR HELICOPTERS 31

5.1 Flight Test Procedures 315.1.1 Noise Certification Guidance 31

5.1.1.1 Helicopter test window for zero adjustment foratmospheric attenuation

31

5.1.1.2 Helicopter test speed 315.1.1.3 Test speed for light helicopters 325.1.1.4 Helicopter test mass 325.1.1.5 Helicopter approach 325.1.1.6 Helicopter flight path tracking 33

5.1.1.6.1 Radar or microwave tracking system 335.1.1.6.2 Kine-theodolite system 345.1.1.6.3 Radar/theodolite triangulation 345.1.1.6.4 Photographic scaling 34

5.1.1.7 Atmospheric test conditions 355.1.1.8 Procedure for determination of source noise correction 35

5.1.2 On board flight data acquisition 365.1.2.6 Magnetic tape recording 375.1.2.7 Automatic still photographic recording 375.1.2.8 Cine recording 375.1.2.9 Video recording 375.1.2.10 Time synchronisation of recorded data 37

5.1.3 Procedures for the determination of changes in noise levels 37

jskaleck


jskaleck

iii

5.1.3.1 Modifications or upgrades involving aerodynamic dragchanges

37

5.1.4 Temperature and relative humidity measurements 385.1.4.5 Testing of light helicopters outside Chapter 11

temperature and humidity limits39

5.1.5 Anomalous Test Conditions 39

SECTION 6 EVALUATION METHODS 41

6.1 Spectral irregularities 416.2 Ambient noise levels 416.3 Establishment and extension of data bases 416.4 Test environment corrections 426.5 Inertial navigation system for aeroplane flight path measure 426.6 Computation of EPNL by the integrated method of adjustment 42

6.6.3 Test aircraft position 426.6.4 Sound propagation times and sound emission angles 436.6.5 Aircraft reference flight path 436.6.6 Time interval computation 446.6.7 Adjusted EPNL 44

6.7 Calculation of the speed of sound 45

SECTION 7 MEASUREMENT AND ANALYSIS EQUIPMENT 47

7.1 Definitions 477.1.1 Measurement system 477.1.2 Microphone system 477.1.3 Sound incidence angle 477.1.4 Reference direction 477.1.5 Free-field sensitivity of a microphone system 477.1.6 Free-field sensitivity level of a microphone system 477.1.7 Time-average band sound pressure level 487.1.8 Level range 487.1.9 Calibration sound pressure level 487.1.10 Reference level range 487.1.11 Calibration check frequency 487.1.12 Level difference 487.1.13 Reference level difference 487.1.14 Level non-linearity 487.1.15 Linear operating range 487.1.16 Windscreen insertion loss 48

7.2 Reference Environmental Conditions 487.3 General 487.4 Windscreen 497.5 Microphone System 497.6 Recording and Reproducing Systems 507.7 Analysis Systems 517.8 Calibration Systems 537.9 Calibration and Checking of System 537.10 Adjustments for Ambient Noise 54

SECTION 8: CONTROL OF NOISE CERTIFICATION COMPUTER PROGRAMMESOFTWARE AND DOCUMENTATION RELATED TO STATIC-TO-FLIGHTPROJECTION PROCESSES

55

8.1 General 558.2 Software control plan 55

8.2.1 Configuration index 558.2.2 Software control plan 558.2.3 Design description 55

jskaleck


iv

8.2.4 Verification process 558.3 Applicability 55

SECTION 9: GUIDELINES FOR NOISE CERTIFICATION OF TILTROTOR AIRCRAFT 57

REFERENCES 59

FIGURES 61

APPENDIX 1 Calculation of confidence intervals 75

APPENDIX 2 Identification of spectral irregularities 89

APPENDIX 3 A procedure for removing the effects of ambient noise levels from aeroplanenoise data

91

APPENDIX 4 Reference tables and figures used in the manual calculation of EffectivePerceived Noise Level

93

APPENDIX 5 Worked example of calculation of reference flyover height and referenceconditions for source noise adjustments for certification of light propellerdriven aeroplanes to Chapter 10

99

APPENDIX 6 Noise data corrections for tests at high altitude test sites 103

APPENDIX 7 Technical background information on the guidelines for the noisecertification of tiltrotor aircraft

105

APPENDIX 8 Reassessment criteria for the re-certification of an aeroplane from ICAOAnnex 16, Volume 1, Chapter 3 to a more stringent standard

109

jskaleck


v

NOMENCLATURE

Symbols and abbreviations employed in this manual are consistent with those contained in ICAO Annex 16,Volume 1, Third Edition, July 1993.

Symbol Unit Description

c m/s Speed of soundCI dB 90 per cent Confidence interval in decibel units relevant to the calculation being

made.D m Jet nozzle diameter based on total nozzle exit area.EPNL EPNdB Effective Perceived Noise LevelF N Engine net thrustf Hz 1/3-octave band centre frequencyICD - Inflow control deviceK - ConstantL dBA 'A' - weighted sound pressure levelM - Mach numberMH - Propeller helical tip Mach numberMAP in. Hg Manifold air pressureNP rpm Propeller rotational speedN1 rpm Low pressure rotor speed of turbine enginesOASPL dB Overall Sound Pressure LevelPNL PNdB Perceived Noise LevelPNLT TPNdB Tone Corrected Perceived Noise LevelPNLTM TPNdB Maximum Tone Corrected Perceived Noise LevelS - Strouhal number (fD/Vj)SHP kW Shaft horse powerSPL dB Sound pressure level based on a reference of 20 µPaTCL °C Air temperature at engine centreline heightTMIC °C Air temperature at the ground plane microphone heightVj m/sec Jet velocity for complete isentropic expansion to ambient pressureV m/sec Aircraft airspeedVy m/sec Aircraft best rate of climb speedWCL Km/h Average wind speed at engine centreline heightx m Distance downstream from nozzle exitδamb - Ratio of absolute static pressure of the ambient air at the height of the aeroplane to

ISA air pressure at mean sea level (i.e. 101.325 kPa)θt2 - Ratio of absolute static temperature of the air at the height of the aeroplane to the

absolute temperature of the air at sea level for ISA conditions (i.e. 288.15 °K)µ - Engine power related parameter, or mean value see Appendix 1λ degrees Angle between the flight path in the direction of flight and a straight line

connecting the aeroplane and the microphone at the time of sound emissionσ - Ratio of atmospheric air density at altitude to that at sea level for ISA conditions

Suffices

flt Quantity related to flight conditionsmax Maximum valueref Quantity related to reference conditionsstatic Quantity related to static conditionstest Quantity related to test conditionsDOP Doppler related quantity

jskaleck


vi

Abbreviations

ESDU Engineering Sciences Data UnitISA International Standard AtmosphereNPD Noise-power-distanceSAE AIR Society of Automotive Engineers - Aerospace Information ReportSAE ARP Society of Automotive Engineers - Aerospace Recommended Practice

Notes: Where log is used in this document it denotes logarithm to the base of 10.

jskaleck


CAEP Steering Group Environmental Technical Manual on the use ofApproved Revision 7 Procedures in the Noise Certification of Aircraft

1

SECTION 1 - GENERAL

1.1 PURPOSE

1.1.1 The aim of this manual is topromote uniformity of implementation of thetechnical procedures of Annex 16, Volume 1,and to provide guidance such that allcertificating authorities can apply the samedegree of stringency and the same criteria foracceptance in approving applications for theuse of equivalent procedures.

1.1.2 This manual provides guidance inthe wider application of equivalent proceduresthat have been accepted as a technical meansfor demonstrating compliance with the noisecertification requirements of Annex 16,Volume 1. Such procedures are referred to inAnnex 16, Volume 1, but are not dealt with inthe same detail as in the Appendices to theAnnex which describe the noise evaluationmethods for compliance with the relevantChapters.

1.1.3 Annex 16, Volume 1, proceduresmust be used unless an equivalent procedure isapproved by the certificating authority.Equivalent procedures should not beconsidered as limited only to those describedherein, as this manual will be expanded as newprocedures are developed.

1.1.4 For the purposes of this manual anequivalent procedure is a test or analysisprocedure which, while differing from onespecified in Annex 16, Volume 1, in thetechnical judgement of the certificatingauthority, yields effectively the same noiselevels as the specified procedure.

1.1.5 References to Annex 16, Volume 1,relate to the Amendment 6 thereof.

1.2 FRAMEWORK

Equivalent procedures fall into two broadcategories; those which are generally applicableand those which are applicable to a particularaircraft type. For example, some equivalenciesdealing with measurement equipment may be usedfor all types of aircraft, but a given test proceduremay only be appropriate for turbojet poweredaeroplanes, and not to turboprop poweredaeroplanes. Consequently this manual is framed toprovide information on equivalent proceduresapplicable to the types of aircraft covered by Annex

16, Volume 1, i.e. jet powered, propeller drivenheavy and light aeroplanes and helicopters.Equivalent procedures applicable to each aircrafttype are identified in separate sections. Eachsection covers, in the main, flight testequivalencies, the use of analytical procedures andequivalencies in evaluation procedures.

1.3 INCORPORATION OF EQUIVALENT

PROCEDURES INTO THE NOISE COMPLIANCE

DEMONSTRATION PLAN

1.3.1 Prior to undertaking a noisecertification demonstration, the applicant isnormally required to submit to the certificatingauthority a noise compliance demonstrationplan. This plan contains the method by whichthe applicant proposes to show compliancewith the noise certification requirements.Approval of this plan and the proposed use ofany equivalent procedure remains with thecertificating authority. The procedures in thismanual are grouped for specific applications.The determination of equivalency for anyprocedure or group of procedures must bebased upon the consideration of all pertinentfacts relating to the application for a certificate.

1.3.2 Use of equivalent procedures maybe requested by certificate applicants for manyreasons, such as:

a) to make use of previously acquiredcertification test data for theaeroplane type;

b) to permit and encourage more reliabledemonstration of small noise leveldifferences among derived versions ofan aeroplane type; and

c) to minimise the costs ofdemonstrating compliance with therequirements of Annex 16, Volume 1,by keeping aircraft test time, airfieldusage, and equipment and personnelcosts to a minimum.

1.3.3 The material included in this manualis for technical guidance only. The use of pastexamples of approved equivalencies does notimply that these equivalencies are the onlyacceptable ones, neither does their presentationimply any form of limitation of their

jskaleck

jskaleck

jskaleck


Environmental Technical Manual on the use of CAEP Steering GroupProcedures in the Noise Certification of Aircraft Approved Revision 7

2

application, nor does it imply commitment tofurther use of these equivalencies.

1.4 CHANGES TO THE NOISE CERTIFICATION

LEVELS FOR DERIVED VERSIONS

1.4.1 Many of the equivalent proceduresgiven in this manual relate to derived versions,where the procedure used yields theinformation needed to obtain the noisecertification levels of the derived version byadjustment of the noise levels of the "flightdatum" aircraft (i.e. the most appropriateaircraft for which the noise levels weremeasured during an approved Annex 16,Volume 1, flight test demonstration).

1.4.2 The physical differences betweenthe "flight datum" aircraft and the derivedversion can take many forms, for example, anincreased take-off weight, an increased enginethrust, changes to the powerplant or propelleror rotor types, etc. Some of these will alter thedistance between the aircraft and the noisecertification reference points, others the noisesource characteristics. Procedures used in thedetermination of the noise certification levelsof the derived versions will therefore dependupon the change to the aircraft beingconsidered. However, where several similarchanges are being made, for example,introduction of engines from differentmanufacturers, the procedures used to obtainthe noise certification levels of each derivativeaircraft should be followed in identical fashion.

1.4.3 Aircraft/engine model designchanges and airframe/ engine performancechanges may result in very small changes inaircraft noise levels that are not acousticallysignificant. These are referred to as no-acoustical changes. For this manual, no-acoustical changes, which do not result inmodification of an aircraft’s certificated noiselevels, are defined as:

a) Changes in aircraft noise levelsapproved by certificating authoritiesas not exceeding 0.1 dB at any noisemeasurement point and which anapplicant does not track;

b) Cumulative changes in aircraft noiselevels approved by certificatingauthorities whose sum is greater than0.1 dB but not more than 0.3 dB atany noise measurement point and for

which an applicant has an approvedtracking procedure.

1.4.4 Noise certification approval hasbeen given for a tracking procedure of the typeidentified in 1.4.3b, based upon the followingcriteria:

a) Certification applicant ownership ofthe noise certification database andtracking process based upon anaircraft/engine model basis;

b) When 0.3 dB cumulative is exceeded,compliance with Annex 16 Volume 1requirements is required. The aircraftcertification noise levels may not bebased upon summation of no-acoustical change noise increments;

c) Noise level decreases should not beincluded in the tracking processunless the type design change will beretrofitted to all aircraft in service andincluded on newly produced aircraft;

d) Aircraft/engine design changesresulting in noise level increasesshould be included in the trackingprocess regardless of the extent ofretrofit to aircraft in service;

e) Tracking of an aircraft /engine modelshould, in addition to engine designchanges, include airframe, andperformance changes;

f) Tracked noise increments should bedetermined on the basis of the mostnoise sensitive condition and beapplied to all configurations of theaircraft/engine model;

g) The tracking should be revised toaccount for a tracked design changeincrement that is no longer applicable;

h) Changes should be tracked to twodecimal places (e.g. 0.01dB). Round-off shall not be considered in judginga no-acoustical change (e.g. 0.29dB =no-acoustical change ; 0.30dB = no-acoustical change ; 0.31dB =acoustical change);

i) An applicant should maintain formaldocumentation of all no-acousticalchanges approved under a trackingprocess for an airframe/engine model.

jskaleck



3

The tracking list will be reproduced ineach noise certification dossierdemonstration; and

j) Due to applicable dates for Chaptersconcerning helicopters and lightpropeller driven aeroplanes, someaircraft are not required to havecertified noise levels. However somemodifications to these aircraft can beapplied which may impact the noisecharacteristics. In which case the no-acoustical criterion application will betreated with a procedure approved bythe certificating authority.

1.5 RE-CERTIFICATION

1.5.1 Re-certification is defined as the“Certification of an aircraft with or withoutrevision to noise levels, to a Standard differentto that which it had been originally certified”.

1.5.2 In the case of an aircraft being re-certificated from the standards of ICAO Annex16, Volume 1, Chapter 3 to a more stringentstandard that may be directly applicable to newtypes only, noise re-certification should begranted on the basis that the evidence used todetermine compliance is as satisfactory as theevidence associated with a new type design.The date used by a certificating authority todetermine the re-certification basis should bethe date of acceptance of the first applicationfor re-certification.

1.5.3 The basis upon which the evidenceassociated with applications for re-certificationdescribed in paragraph 1.5.2 should beassessed is presented in Appendix 8.

jskaleck



4

THIS PAGE IS INTENTIONALLY BLANK

jskaleck



5

SECTION 2 - EQUIVALENT PROCEDURES FOR SUBSONIC JET POWERED AEROPLANES

The objective of a noise demonstration test isto acquire data for establishing an accurate andreliable definition of the aeroplane's noisecharacteristics under reference conditions (seeSection 2.6 of Annex 16, Volume 1, for Chapter 2aeroplanes and Section 3.6 for Chapter 3aeroplanes). In addition, the Annex sets forth arange of test conditions and the procedures foradjusting measured data to reference conditions.

2.1 FLIGHT TEST PROCEDURES

The following methods have been used toprovide equivalent results to Annex 16, Volume 1,Chapters 2 and 3 procedures for turbo-jet andturbo-fan powered aeroplanes.

2.1.1 Flight path intercept procedures

2.1.1.1 Flight path interceptprocedures in lieu of full take-off and/orlanding profiles described in paragraphs9.2 and 9.3 of Appendix 1 or paragraphs9.2 of Appendix 2 of Annex 16, Volume1, have been used to meet thedemonstration requirements of noisecertification. The intercept procedureshave also been used in the implementationof the generalised flight test proceduresdescribed in Section 2.1.2 of this manual.The use of intercepts eliminates the needfor actual take-offs and landings (withsignificant cost and operational advantagesat high gross mass) and substantiallyreduces the test time required. Siteselection problems are reduced and theshorter test period provides a higherprobability of stable meteorologicalconditions during testing. Aeroplanewear, and fuel consumption are reducedand increased consistency and quality innoise data are obtained.

2.1.1.2 Figure l(a) illustrates a typicaltake-off profile. The aeroplane is initiallystabilised in level flight at a point A andcontinues to point B where take-off poweris selected and a steady climb is initiated.The steady climb condition is achieved atpoint C, intercepting the reference take-offflight path and continuing to the end of thenoise certification take-off flight path.Point D is the theoretical take-off rotationpoint used in establishing the referenceflight path. If cutback power is employed,

point E is the point of application ofpower cutback and F, the end of the noisecertification take-off flight path. Thedistance TN is the distance over which theposition of the aeroplane is measured andsynchronised with the noise measurementat K.

2.1.1.3 For approach, the aeroplaneusually follows the planned flighttrajectory while maintaining a constantconfiguration and power until no influenceon the noise levels within ten decibels ofPNLTM. The aeroplane then carries out ago-around rather than continuing thelanding (See Fig. 1(b)).

2.1.1.4 For the development of thenoise-power-distance data for theapproach case (see 2.1.2.1) the speed, andapproach angle constraints imposed byAnnex 16, Volume 1 in 2.6.2 and 3.6.3and 3.7.5 cannot be satisfied over thetypical ranges of thrust needed. For theapproach case speed shall be maintained at1.3 VS + 19 km/h (1.3 VS + 10 kt) towithin ±9 km/h or ±5 kt and flyover heightover the microphone maintained at400 ft ±100 ft. However the approachangle at the test thrust shall be that whichresults from the aircraft conditions, ie.mass, configuration, speed and thrust.

2.1.1.5 The flight profiles should beconsistent with the test requirements of theAnnex over a distance that corresponds atleast to noise levels 10 dB below themaximum tone corrected Perceived NoiseLevel (PNLTM) obtained at themeasurement points during thedemonstration.

2.1.2 Generalised flight test procedures

The following equivalent flight testprocedures have been used for noisecertification compliance demonstrations.

2.1.2.1 Derivation of noise, power,distance data

2.1.2.1.1 For a range of powerscovering full take-off and cut-backpowers, the aeroplane is flown pastlateral and under-flight-pathmicrophones according to either the

jskaleck



6

take-off procedures defined inparagraph 3.6.2 of Volume 1 ofAnnex 16 or, more typically, flightpath intercept procedures described inSection 2.1.1 of this manual. Targettest conditions are established foreach sound measurement. Thesetarget conditions define the flightprocedure, aerodynamic configurationto be selected, aeroplane weight,power, airspeed and, at the closestpoint of approach to the measurementlocation, height. Regarding choice oftarget airspeeds and variation in testweights, the possible combinations ofthese test elements may affect theaeroplane angle-of-attack or aeroplaneattitude and therefore possibly theaeroplane sound generation orpropagation geometry. The aeroplaneangle-of-attack will remainapproximately constant for all testweights if the tests are conducted atthe take-off reference airspeedappropriate for each test weight. (Asan example if the appropriate take-offreference airspeed for the aeroplane isV2+15 kt set the target airspeed foreach test weight at V2+15 kt; theactual airspeed magnitude will varyaccording to each test weight but theaeroplane test angle-of-attack willremain approximately constant.)Alternatively, for many aeroplanes theaeroplane attitude remainsapproximately constant for all testweights if all tests are conducted atthe magnitude of the take-offreference airspeed corresponding tothe maximum take-off weight. (As anexample if the approximate take-offreference airspeed for the aeroplane isV2+15 kt set the target airspeed foreach test weight at the magnitude ofthe V2+15 kt airspeed thatcorresponds to the maximum take-offweight; the airspeed magnituderemains constant for each test weightand the aeroplane attitude remainsapproximately constant.) Review ofthese potential aeroplane sensitivitiesmay dictate the choice of targetairspeeds and/or test weights in thetest plan in order to limit excessivechanges in angle-of-attack oraeroplane attitude that couldsignificantly change measured noisedata. In the execution of eachcondition the pilot should “set up” the

aeroplane in the appropriate conditionin order to pass by the noisemeasurement location within thetarget height window, whilemaintaining target power andairspeed, within agreed tolerances,throughout the 10dB-down timeperiod.

2.1.2.1.2 A sufficient number ofnoise measurements are made toenable noise-power curves at a givendistance for both lateral and flyovercases to be established. These curvesare extended either by calculation orby the use of additional flight test datato cover a range of distances to formthe generalised noise data base for usein the noise certification of the "flightdatum" and derived versions of thetype and are often referred to asNoise-Power-Distance (NPD)plots(see Fig. 2). If over any portionof the range for the NPD plot thecriteria for calculating the EPNdBgiven in paragraphs 9.1.2 and 9.1.3 ofAppendix 2 of Annex 16, Volume 1,requires the use of the integratedprocedure, this procedure shall beused for the whole NPD. The90 per cent confidence intervals aboutthe mean lines are constructedthrough the data (see paragraph 2.2 ofAppendix 1).

Note: The same techniques can beused to develop NPD’s appropriatefor the derivation of approach noiselevels by flying over an under flightpath microphone for a range ofapproach powers using the speed andaeroplane configuration given inparagraph 3.6.3 of Annex 16, Volume1, or more typically, flight testprocedures described in 2.1.1 of thismanual.

2.1.2.1.3 Availability of flighttest data for use in data adjustment,e.g. speed and altitude, should beconsidered in test planning and maylimit the extent to which a derivedversion may be developed withoutfurther flight testing especially wherethe effects of airspeed on source noiselevels become significant. The effectsof high altitude test site location on jetnoise source levels should also beconsidered in test planning. High

jskaleck



7

altitude test site locations have beenapproved under conditions specifiedin Appendix 6 provided that jet noisesource corrections are applied to thenoise data. The correction method ofAppendix 6 has been approved forthis purpose.

2.1.2.1.4 The take-off, lateraland approach noise measurementsshould be corrected to the referencespeed and atmospheric conditionsover a range of distances inaccordance with the proceduresdescribed in Appendix 1 (Chapter 2aeroplanes) or Appendix 2 (Chapter 3aeroplanes) of Annex 16, Volume 1.The NPD plots can then beconstructed from the correctedEffective Perceived Noise Level(EPNL), power and distanceinformation. The curves present theEPNL value for a range of distanceand engine noise performanceparameters, (see Annex 16, Volume 1,paragraph 9.3.4.1 of Appendix 2).The parameters are usually thecorrected low pressure rotor speed

N t1 2θ or the corrected net thrust

FN δamb (see Fig. 2), where:

N1 is the actual low pressurerotor speed;

θt2 is the ratio of absolute static

temperature of the air at theheight of the aeroplane to theabsolute temperature of the airfor an international standardatmosphere (ISA) at mean sealevel (ie. 288.15 °K);

FN is the actual engine net thrustper engine; and

δamb is the ratio of absolute staticpressure of the ambient air at theheight of the aeroplane to ISA airpressure at mean sea level (ie.101.325 kPa).

2.1.2.1.5 Generalised NPD datamay be used in the certification of theflight tested aeroplane and derivativeversions of the aeroplane type. Forderived versions, these data may beused in conjunction with analytical

procedures, static testing of the engineand nacelle or additional limited flighttests to demonstrate compliance.

2.1.2.2 Procedures for thedetermination of changes in noise levels

Noise level changes determined bycomparison of flight test data for differentdevelopments of an aeroplane type havebeen used to establish certification noiselevels of newly derived versions byreference to the noise levels of the "flightdatum" aeroplane. These noise changesare added to or subtracted from the noiselevels obtained from individual flights ofthe "flight datum" aeroplane. Confidenceintervals of new data are statisticallycombined with the "flight datum" data todevelop overall confidence intervals (seeAppendix 1).

2.1.3 The determination of the lateralnoise certification levels

2.1.3.1 Alternative procedures usingtwo microphone stations locatedsymmetrically on either side of the take-off reference track has proved to beeffective in terms of time and costssavings. Such an arrangement avoidsmany of the difficulties encountered inusing the more conventional multi-microphone arrays. The proceduresconsist of flying the test aeroplane at fulltake-off power at one (or more) specifiedheights above a track at right angles to andmidway along the line joining the twomicrophone stations. However, when thisprocedure is used matching data from bothlateral microphones for each fly-by shouldbe used for the lateral noise determination;cases where data from only onemicrophone is available for a given runmust be omitted from the determination.The following paragraphs describe theprocedures for determining the lateralnoise level for subsonic turbo-jet or turbo-fan powered aeroplanes.

2.1.3.2 Lateral noise measurementsfor a range of conventionally configuredaeroplanes with under wing and/or rear-fuselage mounted engines with bypassratio of more than 2, have shown that themaximum lateral noise at full powernormally occurs when the aeroplane is

jskaleck



8

close to 300 m (985 ft) or 435 m*

(1,427 ft) in height during the take-off.Based on this finding it is consideredacceptable to use the following as anequivalent procedure:

a) for aeroplanes to be certificatedunder Chapter 3 or Chapter 2 ofAnnex 16, Volume 1, twomicrophone locations are used,symmetrically placed on eitherside of the aeroplane referenceflight track and 450 m or 650 m*from it;

b) for aeroplanes with engineshaving bypass ratios of more than2, the height of the aeroplane as itpasses the microphone stationsshould be 300 m (985 ft) or435 m* (1,427 ft) and be no morethan +100 m, -50 m (+328 ft, -164 ft) relative to this targetheight. For aeroplanes withbypass ratios of 2 or less it isnecessary to determine the peaklateral noise by undertaking anumber of flights over a range ofheights to define the noise(EPNL) versus heightcharacteristics. A typical heightrange would cover 60 m (200 ft)to 600 m (2000 ft) above theinter-section of a track at rightangles to the line joining the twomicrophone positions and thisline;

c) constant power, configurationand airspeed as described inparagraphs 3.6.2.1 a), 3.6.2.1 d),2.6.1.2 and 2.6.1.3 of Annex 16,Volume 1, should be used duringthe flight demonstration;

d) adjustment of measured noiselevels should be made to theacoustical reference dayconditions and to referenceaeroplane operating conditions asspecified in Section 9 ofAppendix 1 and 2 of Annex 16,Volume 1; and

e) to account for any possibleasymmetry effects in measurednoise levels, the reported lateral

* For Chapter 2 procedures.

noise level for purposes ofdemonstrating compliance withthe applicable noise limit ofChapter 3 or Chapter 2 of Annex16, Volume 1, as applicable,should be the arithmetic averageof the corrected maximum noiselevels from each of the twolateral measurement points andcompliance should be determinedwithin the ±1.5 dB 90 per centconfidence interval required bythe Annex (see paragraph 2 ofAppendix 1 of this manual).

2.1.3.3 Lateral noise certificationlevel determination has also beenaccomplished using multiple pairs oflateral microphones rather than only onepair of symmetrically locatedmicrophones. A sufficient number ofacceptable data points, resulting from aminimum of six runs, must be obtainedfrom sufficiently spaced microphone pairsto adequately define the maximum lateralnoise certification level and provide anacceptable 90 per cent confidence interval.

2.1.4 Take-off flyover noise levels withpower cut-back

Flyover noise levels with power cut-backmay also be established without makingmeasurements during take-off with full powerfollowed by power reduction in accordancewith paragraph 2.2.1 of this manual.

2.1.5 Measurements at non-referencepoints

2.1.5.1 In some instances testmeasurement points may differ from thereference measurement points specified inAnnex 16, Volume 1, Chapters 2 and 3,paragraphs 2.3.1 and 3.3.1. Under thesecircumstances an applicant may requestapproval of data that have been adjustedfrom the actual measurements to representdata that would have been measured at thereference points in reference conditions.

2.1.5.2 Reasons for such a requestmay be:

a) to allow the use of a measurementlocation that is closer to theaeroplane flight path so as toimprove data quality by obtaining

jskaleck



9

a greater ratio of signal tobackground noise. WhereasAppendix 3 describes aprocedure for removing theeffects of ambient noise the useof data collected closer to theaeroplane avoids theinterpolations and extrapolationsinherent in the method;

b) to enable the use of an existing,approved certification data basefor an aeroplane type design inthe certification of a derivative ofthat type when the derivative is tobe certificated under referenceconditions that differ from theoriginal type certificationreference conditions; and

c) to avoid obstructions near thenoise measurement station(s)which could influence soundmeasurements. When a flightpath intercept technique is beingused, take-off and approach noisemeasurement stations may berelocated as necessary to avoidundesirable obstructions.Sideline measurement stationsmay be relocated by distanceswhich are of the same order ofmagnitude as the aeroplanelateral deviations (or offsets)relative to the nominal flightpaths that occur during flighttesting.

2.1.5.3 Approval has been granted toapplicants for the use of data from non-reference noise measurement pointsprovided that measured data are adjustedto reference conditions in accordance withthe requirements of section 9 of Appendix1 or 2 of Annex 16, Volume 1, and themagnitudes of the adjustments do notexceed the limitations in section 5.4 ofAppendix 1 and paragraph 3.7.6 ofChapter 3 of the Annex.

2.1.6 Atmospheric test conditions

It has been found acceptable bycertificating authorities to exceed the soundattenuation limits of Annex 16, Volume 1,Appendix 2, Section 2.2.2(c) when:

a) the dew point and dry bulbtemperature are measured with a

device which is accurate to ± 0.5 °Cand are used to obtain relativehumidity and when 'layered' sectionsof the atmosphere are used tocompute equivalent weighted soundattenuations in each one-third octaveband, sufficient sections being used tothe satisfaction of the certificatingauthority; or

b) where the peak noy values at the timeof PNLT, after adjustment toreference conditions, occur atfrequencies of less than or equal to400 Hz.

2.1.7 Reference approach speed

The reference approach speed is currentlycontained in 3.6.3.1(b) of Chapter 3 of Annex16, Volume 1 as 1.3 VS + 19 km/h(1.3 VS + 10 kt). There is a change beingmade to the definition of stall speed, forairworthiness reasons, to alter the currentminimum speed VS definition to a stall speedduring a 1-g manoeuvre (ie. a flight load factorof unity) VS1g. In terms of the new definitionthe approach reference speed becomes1.23 VS1g + 19 km/h, (1.23 VS1g + 10 kt)which can be taken as equivalent to thereference speed contained in Chapter 3.

2.2 ANALYTICAL PROCEDURES

Analytical equivalent procedures rely uponavailable noise and performance data obtained fromflight test for the aeroplane type. Generalisedrelationships between noise, power and distance(NPD plots see 2.1.2.1) and adjustment proceduresfor speed changes in accordance with the methodsof Appendix 1 or 2 of Annex 16, Volume 1, arecombined with certificated aeroplane aerodynamicperformance data to determine noise level changesresulting from type design changes. These noiselevel increments are then applied to noise levels inaccordance with paragraph 2.1.2.2 of this manual.

2.2.1 Flyover noise levels with powercut-back

Note: The “average engine” spool-down timeshould reflect a 1.0 second minimum altituderecognition lag time to account for pilotresponse.

2.2.1.1 Flyover noise levels withpower cut-back may be established fromthe merging of PNLT versus time

jskaleck



10

measurements obtained during constantpower operations. As seen in Fig. 4(a) the10 dB-down PNLT noise time historyrecorded at the flyover point may containportions of both full power and cut-backpower noise time histories. Provided thesenoise time histories, the average enginespool-down thrust characteristics and theaeroplane flight path during this period(See Fig. 4(b)), which includes thetransition from full to cut-back power, areknown, the flyover noise level may becomputed.

2.2.1.2 Where the full power portionof the noise time history does not intrudeupon the 10 dB-down time history of thecut-back power, the flyover noise levelsmay be computed from a knowledge of theNPD characteristics and the effect of theaverage spool-down thrust characteristicson the aeroplane flight path.

Note: To ensure that the full powerportion of the noise time history does notintrude upon the 10 dB-down noiselevels, PNLTM PNLT dB

after cutback before cutback − ≥ 10 5.

2.2.2 Equivalent procedures based onanalytical methods

Noise certification approval has beengiven for applications based on type designchanges that result in predictable noise leveldifferences including the following:

a) changes to the originally certificatedtake-off or landing mass which lead tochanges in distance between theaeroplane and the microphone for thetake-off case and changes to theapproach power. In this case the NPDdata may be used to determine thenoise certification level of the derivedversion;

b) noise changes due to engine powerchanges. However, care should betaken to ensure that when NPD plotsare extrapolated the relativecontribution of the component noisesources to the Effective PerceivedNoise Level remains essentiallyunchanged and a simple extrapolationof the noise/power and noise/distancecurves can be made. Among theitems which should be considered inextension of the NPD are:

− the 90% confidence interval atthe extended power;

− aeroplane/engine source noisecharacteristics and behaviour;

− engine cycle changes; and− quality of data to be extrapolated.

c) aeroplane engine and nacelleconfiguration and acoustical treatmentchanges, usually leading to changes inEPNL of less than one decibel.However, it should be ensured thatnew noise sources are not introducedby modifications made to theaeroplane, engine or nacelles. Avalidated analytical noise modelapproved by the certificating authoritymay be used to derive predictions ofnoise increments. The analysis mayconsist of modelling each aeroplanecomponent noise source andprojecting these to flight conditions ina manner similar to the static testprocedure described in paragraph 2.3.A model of detailed spectral anddirectivity characteristics for eachaeroplane noise component may bedeveloped by theoretical and/orempirical analysis. Each componentshould be correlated to theparameter(s) which relates to thephysical behaviour of sourcemechanisms. The sourcemechanisms, and subsequently thecorrelating parameters, should beidentified through use of othersupplemental tests such as engine orcomponent tests. As described inparagraph 2.3 an EPNL representativeof flight conditions should becomputed by adjusting aeroplanecomponent noise sources for forwardspeed effects, number of engines andshielding, reconstructing the totalnoise spectra and projecting the totalnoise spectra to flight conditions byaccounting for propagation effects.The effect of changes in acoustictreatment, such as nacelle lining, maybe modelled and applied to theappropriate component noise sources.The computation of the total noiseincrements, the development of thechanged version NPD, and theevaluation of the changed versioncertification levels should be madeusing the procedures in paragraphs2.3.4.12 and 2.3.4.13. Guidancematerial on confidence interval

jskaleck



11

computations is provided in Appendix1.

d) airframe design changes such aschanges in fuselage length, flapconfiguration and engine installation,that could indirectly affect noiselevels because of an effect onaeroplane performance (increaseddrag for example). Changes inaeroplane performance characteristicsderived from aerodynamic analysis ortesting have been used to demonstratehow these changes affect theaeroplane flight path and hence thedemonstrated noise levels of theaeroplane.

In these cases care should be exercised toensure that the airframe changes do notintroduce significant new noise sources normodify existing source generation or radiationcharacteristics. In such instances themagnitude of such effects may need to beestablished by test.

2.3 STATIC TESTS AND PROJECTIONS TO

FLIGHT NOISE LEVELS

2.3.1 General

2.3.1.1 Static test evidence providesvaluable definitive information forderiving the noise levels resulting fromchanges to an aeroplane powerplant or theinstallation of a broadly similarpowerplant into the airframe followinginitial noise certification of the flightdatum aeroplane. This involves the testingof both the flight datum and derivativepowerplants using an open-air test facilitywhereby the effect on the noise spectra ofthe engine modifications in the aeroplanemay be assessed. It can also extend to theuse of component test data to demonstratethat the noise levels remain unchangedwhere minor development changes havebeen made.

2.3.1.2 Approval of equivalentprocedures for the use of static testinformation depends critically upon theavailability of an adequate approved database (NPD plot) acquired from the flighttesting of the flight datum aeroplane.

2.3.1.3 Static tests can providesufficient additional data or noise source

characteristics to allow a prediction to bemade of the effect of changes on the noiselevels from the aeroplane in flight.

2.3.1.4 Types of static test acceptedfor the purposes of certificationcompliance demonstration in aeroplanedevelopment include engine andcomponent noise tests and performancetesting. Such tests are useful for assessingthe effects of mechanical andthermodynamic cycle changes to theengine on the individual noise sources.

2.3.1.5 Static engine testing is dealtwith in detail in subsequent sections. Thecriteria for acceptance of component testsare less definable. There are manyinstances, particularly when only smallEPNL changes are expected, thatcomponent testing provides an adequatedemonstration of noise impact. Theseinclude, for example:

a) changes in the specification ofsound absorbing linings within anengine nacelle;

b) changes in the mechanical oraerodynamic design of the fan,compressor or turbine;

c) changes to combustor designs;and

d) minor exhaust system changes.

2.3.1.6 Each proposal by theapplicant to use component test datashould be considered by the certificatingauthority with respect to the significanceof the relevant affected source on theEPNL of the aeroplane.

2.3.2 Limitation on the projection ofstatic to flight data

Details of the acceptability, use andapplicability of static test data are contained insubsequent sections.

2.3.2.1 The amount by which themeasured noise levels of a derivativeengine will differ from the referenceengine is a function of several factors,including:

jskaleck



12

a) thermodynamic changes to theengine cycle, including increasesin thrust;

b) design changes to majorcomponents, e.g. the fan,compressor, turbine, exhaustsystem, etc.; and

c) changes to the nacelle.

2.3.2.2 Additionally, day-to-day andtest site-to-site variables can influencemeasured noise levels and therefore thetest, measurement and analysis proceduresdescribed in this manual are designed toaccount for these effects. In order that thedegree of change resulting from aspectssuch as (a), (b) and (c) above, whenextrapolated to flight conditions, areconstrained to acceptable amounts beforea new flight test is required, a limit isneeded that can be used uniformly bycertificating authorities.

2.3.2.3 The recommended guidelinefor this limit is that the summation of themagnitudes, neglecting signs, of the noisechanges, for the three referencecertification conditions, between the flightdatum aeroplane and the derived version,at the same thrust and distance (for thederived version), is no greater than5 EPNdB with a maximum of 3 EPNdB atany one of the reference conditions (seefigure 5).

2.3.2.4 For differences greater thanthis additional flight testing at conditionswhere noise levels are expected to changeis recommended to establish a new flightNPD data base.

2.3.2.5 Provided the detailedprediction procedures used are verified byflight test for all the types of noise sources,ie. tones, non-jet broadband and jet noiserelevant to the aeroplane underconsideration and there are no significantchanges in installation effects between theaeroplane used for the verification of theprediction procedures and the aeroplaneunder consideration, the procedure may beemployed without the limitationsdescribed above.

2.3.2.6 In addition to the limitationsdescribed above a measure of acceptabilityregarding methodologies for static to flight

projection is also needed that can be useduniformly by certificating authorities.This measure can be derived as residualNPD differences between the flight testdata and the projected static to flight datafor the original aeroplane version. Theguideline for a measure of acceptability isto limit these residuals to 3 EPNdB at anyone of the reference conditions.

2.3.2.7 In the determination of thenoise levels of the modified or derivedversion the same analytical procedures asused in the first static to flight calculationsfor the noise certification of the aeroplanetype shall be used.

2.3.3 Static engine tests

2.3.3.1 General

2.3.3.1.1 Data acquired fromstatic tests of engines of similardesigns to those that were flight testedmay be projected, when appropriate,to flight conditions and, afterapproval, used to supplement anapproved NPD plot for the purpose ofdemonstrating compliance with theAnnex 16, Volume 1, provisions insupport of a change in type design.This section provides guidelines onstatic engine test data acquisition,analysis and normalisation techniques.The information provided is used inconjunction with technicalconsiderations and the generalguidelines for test site, measurementand analysis instrumentation, and testprocedures provided in the latestversion of the Society of AutomotiveEngineers (SAE) ARP 1846,"Measurement of Noise from GasTurbine Engines During StaticOperation". The engine designs andthe test and analysis techniques to beused should be presented in the testplan and submitted, for approval, tothe certificating authority forconcurrence prior to testing. Notethat test restrictions defined for flighttesting in conformity with Annex 16,Volume 1, are not necessarilyappropriate for static testing. (SAEARP 1846 provides guidance on thissubject).

2.3.3.1.2 For example, themeasurement distances associated

jskaleck



13

with static tests are substantially lessthan those encountered in flighttesting and may permit testing inatmospheric conditions not permittedfor flight testing by Annex 16,Volume 1. Moreover, since staticengine noise is a steady soundpressure level rather than the transientnoise level of a flyover, themeasurement and analysis techniquesmay be somewhat different for staticnoise testing.

2.3.3.2 Test site requirements

The test site should meet at least thecriteria specified in SAE ARP 1846.Different test sites may be selected fortesting differing engine configurationsprovided the acoustic measurements fromthe different sites can be adjusted to acommon reference condition.

2.3.3.3 Engine inlet bell mouth

The installation of a bell mouthforward of the engine inlet may be usedwith turbofan or turbojet engines duringstatic noise tests. Such an installation isused to provide a simulated flightcondition of inlet flow during statictesting. Production inlet acoustic liningand spinners are also to be installed duringnoise testing.

2.3.3.4 Inflow control devices

2.3.3.4.1 The use of static enginetest noise data for the noisecertification of an aeroplane with achange of engine to one of a similardesign requires the use of an approvedInflow Control Device (ICD) for highbypass engines (BPR > 2.0). TheICD should meet the followingrequirements:

a) The specific ICD hardwaremust be inspected by thecertificating authority toensure that the ICD is freefrom damage andcontaminants that may affectits acoustic performance;

b) The ICD must beacoustically calibrated by anapproved method (such asthat provided in paragraph

2.3.3.4.3) to determine itseffect on sound transmissionin each one-third octaveband;

c) Data obtained during statictesting must be corrected toaccount for soundtransmission effects that arecaused by the ICD. Thecorrections shall be appliedto each one-third octaveband of data measured;

d) The ICD position relative tothe engine inlet lip must bedetermined and thecalibration must beapplicable to that position;and

e) No more than one calibrationis required for an ICDhardware design, providedthat there is no deviationfrom the design for any oneICD serial number hardwareset.

2.3.3.4.2 It is not necessary toapply the ICD calibration correctionsif the same ICD hardware (identicalserial number) is used as waspreviously used in the static noise testof the flight engine configuration, andthe fan tones for both engines remainin the same one-third octave bands.

2.3.3.4.3 ICD calibration

An acceptable ICD calibrationmethod is as follows:

a) Place an acoustic driver(s)on a simulated enginecentreline in the plane of theengine inlet lip. Locate thecalibration microphones onthe forward quadrantazimuth at a radius between50 ft and 150 ft that providesa good signal-to-ambientnoise ratio and at eachmicrophone angle to be usedto analyse static engine noisedata. Locate a referencenear-field microphone on thecentreline of and within 2 ft

jskaleck

jskaleck

jskaleck



14

of the acoustic centre of theacoustic driver(s);

b) Energise the acoustic driverwith pink noise without theICD in place. Record thenoise for a minimum of 60 sduration following systemstabilisation. The proceduremust be conducted at aconstant input voltage to theacoustic driver(s);

c) Repeat item (b), alternatelywith and without the ICD inplace. A minimum of threetests of each configuration(with and without ICD inplace) is required. To beacceptable, the total variationof the 55° microphone on-line OASPL signal (averagedfor a 1 minute duration) forall three test conditions ofeach configuration shall notexceed 0.5 dB;

Note: Physically moving the ICDalternately in and out of place forthis calibration may beeliminated if it is demonstratedthat the ICD positioning does notaffect the calibration results.

d) All measured data are to becorrected for sound pressurelevel variations as measuredwith the near-fieldmicrophone and foratmospheric absorption to 77°F and 70 per cent RHconditions using the slantdistance between the outermicrophones and theacoustic driver(s);

e) The calibration for each one-third octave band at eachmicrophone is the differencebetween the average of thecorrected SPL's without theICD in place and the averageof the corrected SPL's withthe ICD in place; and

f) The tests must be conductedunder wind and thermalconditions that precludeacoustic shadowing at the

outer microphones andweather induced variations inthe measured SPL data.Refer to Figure 6, Section2.3.3.7.

In some cases large fluctuations in thevalue of the calibrations acrossadjacent 1/3 octave bands andbetween closely spaced angularpositions of microphones can occur.These fluctuations can be related toreflection effects caused by thecalibration procedure and care mustbe taken to ensure that they do notintroduce or suppress engine tones.This may be done by comparingeffective perceived noise levelscomputed with:

a) the ICD calibrations asmeasured;

b) a mean value of thecalibration curves; and

c) the calibration values set tozero.

2.3.3.5 Measurement and analysis

Measurement and analysis systemsused for static test, and the modusoperandi of the test programme, may wellvary according to the specific testobjectives, but in general they shouldconform with those outlined in SAE ARP1846. Some important factors to be takeninto account are highlighted in subsequentsections.

2.3.3.6 Microphone locations

2.3.3.6.1 Microphones should belocated over an angular rangesufficient to include the 10 dB-downtimes after projection of the staticnoise data to flight conditions.General guidance in SAE ARP 1846,describing microphone locations issufficient to ensure adequatedefinition of the engine noise sourcecharacteristics.

2.3.3.6.2 The choice ofmicrophone location with respect tothe test surface depends on thespecific test objectives and themethods to be used for data

jskaleck

jskaleck



15

normalisation. Certificationexperience with static engine testinghas been primarily limited tomicrophone installations near theground or at engine centreline height.In general, because of the difficultiesassociated with obtaining free-fieldsound pressure levels that are oftendesirable for extrapolating to flightconditions, near-ground-planemicrophone installations or acombination of ground-plane andelevated microphones have been used.Consistent microphone locations,heights, etc. are recommended fornoise measurements of both the priorapproved and changed version of anengine or nacelle.

2.3.3.7 Acoustic shadowing

2.3.3.7.1 Where ground planemicrophones are used, specialprecautions are necessary to ensurethat consistent measurements, e.g. freefrom "acoustic shadowing"(refraction) effects, will be obtained.When there is a wind in the oppositedirection to the sound wavepropagating from the engine, or whenthere is a substantial thermal gradientin the test arena, refraction caninfluence near ground planemicrophone measurements to a largerdegree than measurements at greaterheights.

2.3.3.7.2 Previous evidence, ordata from a supplemental test, may beused to demonstrate that testing at aparticular test site results in consistentmeasurements, including the absenceof shadowing. In lieu of thisevidence, a supplemental noisedemonstration test should include anapproved method to indicate theabsence of shadowing effects on theground plane measurements.

2.3.3.7.3 The following criteriaare suggested for certain testgeometries, based on measurements ofthree weather parameters as follows:

− average wind speeds at enginecentreline height (WCL);

− air temperature at enginecentreline height (TCL); and

− air temperature near ground planemicrophone height (TMIC).

a) The instruments for thesemeasurements should be co-located and placed close tothe 90° noise measurementposition without impedingthe acoustic measurement;

b) The suggested limits areadditional to the wind andtemperature limitsestablished by other criteria(such as the maximum windspeed at the microphone ifwind screens are not used);and

c) Wind and temperaturecriteria that have beenobserved to provideconsistent measurements thatpreclude any influence ofacoustic shadowing effectson ground planemeasurements are defined inFigure 6.

The line defines a boundary betweenthe absence of shadowing and thepossible onset of spectral deficienciesin the very high frequencies. Testingis permitted provided that the test dayconditions are such that the average(typically 30 s) wind speed at enginecentreline height falls below the lineshown, and that wind gusting does notexceed the value of the line shown bymore than 5.5 km/h (3 kt). Windspeeds in excess of the linearrelationship shown, between 7 and22 km/h (4 and 12 kt), may indicatethe need to demonstrate the absenceof spectral abnormalities, either priorto or at the time of test when the winddirection opposes the direction ofsound propagation.

When the temperature at the groundmicrophone height is not greater thanthe temperature at the enginecentreline height plus 4° K,shadowing effects due to temperaturegradients can be expected to benegligible.

Note: Theoretical analyses and theexpression of wind criteria in terms of

jskaleck

jskaleck


jskaleck


16

absolute speed rather than the vectorreduction suggest that the noted limitsmay be unduly stringent in somedirections.

2.3.3.8 Engine power test conditions

A range of static engine operatingconditions should be selected tocorrespond to the expected maximumrange of in-flight engine operatingconditions for the appropriate enginepower setting parameter. A sufficientnumber of stabilised engine power settingsover the desired range should be includedin the test to ensure that the 90 per centconfidence intervals in flight projectedEPNL can be established (see paragraph 3of Appendix 1 to this manual).

2.3.3.9 Data system compatibility

2.3.3.9.1 If more than one dataacquisition system and/or dataanalysis system is used for theacquisition or analysis of static data,compatibility of the airframe andengine manufacturers’ systems isnecessary. Compatibility of the dataacquisition systems can beaccomplished through appropriatecalibration. Compatibility of the dataanalysis systems can be verified byanalysing the same data samples onboth systems. The systems arecompatible if the resulting differencesare no greater than 0.5 EPNdB.Evaluation should be conducted atflight conditions representative ofthose for certification.

2.3.3.9.2 The use of pseudorandom noise signals with spectralshape and tonal content representativeof turbo-fan engines is an acceptablealternative to using actual enginenoise measurements for analysissystem compatibility determination.The systems are compatible if theresulting differences are no greaterthan 0.5 PNdB for an integration timeof 32 seconds.

2.3.3.10 Data acquisition, analysisand normalisation

For each engine power settingdesignated in the test plan, the engineperformance, meteorological and sound

pressure level data should be acquired andanalysed using instrumentation and testprocedures described in SAE ARP 1846.Sound measurements should benormalised to consistent conditions andinclude 24 1/3-octave band sound pressurelevels between band centre frequencies of50 Hz to 10 kHz for each measurement(microphone) station. Before projectingthe static engine data to flight conditions,the sound pressure level data should becorrected for:

a) the frequency responsecharacteristics of the dataacquisition and analysis system;and

b) contamination by backgroundambient or electrical systemnoise. (See Appendix 3).

2.3.4 Projection of static engine data toaeroplane flight conditions

2.3.4.1 General

2.3.4.1.1 The static engine soundpressure level data acquired at eachangular location should be analysedand normalised to account for theeffects identified in 2.3.3.8. Theyshould then be projected to the sameaeroplane flight conditions used in thedevelopment of the approved NPDplot.

2.3.4.1.2 As appropriate, theprojection procedure includes:

a) effects of source motionincluding Doppler effects;

b) number of engines andshielding effects;

c) installation effects;

d) flight geometry;

e) atmospheric propagation,including spherical wavedivergence and atmosphericattenuation; and

f) flight propagation effectsincluding ground reflectionand lateral attenuation. (Seeparagraph 2.3.4.11).

jskaleck



17

2.3.4.1.3 To account for theseeffects, the measured total static noisedata should be analysed to determinecontributions from individual noisesources. After projecting the 1/3octave-band spectral data to flightconditions, Effective Perceived NoiseLevels should be calculated for therevised NPD plot. Guidelines on theelements of an acceptable projectionprocedure are provided in this section.The process is also illustrated in Figs7 and 8.

2.3.4.1.4 It is not intended thatthe procedure illustrated in Figs 7 and8 should be exclusive. There areseveral options, depending upon thenature of the powerplant noise sourcesand the relevance of individual noisesources to the Effective PerceivedNoise Level of the aeroplane. Themethod presented does, however,specify the main features that shouldbe considered in the computationalprocedure.

2.3.4.1.5 It is also not necessarythat the computations should alwaysbe carried out in the order specified.There are interrelations between thevarious steps in the procedure whichdepend on the particular form of thecomputation being followed. Hencethe most efficient manner ofstructuring the computation cannotalways be pre-determined.

2.3.4.1.6 There are severalengine installation effects which canmodify the generated noise levels butwhich cannot be derived from statictests. Additional noise sources suchas jet/flap or jet/wind interactioneffects may be introduced on aderived version of the aeroplanewhich are not present on the flightdatum aeroplane. Far-field noisedirectivity patterns (field shapes) maybe modified by wing/nacelle or jet-by-jet shielding, tailplane and fuselagescattering or airframe reflectioneffects. However, general methods toadjust for these effects are not yetavailable. It is therefore importantthat, before the following proceduresare approved for the derived versionof the aeroplane, the geometry of the

airframe and engines in the vicinity ofthe engines be shown to be essentiallyidentical to that of the flight datumaeroplanes so that the radiated noise isessentially unaffected.

2.3.4.2 Normalisation to referenceconditions

2.3.4.2.1 The analysed static testdata should be normalised to freefieldconditions in the Annex 16, Volume 1reference atmosphere. Thisadjustment can only be applied with aknowledge of the total spectra beingthe summation of all the noise sourcespectra computed as described inparagraphs 2.3.4.3 to 2.3.4.5.

2.3.4.2.2 The requiredadjustments include:

a) Atmospheric absorption:adjustments to account forthe acoustical reference dayatmospheric absorption aredefined in SAE ARP 866A(revised 15th March 1975).In the event that minordifferences in absorptionvalues are found in SAEARP 866A betweenequations, tables or graphs,the equations should be used.

The atmospheric absorptionshould be computed over theactual distance from theeffective centre of each noisesource to each microphone,as determined in 2.3.4.5; and

b) Ground reflection: examplesof methods for obtainingfreefield sound pressurelevels are described in SAEAIR 1672B-1983 orEngineering Sciences DataUnit, ESDU Item 80038Amendment A.

Spatial distribution of noisesources do not have a firstorder influence on groundreflection effects and hencemay be disregarded. It isalso noted that measurementsof far-field sound pressurelevels with ground-plane

jskaleck



18

microphones may be used toavoid the large spectralirregularities caused byinterference effects atfrequencies less than 1 kHz.

2.3.4.3 Separation into broadbandand tone noise

2.3.4.3.1 The purpose ofprocedures described in this section isto identify all significant tones in thespectra, firstly to ensure that tones arenot included in the subsequentestimation of broadband noise, andsecondly to enable the Doppler-shifted tones (in-flight) to be allocatedto the correct 1/3 octave band atappropriate times during a simulatedaeroplane flyover.

2.3.4.3.2 Broadband noiseshould be derived by extracting allsignificant tones from the measuredspectra. One concept for theidentification of discrete tones is thatused in Annex 16, Volume 1, Chapter3 (Appendix 2) for tone correctionpurposes (that is, considering theslopes between adjacent 1/3 octaveband levels). Care must be taken toavoid regarding tones as "nonprotrusive" when the surroundingbroad band sound pressure level islikely to be lower when adjusted fromstatic to flight conditions, orclassifying a closely grouped pair, orseries, of tones as broadband noise.One technique for resolving suchproblems is the use of narrow bandanalysis with a bandwidth of less than50 Hz.

2.3.4.3.3 Narrow band analysiscan also be used to check the validityof other tone identification proceduresin establishing the spectral characterat critical locations in the sound field,e.g. around the position of peakPNLT, or where predominant turbo-machinery tones exist.

2.3.4.4 Separation into contributingnoise sources

2.3.4.4.1 The number of noisesources which require identificationwill to some extent depend on theengine being tested and the nature of

the change to the engine or nacelle.Separation of broadband noise intothe combination of noise generated byexternal jet mixing and by internalnoise sources is the minimum andsometimes adequate requirement. Amore sophisticated analysis may benecessary depending upon thesignificance of the contribution fromother individual sources, which couldinvolve identifying broadband noisefrom fan, compressor, combustor andturbine. Furthermore, for fan andcompressor noise, the split of both thebroadband and the tone noise betweenthat radiating from the engine intakeand that from the engine exhaustnozzle(s) could be a furtherrefinement.

2.3.4.4.2 To meet the minimumrequirement, separation of sources ofbroadband noise into those due toexternal jet mixing and thosegenerated internally can be carried outby estimating the jet noise by one ormore of the methods identified below,and adjusting the level of thepredicted spectrum at each angle to fitthe measured low frequency part ofthe broadband spectrum at which jetnoise can be expected to be dominant.

2.3.4.4.3 There are three meanswhich have been used to obtainpredicted jet noise spectra shapes:

a) For single-stream engineswith circular nozzles theprocedure detailed in SAEARP 876C-1985 may beused. However, the enginegeometry may possessfeatures which can renderthis method inapplicable.Example procedures for co-axial flow engines areprovided in SAE AIR 1905-1985;

b) Analytical procedures basedon correlating full scaleengine data with modelnozzle characteristics may beused. Model data have beenused to supplement full scaleengine data, particularly atlow power settings, becauseof the uncertainty in defining

jskaleck



19

the level of jet noise at thehigher frequencies wherenoise from other enginesources may make asignificant contribution tothe broadband noise; and

c) Special noise source locationtechniques are availablewhich, when used duringfull-scale engine tests, canidentify the positions andlevels of separate enginenoise sources.

2.3.4.5 Noise source position effects

2.3.4.5.1 Static engine noisemeasurements are often made atdistances at which engine noisesources cannot be truly treated asradiating from a single acousticcentre. This may not give rise todifficulties in the extrapolation todetermine the noise increments fromstatic data to flight conditions becausenoise increments in EPNL are notparticularly sensitive to theassumption made regarding the spatialdistribution of noise sources.

2.3.4.5.2 However, in somecircumstances (for example, wherechanges are made to exhauststructures, and the sources of externaljet-mixing noise are of overridingsignificance) it may be appropriate toidentify noise source positions moreaccurately. The jet noise can beconsidered as a noise sourcedistributed downstream of the engineexhaust plane. Internal sources ofbroadband engine noise may beconsidered as radiating from theintake and the exhaust.

2.3.4.5.3 There are threeprincipal effects to be accounted foras a consequence of the position ofthe noise source differing from the"nominal" position assumed for the"source" of engine noise:

a) Spherical divergence: thedistance of the source fromthe microphone differs fromthe nominal distance; aninverse square law

adjustment needs to beapplied.

b) Directivity: the anglesubtended by the line fromthe source to the microphoneand the source to the enginecentreline differs from thenominal angle; a linearinterpolation should be madeto obtain data for the properangle.

c) Atmospheric attenuation: thedifference between the trueand the nominal distancebetween the source and themicrophone alters theallowance made foratmospheric attenuation inparagraph 2.3.4.2 above.

2.3.4.5.4 Source position can beidentified either from noise sourcelocation measurements (made either atfull or model scale), or from ageneralised data base.

Note: No published standard oncoaxial jet noise source distribution iscurrently available. An approximatedistribution for a single jet is given bythe following equation (seeReferences 1 and 2):

( )x D S S= +−

0 057 0 021 21

2. .

where:

S is the Strouhal numberfD Vj ;

x is the distance downstreamfrom the nozzle exit;

D is the nozzle diameter based ontotal nozzle exit area;

Vj is the average jet velocity forcomplete isentropic expansion toambient pressure from averagenozzle-exit pressure andtemperature; and

f is the 1/3 octave band centrefrequency.

jskaleck



20

2.3.4.6 Engine flight conditions

2.3.4.6.1 Some thermodynamicconditions within an engine testedstatically differ from those that existin flight and account should be takenof this. Noise source strengths maybe changed accordingly. Therefore,values for key correlating parametersfor component noise sourcegeneration should be based on theflight condition and the static database should be entered at theappropriate correlating parametervalue. Turbo-machinery noise levelsshould be based on the inflight

corrected rotor speeds N t1 2θ and

jet noise levels should be based on therelative jet velocities that exist at theflight condition.

2.3.4.6.2 The variation of sourcenoise levels with key correlatingparameters can be determined fromthe static data base which includes anumber of different thermodynamicoperating conditions.

2.3.4.7 Noise source motion effects

The effects of motion on jet noisediffer from speed effects on other noisesources, and hence are consideredseparately during static-to-flightprojection.

2.3.4.7.1 External jet noise

Account should be taken of thefrequency-dependent jet relativevelocity effects and convectiveamplification effects. Broadly, twosources of information may be used todevelop an approved method fordefining the effect of flight onexternal jet noise:

a) For single-stream engineshaving circular exhaustgeometries, SAE ARP 876C-1985 provides guidance.However, additionalsupporting evidence may beneeded to show when jetnoise is the major contributorto the noise from an enginewith a more complex nozzleassembly; and

b) Full scale flight data on asimilar exhaust geometry canprovide additional evidence.In general, however, becauseof the difficulty of defininghigh frequency effects in thepresence of internally-generated engine noise, itmay be necessary to provideadditional supportinginformation to determine thevariation of EPNL, withchanges of jet noise spectraat high frequencies.

2.3.4.7.2 Noise sources otherthan jet noise

In addition to the Dopplerfrequency effect on the non-jet noiseobserved on the ground from anaeroplane flyover, the noise generatedby the engine's internal componentsand the airframe can be influenced bysource amplitude modification, anddirectivity changes:

a) Doppler Effect: frequencyshifting that results frommotion of the source(aeroplane) relative to amicrophone is accounted forby the following equation:

( )ff

Mflightstatic=

−1 cosλ

where:

fflight = flight frequency;

fstatic= static frequency;

M = Mach number ofaeroplane; and

λ = angle between theflight path in thedirection of flight and astraight line connectingthe aeroplane and themicrophone at the timeof sound emission.

It should be noted for those1/3 octave band sound

jskaleck



21

pressure levels dominated bya turbo-machinery tone, theDoppler shift may move thetone (and its harmonics) intoan adjacent band.

b) Source amplitudemodification and directivity

changes: sound pressurelevel adjustments to airframegenerated noise that resultfrom speed changes betweenthe datum and derivativeversion is provided for inparagraph 2.3.4.9, Airframenoise.

For noise generatedinternally within the engine,e.g. fan noise, there is noconsensus of opinion on themechanisms involved or aunique adjustment methodthat accounts for the detailedsource modification andsound propagation effects.

If an adjustment is used, thesame technique must beapplied to both the flightdatum and derivativeconfiguration whenestablishing noise changes.In such instances theadjustment for soundpressure level changes thatresult form the motion of thesource (aeroplane) relative tothe microphone may beaccounted for using theequation:

( )SPL SPL K Mcosflight static= − −log 1 λ

where:

SPLflight = flight sound

pressure level;

SPLstatic= static soundpressure level; and

M and λ are definedabove and K is aconstant.

Theoretically K has a valueof 40 for a point noise source

but a more appropriate valuemay be obtained bycomparing static and flightdata for the flight datumaeroplane.

2.3.4.8 Aeroplane configuration effects

2.3.4.8.1 The contribution frommore than one engine on an aeroplaneis normally taken into account byadding 10 log N, where N is thenumber of engines, to eachcomponent noise source. However, itmight be necessary to compute thenoise from engines widely spaced onlarge aeroplanes particularly in theapproach case if they include bothunderwing and fuselage mountings.The noise from the intakes of enginesmounted above the fuselage is knownto be shielded.

2.3.4.8.2 If engine installationeffects change between the flightdatum aeroplane and a derivedversion, account should be taken ofthe change on sound pressure levelswhich should be estimated accordingto the best available evidence.

2.3.4.9 Airframe noise

2.3.4.9.1 To account for thecontribution of airframe noise,measured flight datum airframe noiseon its own or combined with anapproved airframe noise analyticalmodel may be used to develop anairframe noise data base. Theairframe-generated noise, which canbe treated as a point source foradjustment purposes, is normalised tothe same conditions as those of theother (engine) sources, with dueaccount for the effects of sphericaldivergence, atmospheric absorptionand airspeed as described in section 8and 9 of Appendix 2 of Annex 16,Volume 1.

2.3.4.9.2 Airframe noise for aspecific configuration varies withairspeed (see Reference 3) as follows:

( )∆SPL V Vairframe REF TEST= 50 log

where:

jskaleck



22

VREF is approved referenceairspeed for the flight datumaeroplane; and

VTEST is model or measuredairspeed.

2.3.4.9.3 The above equation isalso valid for adjustments to EPNLwhere an empirically derivedcoefficient replaces the coefficient 50since that number may be somewhatconfiguration dependent. However,approval of the certification authorityis required for values other than 50.

2.3.4.10 Aeroplane flight pathconsiderations

When computing noise levelscorresponding to the slant distance of theaeroplane in flight from the noisemeasuring point, the principal effects arespherical divergence (inverse square lawadjustments from the nominal staticdistance) and atmospheric attenuation, asdescribed in sections 8 and 9 of Appendix2 of Annex 16, Volume 1. Further,account should be taken of the differencebetween the static engine axis and that axisin flight relative to the reference noisemeasuring points. The adjustments shouldbe applied to the component noise sourcelevels that have been separately identified.

2.3.4.11 Total noise spectra

2.3.4.11.1 Both the engine tonaland broadband noise sourcecomponents in flight, as discussedearlier, together with the airframenoise and any installation effects, aresummed on a mean-square pressurebasis to construct the spectra of totalaeroplane noise levels.

2.3.4.11.2 During the merging ofbroadband and tonal componentsconsideration should be given toappropriate band-sharing of discretefrequency tones.

2.3.4.11.3 The effects of groundreflections must be included in theestimate of freefield sound pressurelevels to simulate the sound pressurelevels that would be measured by a

microphone at a height of 1.2 mabove a natural terrain. Informationin SAE AIR 1672B-1983 orEngineering Science Data Unit,ESDU data item 80038 AmendmentA may be used to apply adjustmentsto the freefield spectra to allow forflight measurements being made at1.2 m (4 ft). Alternatively, the groundreflection correction can be derivedfrom other approved analytical orempirically derived models. Note thatthe Doppler correction for a staticsource at frequency fstatic applies to amoving (aeroplane) source at afrequency fflight where

( )f f 1-flight static= M cosλ using

the terminology of 2.3.4.7.2(a).

2.3.4.11.4 This process isrepeated for each measurement angleand for each engine power setting.

2.3.4.11.5 With regard to lateralattenuation, information in SAE AIR1751-1981 applicable to thecomputation of lateral noise may beapplied.

2.3.4.12 EPNL computations

For EPNL calculations, a time isassociated with each extrapolatedspectrum along the flightpath. (Note:Time is associated with each measurementlocation with respect to theengine/aeroplane reference point and theaeroplane's true airspeed along thereference flight path assuming zero wind).For each engine power setting andminimum distance, an EPNL is computedfrom the projected time history using themethods described in Annex 16, Volume1, Appendices 1 and 2.

2.3.4.13 Changes to noise levels

2.3.4.13.1 An NPD plot can beconstructed from the projected staticdata for both the original (flightdatum) and the changed versions ofthe engine or nacelle tested.Comparisons of the noise v's enginepower relationships for the twoconfigurations at the same appropriateminimum distance, will determinewhether or not the changedconfiguration resulted in a change to

jskaleck



23

the noise level from an engine noisesource. If there is a change in thelevel of source noise, a new in-flightaeroplane NPD plot can be developedby adjusting the measured originalNPD plot by the amount of changeindicated by the comparison of thestatic-projected NPD plots for theoriginal and changed versions withinthe limitations specified in 2.3.2 forEffective Perceived Noise Level.

2.3.4.13.2 The noise certificationlevels for the derived version may beobtained by entering the NPD plots atthe relevant reference engine powerand distance.

jskaleck



24


jskaleck



25

SECTION 3 - EQUIVALENT PROCEDURES FOR PROPELLER DRIVEN AEROPLANES OVER9000 kg

The following procedures have been used asequivalent in stringency to Chapter 3 and Chapter5, Annex 16, Volume 1 for propeller drivenaeroplanes with maximum certificated take-offmass exceeding 9000 kg.


3.1.1 Flight path intercept procedures

Flight path intercept procedures, asdescribed in 2.1.1 of this manual, have beenused to meet the demonstration requirements ofnoise certification in lieu of full take-offsand/or landings.

3.1.2 Generalised flight test procedures

Generalised flight test procedures (otherthan normal noise demonstration take-offs andapproaches) have been used to meet twoequivalency objectives:

a) to acquire noise data over a range ofengine power settings at one or moreheights: this information permits thedevelopment of generalised noisecharacteristics necessary forcertification of a "family" of similaraeroplanes. The procedures used aresimilar to those described in 2.1.2.1with the exception that the NPD plotsemploy engine noise performanceparameters (µ) of MH (propellerhelical tip Mach number) andSHP ambδ (shaft horse power) (see

Fig 3), where δamb is defined in2.1.2.1.3.

In order to ensure that propellerinflow angles are similar throughoutthe development of the noise-sensitivity data as the aeroplane masschanges, the airspeed of the aeroplaneused in the flight tests for developingthe lateral and flyover data shall beV2 + 19 km/h (V2 + 10 kt) to within±6 km/h or ±3 kt appropriate for themass of the aeroplane during the test.

For the development of the NPD datafor the approach case the speed andapproach angle constraints imposedby 5.6.3(b), 5.7.5, 3.6.3 and 3.7.5 of

Chapters 5 and 3 of Annex 16,Volume 1, cannot be satisfied over thetypical range of power needed. Forthe approach case the speed shall bemaintained at 1.3 VS + 19 km/h(1.3 VS + 10 kt) to within ±6 km/h or±3 kt and flyover height over themicrophone maintained at 400 ft±100 ft. However the approach angleat the test power shall be that whichresults from the aeroplane conditions,ie. mass, configuration, speed andpower; and

b) Noise level changes determined bycomparisons of fly-over noise testdata for different developments of anaeroplane type, for example, a changein propeller type, have been used toestablish certification noise levels of anewly derived version as described in2.1.2.2.

3.1.3 Determination of the lateral noisecertification level

Determination of the lateral noisecertification level employing an alternativeprocedure using two microphone stationslocated symmetrically on either side of thetake-off flight path similar to that as describedin 2.1.3 has been approved. However, whenthis procedure is used, matching data from bothlateral microphones for each fly-by must beused for the lateral noise determination; caseswhere data from only one microphone isavailable for a given fly-by must be omittedfrom the determination. The followingparagraphs describe the procedures forpropeller-driven heavy aeroplanes.

3.1.3.1 The lateral EffectivePerceived Noise Level from propellerdriven aeroplanes when plotted againstheight opposite the measuring sites canexhibit distinct asymmetry. The maximumEPNL on one side of the aeroplane isoften at a different height and noise levelfrom that measured on the other side.

3.1.3.2 In order to determine theaverage maximum lateral EPNL, i.e. thecertification sideline noise level, it istherefore necessary to undertake a numberof flights over a range of heights to define

jskaleck



26

the noise versus height characteristics foreach side of the aeroplane. A typicalheight range would cover between 30 m(100 ft) and 550 m (1,800 ft) above a trackat right angles to and midway along theline joining the two microphone stations.The inter-section of the track with this lineis defined as the reference point.

3.1.3.3 Since experience has shownthe maximum lateral noise level may oftenbe near the lower end of this range aminimum of six good sets of data,measured simultaneously from both sidesof the flight track, should be obtained for arange of aeroplane heights as low aspossible. In this case take-offs may benecessary, however care should be takento ensure that the airspeed is stabilised toat least V2 + 19 km/h (V2 + 10 kt) overthe 10 dB-down time period.

3.1.3.4 The aeroplane climbs over thereference point using take-off power,speeds and configuration as described in3.6.2.1 c) and d) of Chapter 3 or 5.6.2.1 c)and d) of Chapter 5 of Annex 16, Volume1.

3.1.3.5 The lateral certification noiselevel is obtained by finding the peak of thecurve of noise level (EPNL) corrected toreference day atmospheric absorptionvalues, plotted against aeroplane heightabove the reference point (see Fig. 9).This curve is described as a least squarescurve fit through the data points definedby the median values of each pair ofmatched data measured on each side of thetrack (i.e. the average of the twomicrophone measurements for a givenaeroplane height).

3.1.3.6 To ensure that therequirements of 5.4.2 of Appendix 2 or5.5.2 of Appendix 1 of Annex 16, Volume1, are met the 90 per cent confidencelimits should be determined in accordancewith paragraph 2.2 of Appendix 1 of thismanual.

3.1.4 Measurements at non-referencepoints

3.1.4.1 In some instances testmeasurement points may differ from thereference measurement points specified inChapters 3 and 5 of Annex 16, Volume 1.Under these circumstances an applicant

may request approval of data that havebeen adjusted from actual measurementsto the reference conditions for reasonsdescribed in 2.1.5.2 a), b) and c).

3.1.4.2 Noise measurements collectedcloser to the test aeroplane than at thecertification reference points areparticularly useful for correcting propellernoise data as they are dominated by lowfrequency noise. The spectra rolls offrapidly at higher frequencies and is oftenlost in the background noise at frequenciesabove 5000 Hz. Appendix 3 describes analternative procedure.

3.1.4.3 Non-reference measurementpoints may be used provided thatmeasured data are adjusted to referenceconditions in accordance with therequirements of section 9 of Appendix 1or 2 of Annex 16, Volume 1 and themagnitude of the adjustments does notexceed the limits in paragraphs 3.7.6 ofChapter 3 and 5.7.6 of Chapter 5 of theAnnex.

3.1.5 Reference approach speed

The reference approach speed is currentlycontained in 3.6.3.1(b) of Chapter 3 of Annex16, Volume 1 as 1.3 VS + 19 km/h(1.3 VS + 10 kt). There is a change beingmade to the definition of stall speed, forairworthiness reasons, to alter the currentminimum speed VS definition to a stall speedduring a 1-g manoeuvre (i.e. a flight load factorof unity) VS1g. In terms of the new definitionthe approach reference speed becomes1.23 VS1g + 19 km/h, (1.23 VS1g + 10 kt)which can be taken as equivalent to thereference speed contained in Chapter 3.

3.2 ANALYTICAL PROCEDURES

3.2.1 Analytical equivalent proceduresrely upon available noise and performance datafor the aeroplane type. Generalisedrelationships between noise levels, propellerhelical tip Mach number, and shaft horsepowerand correction procedures for speed and heightchanges in accordance with the methods ofAppendix 2 of Annex 16, Volume 1, arecombined with certificated aeroplaneperformance data to determine noise levelchanges resulting from type design changes.The noise level changes are then added to orsubtracted from the noise certification levels

jskaleck



27

demonstrated by flight test measurements forthe flight datum aeroplane.

3.2.2 Certifications using analyticalprocedures have been approved for type designchanges that result in predictable noise leveldifferences including the following:

a) an increase or decrease in maximumtake-off and/or landing mass of theaeroplane from the originallycertificated mass;

b) power increase or decrease forengines that are acoustically similarand fitted with propellers of the sametype;

c) aeroplane, engine and nacelleconfiguration changes, usually minorin nature, including derivativeaeroplane models with changes infuselage length and flap configuration.However, care is needed to ensurethat existing noise sources are notmodified by these changes, e.g. bychanging the flow field into thepropellers; and

d) minor airframe design changes thatcould indirectly affect noise levelsbecause of an impact on aeroplaneperformance (increased drag forexample). Changes in aeroplaneperformance characteristics derivedfrom aerodynamic analysis or testinghave been used to demonstrate howthese changes affect the aeroplaneflight path and hence thedemonstrated noise levels of theaeroplane.

3.3 GROUND STATIC TESTING PROCEDURES

3.3.1 General

Unlike the case of a turbojet or turbo fanpowerplant, static tests involving changes tothe propeller are not applicable to determiningnoise level changes in the development of apropeller driven aeroplane/powerplant family.This is caused by changes in the aero-acousticoperating conditions of the propeller when runstatically compared with conditions existingduring flight. The propeller noise levelsmeasured during a static test can includesignificant contributions from noise sourcecomponents not normally important in flight.However, limited static tests on engines with

propellers, which are used as engine loadingdevices can be utilised to determine smallnoise changes, as described below.

3.3.2 Guidance on the test sitecharacteristics

Guidance on the test site characteristicsdata acquisition and analysis systems,microphone locations, acoustical calibrationand measurement procedures for static testingis provided in SAE ARP 1846 and is equallyvalid in these respects for propeller powerplants.

3.3.3 Static tests of the gas generator

3.3.3.1 Static tests of the gasgenerator can be used to identify noisechanges resulting from changes to thedesign of the gas generators or the internalstructure of the engine in the frequencyranges where there is a contribution to theaeroplane EPNL, or where that part of thespectrum is clearly dominated by the gasgenerator or ancillary equipment undercircumstances where the propeller and itsaerodynamic performance remainsunchanged.

3.3.3.2 Such circumstances include,for example, changes to the compressor,turbine or combustor of the powerplant.The effect of such changes should beconducted under the same test,measurement, data reduction andextrapolation procedures as described inparagraph 2.3 for turbojet and turbofanengines. The noise from any propeller orother power extraction device used instatic tests should be eliminated orremoved analytically. For the purposes ofaeroplane EPNL calculation, the measuredflight datum aeroplane propellercontributions should be included in thecomputation process.

jskaleck



28


jskaleck



29

SECTION 4 - EQUIVALENT PROCEDURES FOR PROPELLER DRIVEN AEROPLANES NOTEXCEEDING 9000 kg

The following procedures have been used asequivalent in stringency to Chapter 6 and Chapter10 of Annex 16, Volume 1 for propeller-drivenaeroplanes with maximum certificated take-offmass not exceeding 9000 kg.

4.1 SOURCE NOISE ADJUSTMENTS

Source noise adjustment data for propellerdriven light aeroplanes may be established by flyingthe test aeroplane with a range of propeller speeds(for fixed propellers) and torque or manifold airpressure (MAP) values (for variable pitchpropellers).

4.1.1 Fixed pitch propellers

For aeroplanes fitted with fixed pitchpropellers source noise sensitivity curves aredeveloped from data taken by measuring thenoise level for the aeroplane flying at 300 m(985 ft) as described in paragraph 6.5.2 ofAnnex 16, Volume 1 at the propeller speed formaximum continuous power NMCP.Aeroplanes demonstrating compliance withAnnex 16, Chapter 10, should be flownaccording to paragraph 2.3 of Appendix 6 ofAnnex 16, Volume 1 such that the aircraftoverflys the microphone at the reference heightHREF (defined in paragraph 5.1 of Appendix 6),the best rate of climb speed Vy and at thepropeller speed, NMAX, corresponding to thatdefined in paragraph d) of Second Phase ofparagraph 10.5.2 of Annex 16. Noisemeasurements are repeated at two lowerpropeller speeds typically 200 rpm and400 rpm lower than NMCP or NMAX. ForChapter 10 aeroplanes these should be flown atspeed Vy. The maximum A-weighted noisepeak noise level (LAmax) is plotted againstpropeller helical tip Mach number (MH) toobtain the curve from which the source noisecorrection may be derived.

For fixed pitch propellers it is generally notpossible via flight tests to separate out the twosignificant noise generating parameters, helicaltip Mach number and power absorbed by thepropeller. A sensitivity curve of Mach numberversus noise level derived from flight tests of afixed pitch propeller (either level flyovers orfixed speed climbs) will therefore includewithin it the effects not only of Mach numberbut also power. Under these circumstances it

is not appropriate to apply a separate powercorrection.

4.1.2 Variable pitch propellers

4.1.2.1 For variable pitch propellersthe data is taken with the aircraft flyingover a range of propeller speeds, typicallythree, at a fixed torque or MAP in amanner similar to that described in 4.1.1where NMCP or NMAX would, in this case,be the maximum propeller speed at themaximum permitted torque or MAP. Thisis repeated for two lower torque or MAPvalues to establish a carpet plot ofmaximum A-weighted noise levels againstpropeller speed and torque, MAP or shafthorse power (SHP).

4.1.2.2 A plot of maximum A-weighted noise level (LAmax), helical tipMach number (MH) and torque or MAP isdeveloped which is used to derive thesource noise adjustment ( LAmax) beingthe difference between reference and testconditions at the noise certification power.

4.1.2.3 Generally the test andreference engine SHP can be derived fromthe engine manufacturer's performancecurves. However, where such curves arenot available a correction should beapplied to the manufacturer's publishedengine SHP (normally presented for arange of engine speeds under ISA and sealevel conditions) to establish the enginepower level under the test conditions ofambient temperature and air density, asfollows:

( ) ( )[ ]P P T TT R R T=

−1

2 0117 0 883σ . / .

for normally aspirated engines; and

( )P P T TT R R T=

12

for turbo-charged engines,

where PT and PR are the test and referenceengine powers, TT and TR are the test andreference ambient temperatures, and σ is

jskaleck



30

the air density ratio. Note that in thiscontext reference denotes the referenceconditions for which the engine SHP isknown.

4.2 TAKE-OFF TEST AND REFERENCE

PROCEDURES

In planning a test programme for noisecertification under Chapter 10 and Appendix 6, it ishelpful to note the differences between test dayflight procedures and the standardised take-offreference profile.

4.2.1 The take-off reference profile isused to compute the altitude and speed of theaircraft passing over the microphone on astandard day. The requirements for this profileare contained in paragraph 10.5.2 of Chapter10. They require that the first segment becomputed, using airworthiness approved data,assuming take-off power is used from the brakerelease point to 15 m (50 feet) above therunway. The second segment is assumed tobegin precisely at the end of the first segment,with the aeroplane in a climb configuration(gear up and climb flaps) and operating at thecertificated speed for best rate of climb Vy.

A worked example of the calculation ofreference flyover height and referenceconditions for correction of source noise foraeroplanes certificated according to thestandards of Annex 16, Chapter 10 is presentedin Appendix (5) of this manual.

4.2.2 Requirements for aeroplane testprocedures are contained in two places:Section 10.6 of Chapter 10 and Section 2.3 ofAppendix 6. They basically only speak to testtolerances and approval of test plans bycertificating authorities.

4.2.3 Figure 13 illustrates the differencebetween the two. Note that the actual flighttest path need not include a complete take-offfrom a standing condition. Rather, it assumesthat a flight path intercept technique is used.As with the turbojet and helicopter standards,the aeroplanes would be flown to intersect thesecond phase (segment) climb path at the rightspeed and angle of climb when going over themicrophone within 20 per cent of the referenceheight.

jskaleck



31

SECTION 5 - EQUIVALENT PROCEDURES FOR HELICOPTERS

The objective of a noise certificationdemonstration test is to acquire data forestablishing an accurate and reliable definition of ahelicopter's noise characteristics (see section 8.7 ofAnnex 16, Volume 1). In addition, the Annexestablishes a range of test conditions andprocedures for adjusting measured data to referenceconditions.


5.1.1 Noise certification guidance

Noise certification of helicopters has beenrequired fairly recently and many applicantsmay find that the current standards andprocedures differ from those that they haveused hitherto. The following paragraphs aredesigned to provide clarification of therequirements as set out in Chapter 8 andAppendix 2 of Annex 16, Volume 1.

5.1.1.1 Helicopter test window forzero adjustment for atmosphericattenuation

5.1.1.1.1 There is currently a"test window" contained in Annex 16,Volume 1 (paragraph 2.2.2 ofAppendix 2) which needs to be metbefore test results are acceptable tocertificating authorities. In addition ifthe test conditions fall within a “zeroattenuation adjustment window”(Figure 16) defined as the areaenclosed by 2°C, 95% RH; 30°C,95%RH; 30°C, 35% RH; 15°C, 50%RH; and 2°C, 90% RH, theatmospheric attenuation adjustment ofthe test data may be taken as zero.That is the terms

( ) ( )[ ]0 01 0. α αi i QK− and

( ) ( )0 01 0. α i QK Q Kr r− from the

equation for ( )SPL i r in paragraph

8.3.1 of Appendix 2 of Annex 16,Volume 1, become zero and theadjustment becomes:

( ) ( ) ( )SPL i SPL i QK Q Kr r r= + 20log

In addition, provided that all themeasured points for a particular flightcondition are within the “zero

attenuation adjustment window”defined in Figure 16 and are withinthe appropriate height tolerances forflyover of ±9 m (±29.5 ft), forapproach ±10 m (±32.8 ft) and the 2EPNdB limit on the take-offadjustment for height given inparagraph 8.7.4a of Chapter 8 ofAnnex 16, Volume 1, the ratios of thereference and test slant distances forthe propagation path adjustments maybe replaced by the ratios of thereference and test distances to thehelicopter when it is over the centrenoise measuring point.

The total effect of both simplificationsis that the equation of paragraph 8.3.1of Appendix 2 of Annex 16, Volume1 becomes:

( ) ( ) ( )SPL i SPL i HK H Kr r r= + 20log

and the duration adjustment specifiedin paragraph 8.4.2 of Appendix 2 ofAnnex 16, Volume 1, becomes:

( ) ( )∆ 2 7 5 10= − +. log logHK H K V Vr r r

where HK is the measured distancefrom the helicopter to the noisemeasuring point when the helicopteris directly over the centre noisemeasuring point and HrKr is thereference distance.

5.1.1.2 Helicopter test speed

5.1.1.2.1 There are tworequirements on helicopter testspeeds. Firstly, the airspeed duringthe 10 dB-down time period should beclose to, ie. within 9 km/h (5 kt), ofthe reference speed (see 8.7.6 ofVolume 1 of Annex 16) to minimisespeed adjustments for the threecertification conditions of take-off,flyover and approach.

5.1.1.2.2 The second speedrequirement applies to the flyovercase (see 8.7.7 of Volume 1 of Annex16). When the absolute wind speedcomponent in the direction of flight

jskaleck



32

exceeds 9 km/h (5 kt) the leveloverflights shall be made in equalnumbers with a head wind componentand tail wind component. Theobjective is to counter the effect of theshort duration correction obtained inthe downwind direction with thelarger one into wind. In steady windconditions the net effect of flying intoand with the wind would be zero.

5.1.1.2.3 The measurement ofgroundspeed may be obtained bytiming the helicopter as it passes overtwo points a known distance apart onthe helicopter track during theoverflight noise measurements. Thesetwo points should straddle the noisemeasurement microphone array.

5.1.1.3 Test speed for lighthelicopters

For the purposes of compliance withChapter 11 of Annex 16, Volume 1, thehelicopter should be flown at test speedVar which will produce the sameadvancing blade Mach number as thereference speed in reference conditionsgiven in paragraphs 11.5.1.4 and 11.5.2 b)of Chapter 11 of Annex 16.

The reference advancing blade Machnumber MR is defined as the ratio of thearithmetic sum of the blade tip rotationalspeed Vtip and the helicopter true airspeedVREF divided by the speed of sound, c, at

25 °C (346.1 m/sec) such that:

MV V

cRTIP REF= +

The test airspeed VAR is calculated from:

V cV V

cVAR T

TIP REFTIP=

+

−

where: cT is the speed of sound obtainedfrom the onboard measurements of outsideair temperature.

Since the ground speed obtained from theoverflight tests will differ from that forreference conditions an adjustment ∆2 ofthe form:

( )∆2 10= log V Var ref

will need to be applied. ∆2 is theincrement in decibels that must be addedto the measured sound exposure level(SEL).

5.1.1.4 Helicopter test mass

5.1.1.4.1 The mass of thehelicopter during the noisecertification demonstration (see8.7.11 of Volume 1 of Annex 16)must lie within the range 90 per cent - 105 per cent of the maximum take-offmass for the take-off and flyover andbetween 90 per cent - 105 per cent ofthe maximum landing mass for theapproach demonstration. For noisecertification purposes the effect ofchange of mass is to change the testday flight path for take-off andadjustments to the reference flightpath should be made for sphericalspreading and atmospheric attenuationas described in section 9 of Appendix2, Annex 16, Volume 1.

5.1.1.4.2 In some cases, such aswhen the test aircraft weight isrestricted to a value somewhat lessthan the anticipated final certificationweight, the applicant may, subject tothe approval of the certificatingauthority, apply specific correctionsfor weight variations. The applicantmay be approved to use a 10 logrelationship correction or otherwisedetermine, by flight test, the variationof EPNL with weight. In such a casethe weights tested should include themaximum allowable test weight.

Note: A similar correction proceduremay be acceptable when thecertificated weight is increased by asmall amount subsequent to the flighttests.

5.1.1.5 Helicopter approach

Section 8.7.10 of Chapter 8 in Annex16 constrains the approach demonstrationto within ±0.5° of the reference approachangle of 6°. Adjustments to the referenceapproach angle are required to account forspherical spreading effects andatmospheric attenuation as described in

jskaleck



33

section 9 of Appendix 2, Annex 16,Volume 1.

5.1.1.6 Helicopter flight pathtracking

Annex 16, Volume 1, Appendix 2,Section 2.3 requires that the helicopterposition relative to the flight pathreference point be determined andsynchronised adequately with the noisedata between the 10 dB-down points.

Methods which have been used include:

a) radar or microwave trackingsystem;

b) theodolite triangulation; and

c) photographic scaling

These techniques may be used singly or incombination. Practical examples ofaircraft tracking systems employing one ormore of these techniques, are describedbelow. This material is not intended to bean exhaustive list and additionalinformation will be included as moreexperience is acquired.

5.1.1.6.1 Radar or microwavetracking system

One example of a radar positiontracking system is shown in Figure 14.It operates on a principle of the pulseradar with a radar interrogator(receiver/transmitter) located on theaircraft and a radar transponder(reference/station) positioned at eachreference station. The elapsed timebetween the receiver/transmitter pulseand reception of the pulse returnedfrom the reference station transponderis used as the basis for determiningthe range of each reference station.

This range information together withthe known location of the referencestations can be used to obtain a fix onthe position of the aircraft in threedimensions. A pulse coding system isemployed to minimise false returnscaused by radar interference onreflected signals.

The system performs the followingbasic functions during noisecertification:

a) continuously measures thedistance between thehelicopter and four fixedground sites;

b) correlates these ranges withIRIG-B time code and heightinformation and outputs thisdata to a PCM recorder;

c) converts the aircraft rangeand height information intoX, Y and Z position co-ordinates in real time; and

d) uses the X, Y, Z data to drivea cockpit display providingthe pilot with steering andposition cueing.

The accuracy of the co-ordinatecalculation depends on the flight pathand transponder geometry. Errors areminimised when ranges intersect andthe recommended practice is to keepthe intersection angle near to 90°.The four transponder arrangementsshown in Figure 14 produce positionuncertainties from ±1.0 to ±2.0 m.

Some inaccuracies at low aircraftheights can be introduced with the useof microwave systems and the use of aradio-altimeter can reduce the errors.The height data is recorded andsynchronised with the microwave.

Helicopter equipment: The distancemeasuring unit computer andtransponder beacon are connected to ahemispherical antenna which ismounted under the fuselage, on theaircraft centreline, as close to thehelicopter centre of gravity aspossible.

Ground equipment: The four beaconsare located on either side of theaircraft track to permit the optimumlayout, ie. covering the helicopterwith angles between 30° and 150° (90° being the ideal angle).

For example, two beacons can belocated in the axis of the noise

jskaleck



34

measurement points at ±500 m ofcentral microphone, another twobeacons can be located under track at±600 m from the central microphone.

5.1.1.6.2 Kine-theodolite system

It is possible to obtain helicopterposition data with classical kine-theodolites, but it is also possible tomake use of a system composed oftwo simplified theodolites including amotorised photocamera on a movingplatform reporting site and elevation.These parameters are synchronisedwith coded time and the identificationnumber of every photograph recorded.

Each 0.1 s site and elevation data aresent by UHF to a central computerwhich calculates the helicopterposition X, Y, Z, versus time for eachtrajectory.

Photography stations are located atsideline positions, about 300 m fromthe track, and at 200 m either side ofthe 3 noise measurement points.

The accuracy of such a system can be±1.5 m in (X, Y, Z) over the workingarea.

5.1.1.6.3 Radar / theodolitetriangulation

The opto-electronic systemshown diagramatically in Figure 15uses a single optical theodolite toprovide azimuth and elevation whilerange data are obtained from a radartracking system using a singletransponder. Data from these twosources are transferred to a desk topcalculator at a rate of 20samples/second from which threedimensional position fixes can bederived. The system also providestape start and stop times to themeasuring sites, synchronising all taperecording times. Accuracy of thesystem is approximately ±2.0 m,±1.0 m and ±2.0 m for horizontalrange (x), cross-track (y) andheight (z) respectively. Uncertaintiesassociated with determination of thevisual glide slope indicator andground speed are ±0.1° and ±0.5 kt.

5.1.1.6.4 Photographic scaling

The flight path of the helicopterduring the noise certificationdemonstration may be determined bythe use of a combination of groundbased cameras and height datasupplied as a function of time fromthe on-board radio or pressurealtimeters.

In this method three cameras areplaced along the intended track suchthat one is sited close to the centremicrophone position and the othertwo sited close to each of the 10 dB-down points typically 500 m eitherside of the microphone, dependingupon the flight procedure being used.The cameras are mounted verticallyand are calibrated so that the imagesize, obtained as the helicopter passesoverhead, can be used to determinethe height of the aircraft. It isimportant that the time at which eachcamera fires is synchronised with theon-board data acquisition system sothat the height of the aircraft as itpasses over each of the cameras, canbe correlated with the heightsobtained from the photographs.

The flight path of the helicopter as afunction of distance may be obtainedby fitting the aircraft data to thecamera heights.

The aircraft reference dimensionshould be as large as possible tomaximise photograph image size butshould be chosen and used with careif errors in aircraft position are to beavoided. Foreshortening of the imagedue to main rotor coning (bending ofthe blades), disc tilt or fuselage pitchattitude if not accounted for will resultin over estimates of height, lateral andlongitudinal offsets.

By erecting a line above each of thecameras at right angles to the intendedtrack, at a sufficient height above thecamera to provide a clearphotographic image of both the lineand the helicopter, the applicant mayobtain the lateral offset of thehelicopter as it passes over each of thecameras. This can be done byattaching marks to the line showing

jskaleck



35

the angular distances from overheadat 5° intervals either side of thevertical.

This method may be used to confirmthat the helicopter follows a 6° ±0.5°glideslope within 10° of overhead thecentre microphone as required byAnnex 16, Volume 1, Chapter 8,paragraph 8.7.10 and 8.7.8.

Further, from the synchronised timesof the helicopter passing over thethree camera positions the groundspeed can be determined for later usein the duration correction adjustment.

Overall accuracy of the system is±1.0 per cent of height and±1.3 per cent of longitudinal andlateral displacements. Meanapproach/climb angles and meanground speed can be determinedwithin ±0.25° and ±0.7 per centrespectively.

5.1.1.7 Atmospheric test conditions

The temperature, relative humidityand wind velocity limitations arecontained in Appendix 2 of Annex 16,Volume 1 (see 2.2.2). The parameters aremeasured at 10 m (33 ft). For adjustmentpurposes the measured values of theseparameters are assumed to berepresentative of the air mass between thehelicopter and the microphones. Nocalculation procedures based on thedivision of the atmosphere into layers arerequired, but such a method of analysiscould be accepted by the certificatingauthority.

5.1.1.8 Procedure for thedetermination of source noise correction

5.1.1.8.1 For the demonstrationof overflight reference noisecertification levels off-referenceadjustments shall normally be madeusing a sensitivity curve of PNLTMversus advancing blade tip Machnumber deduced from flyovers carriedout at different airspeeds around thereference airspeed; however, theadjustment may be made using analternative parameter, or, parameters,approved by the certificatingauthority. If the test aircraft is unable

to attain the reference value ofadvancing blade tip Mach number orthe agreed reference noise correlatingparameter, then an extrapolation ofthe sensitivity curve is permittedproviding that the data cover a rangeof values of the noise correlatingparameter agreed by the certificatingauthority between test and referenceconditions. The advancing blade tipMach number or agreed noisecorrelating parameter shall becomputed from as measured datausing true airspeed, onboard outsideair temperature (OAT) and rotorspeed. A separate curve of sourcenoise versus advancing blade tip machnumber or another agreed noisecorrelating parameter shall be derivedfor each of the three certificationmicrophone locations i.e. centreline,sideline left and sideline right.Sidelines left and right are definedrelative to the direction of the flighton each run. PNLTM adjustments areto be applied to each microphonedatum using the appropriate PNLTMfunction.

5.1.1.8.2 In order to eliminatethe need for a separate source noisecorrection to the overflight test resultsthe following test procedure isconsidered acceptable when thecorrelating parameter is the main rotoradvancing blade tip Mach number.

Each overflight noise test must beconducted such that:

a) The adjusted reference trueair speed (VAR) is thereference airspeed (VR)specified in Section 8.6.3 ofChapter 8 of Annex 16,Volume 1, adjusted asnecessary to produce thesame main rotor advancingblade tip Mach number asassociated with referenceconditions;

Note: The reference advancingblade tip Mach number (MR) isdefined as the ratio of thearithmetic sum of the main rotorblade tip rotational speed (VTIP)and the helicopter referencespeed (VR) divided by the speed

jskaleck



36

of sound (c) at 25 ° C (346.1 m/s)such that :

MV V

cRTIP REF= +

,

and the adjusted referenceairspeed ((VAR) is calculatedfrom:

V cV V

cVAR T

TIP REFTIP=

+

− ,

where cT is speed of sound fromthe onboard measurement ofoutside air temperature (seeParagraph 6.7).

b) The test true airspeed (V)shall not vary from theadjusted reference trueairspeed (VAR) by more than ±5 km/h (±3 kt) or anequivalent approvedvariation from the referencemain rotor advancing bladetip Mach number; and

c) In practice the tests will beflown to an indicatedairspeed which is theadjusted reference trueairspeed (VAR) corrected forcompressibility effects andinstrument position errors.

d) The onboard outside staticair temperature must bemeasured at the overflightheight just prior to eachflyover.

The calculation of noise levels,including the corrections, is the sameas that described in Chapter 8 andAppendix 2 of Annex 16, Volume 1,except that the need for source noiseadjustment is eliminated. It should beemphasised that in the determinationof the duration correction (∆2), thespeed adjustment to the durationcorrection is calculated as10 log(Vg/Vgr), where Vg is the testground speed and Vgr is the referenceground speed.

5.1.2 On board flight data acquisition

5.1.2.1 It is necessary to obtain thevalues of a variety of flight and engineparameters during the noise measurementperiod in order to:

a) determine the acceptability ofhelicopter noise certificationflight tests;

b) obtain data to adjust noise data;and

c) to synchronise flight, engine andnoise data.

Typical parameters would includeairspeed, height/altitude, rotor speed,torque, time etc.

5.1.2.2 A number of methods forcollecting this information have beenemployed:

a) manual recording;

b) magnetic tape recording;

c) automatic still photographicrecording;

d) cine recording; and

e) video recording.

5.1.2.3 Clearly, when a large numberof parameters are required to be collectedat relatively short time intervals, it may notbe practicable to manually record the dataand the use of one of the automaticsystems listed as 5.1.2.2 b) to e) becomesmore appropriate. The choice of aparticular system may be influenced by anumber of factors such as the spaceavailable, cost, availability of equipmentetc.

5.1.2.4 For systems which opticallyrecord the flight deck instruments (5.1.2.2c) to e)) care must be taken to avoid stronglighting contrast, such as would be causedby sunlight and deep shadow, andreflections from the glass fronts ofinstruments which would make dataunreadable. To avoid this it may benecessary to provide additional lighting to"fill in" deep shadow regions. To preventreflections from the front of instruments itis recommended that light coloured

jskaleck



37

equipment or clothing on the flight deck isavoided. Flight crews should be requiredto wear black or dark coloured clothingand gloves.

5.1.2.5 Further, for systems whichrecord the readings of dials it is importantthat the recording device is as near aspossible directly in front of the instrumentsto avoid parallax errors.

5.1.2.6 Magnetic tape recording

Multi-channel instrumentation taperecorders designed for airborneenvironments are employed for continuousrecording of flight and engine performanceparameters. Typical recorders arecompact intermediate/wide band and cantake both ½" and 1" magnetic tapes with a24 to 28 Volt DC power requirements.Six tape speeds and both direct and FMrecording are available in a tape recorderweighing about 27 kg.

5.1.2.7 Automatic still photographicrecording

Photographs of the flight deckinstrument panel can be taken using a handheld 35 mm SLR camera with an 85 mmlens and high speed slide film. Theindications on the instruments can be readby projecting the slides onto a screen.

5.1.2.8 Cine recording

Cine cameras with a one frame persecond exposure rate have been used toacquire flight deck data. Care must betaken in mounting the camera to ensurethat all the instruments required to bephotographed are within the field of view.Typical film cassettes containing about2000 frames have been used with a framecounter to allow film changes to beanticipated.

5.1.2.9 Video recording

Flight and engine performanceparameters can be recorded with a videocamera, although as with cine cameras,care must be taken to ensure that all theinstruments required are within the field ofview. The recorded information is playedback using freeze frame features to obtainindividual instrument readings.

5.1.2.10 Time synchronisation ofrecorded data

The need to synchronise the noiserecordings with the on board recordedflight deck data is important. This willinvolve radio communication between thehelicopter and the noise recordingpositions. Several methods have beenused such as noting the synchronisationtime on a clock mounted on the instrumentpanel which itself is recorded by the dataacquisition system. One such system usesa ground camera which operates a radiotransmission which, when received by thehelicopter lights two high intensity LED'smounted in an analogue clock attached tothe instrument panel.

5.1.3 Procedures for the determinationof changes in noise levels

Noise level changes determined bycomparison of flight test data for differenthelicopter model series have been used toestablish certification noise levels of modifiedor newly derived versions by reference to thenoise levels of the baseline or "flight datum"helicopter model. These noise changes areadded to or subtracted from the noise levelsobtained from individual flights of the "flightdatum" helicopter model. Confidence intervalsof new data are statistically combined with the"flight datum" data to develop overallconfidence intervals (see Appendix 1).

5.1.3.1 Modifications or upgradesinvolving aerodynamic drag changes

Use of drag devices, such as dragplates mounted beneath or on the sides ofthe "flight datum" helicopter, has provedto be effective in noise certification ofmodifications or upgrades involvingaerodynamic drag changes. Externalmodifications of this type are made bymanufacturers and aircraft "modifiers".Considerable cost savings are realised bynot having to perform noise testing ofnumerous individual modifications to thesame model series.

Based on these findings it is consideredacceptable to use the following as anequivalent procedure:

a) For helicopters to be certificatedunder Chapter 8 or Chapter 11 ofAnnex 16, Volume 1, a drag

jskaleck



38

device is used that produces theaerodynamic drag calculated forthe highest drag modification orcombination of modifications;

b) With the drag producing deviceinstalled, a flyover test and take-off and/or approach test ifconsidered appropriate by thecertificating authority, in the caseof a Chapter 8 certification, orflyover for the case of a Chapter11 certification, are performedusing the appropriate noisecertification reference and testprocedures;

c) A relationship of noise levelversus change in aerodynamicdrag or airspeed is developedusing noise data, adjusted asspecified in Appendix 2 or 4 ofAnnex 16, Volume 1, of the"flight datum" helicopter and ofthe "high drag" configuration;

d) The actual airspeed of themodification to be certificated isdetermined from performanceflight testing of the baselinehelicopter with the modificationinstalled; and

e) Using the measured airspeed ofthe modification, certificationnoise levels are determined byinterpolation of the relationshipdeveloped in c) above.

5.1.4 Temperature and RelativeHumidity Measurements

5.1.4.1 Temperature and relativehumidity measurements as defined inparagraph 2.2.3 of Appendix 2, Annex 16,Volume 1, are required to be made at aheight of 10 m (33 ft) above the ground.The measured values are used in theadjustment of the measured one-thirdoctave band sound pressure levels toaccount for the difference in the soundattenuation coefficients in the test andreference atmospheric conditions as givenin paragraph 8.3.1 of Appendix 2, Annex16, Volume 1. The distances QK andQrKr in the equations of paragraph 8.3.1refer to the distances between positions onthe measured and reference flight paths

corresponding to the apparent PNLTMposition and the noise measurement point.

5.1.4.2 As a consequence theprocedure assumes that the differencebetween the temperature and relativehumidity at 10 m and the PNLTM positionis zero or small and that the atmospherecan be represented by the values measuredat 10 m (33 ft) above the ground in thevicinity of the noise measurement point.Data obtained from European and U.S.certification tests over a number of years,and records provided by the U.K.Meteorological Office, have confirmedthis assumption is valid over a wide rangeof meteorological conditions.

5.1.4.3 Noise certificationmeasurements made under test conditionswhere significant changes in temperatureand/or relative humidity with height areexpected, particularly when a significantdrop in humidity with altitude is expected,should be adjusted using the average ofthe temperature and relative humiditymeasured at 10 m (33 ft) and at the heightassociated with the PNLTM point in orderto eliminate errors associated by use ofdata measured at 10 m (33 ft) only. Suchspecial conditions might be encountered indesert areas shortly after sunrise where thetemperature near the ground is lower, andthe relative humidity considerable higher,than at the height associated with thePNLTM point. Except for tests madeunder such conditions experience fromnoise certification tests over many yearsclearly indicates that the calculations ofparagraph 8.3.1 of Appendix 2 of Annex16, Volume 1, can be based onmeteorological data measured at 10 m (33ft) only.

5.1.4.4 Paragraph 2.2.2 of Appendix2, Annex 16, Volume 1 limits testing toconditions where the sound attenuationrate in the 8 kHz one-third octave band isnot more than 12 dB/100 m. If, however,the dew point and dry bulb temperatureare measured with a device which isaccurate to within ±0.5°C it has beenfound acceptable by certificatingauthorities to permit testing in conditionswhere the 8 kHz sound attenuation rate isnot more than 14 dB/100 m.

jskaleck



39

5.1.4.5 Testing of light helicoptersoutside Chapter 11 temperature andhumidity limits

With the approval of the certificatingauthority it may be possible to conducttesting of light helicopters outside the testenvironment specified in paragraph 2.2 ofICAO Annex 16, Volume 1, Appendix 4,provided the test environment is within thetemperature and relative humidity limitsspecified in paragraph 2.2 of ICAO Annex16, Volume 1, Appendix 2. In suchcircumstances it will be necessary toconduct a one-third octave band analysisof a noise recording of each overflight.The measured value of SEL shall becorrected from the test values oftemperature and humidity measuredaccording to paragraph 2.2.2 of ICAOAnnex 16, Volume 1, Appendix 4, to thereference conditions defined in paragraph11.5.1.4 of ICAO Annex 16, Volume 1,Chapter 11. The correction procedureshall be similar to that defined inparagraph 8.3.1 of ICAO Annex 16,Volume 1, Appendix 2 with thepropagation distances QK and QrKr

replaced by H, the height of the testhelicopter when it passes over the noisemeasurement point and the referenceheight, 150 m, respectively.

5.1.5 Anomalous Test Conditions

5.1.5.1 Annex 16, Volume 1,Appendix 2, Paragraph 2.2.2(f) requiresthat the tests be conducted underconditions that no anomalousmeteorological conditions exist. Thepresence of anomalous atmosphericconditions can be determined to asufficient level of certainty by monitoringthe outside air temperature (OAT) usingthe aircraft instruments. Anomalousconditions which could impact themeasured levels can be expected to existwhen the OAT at 150 m (492 ft) is 2°C(3.2°F) or more than the temperaturemeasured at 10 m (33 ft) above groundlevel. This check can be made in levelflight at a height of 150 m (492 ft) within30 minutes of each noise measurement.

5.1.5.2 Since the actual heightsassociated with the PNLTM points willnot be known until the analysis is made,measurements of temperature and relativehumidity can be made at a number of

heights and the actual value determinedfrom a chart of temperature and relativehumidity versus height. Alternativelysince the influence of height is small,measurements at a fixed height, in theorder of 120 m (400 ft) and 150 m (492 ft)depending on the flight condition andagreed with the certificating authorityprior to the tests being conducted can beused.

5.1.5.3 If tests are adjusted using the“average” of the temperature and relativehumidity measured at 10 m (33 ft) and theheight association with the PNLTM pointdescribed in 5.1.4.4, the provisions of5.1.5.1 do not apply. This is because theimpact of any anomalous meteorologicalconditions are taken into account by usingthe average of the temperature and relativehumidity at 10 m (33 ft) and the heightassociated with the PNLTM point.

jskaleck



40


jskaleck



41

SECTION 6 - EVALUATION METHODS

Several methods used to resolve difficultiesencountered during the measurement and analysisof aircraft noise data have been developed. Someof these procedures are applicable to all aircrafttypes whereas others have limited application toone or more. This section presents these evaluationmethods and describes specific procedures whichhave been approved to deal with:

a) spectral irregularities which are not relatedto the aircraft noise sources;

b) ambient noise levels both acoustic andelectrical;

c) the establishment and extension of databases;

d) the use of inertial navigation systems foraeroplane flight path measurements;

e) integrated analysis of noise data;

f) computation of EPNL by the integratedmethod of adjustment; and

g) calculation of the speed of sound.

6.1 SPECTRAL IRREGULARITIES

Tone corrections are required for tones orirregularities in the spectra of aeroplane noise.Irregularities which occur in measured spectra dueto interference effects caused by reflection of soundfrom the ground surface or by perturbations duringthe propagation of the noise from the aeroplane tothe microphone need to be identified so that tonecorrections are not applied to spectralcharacteristics which are not related to theaeroplane noise source. As specified in paragraph4.3.1 of Appendix 2 of Annex 16, Volume 1,narrow band analysis is one recommendedprocedure for identifying these false tones. Othermethods of identification that have been approvedfor use in certification are included in Appendix 2of this manual. However, the effect of thesespectral irregularities are not normally removedfrom data from propeller-driven aeroplanes orhelicopters since they are difficult to differentiatefrom engine and propeller or rotor associated tones.

6.2 AMBIENT NOISE LEVELS

Ambient noise levels including backgroundacoustic noise levels and electronic noise frommeasuring and analysis equipment can mask noise

levels from aeroplanes over parts of the frequencyrange of interest for effective perceived noise levelcalculations. The effects of ambient noise may beremoved by use of an approved procedure such asdescribed in Appendix 3.

6.3 ESTABLISHMENT AND EXTENSION OF

DATA BASES

6.3.1 Noise certification levels may bedetermined from a number of repeat noisemeasurements, at least six, at the same enginepower (transmitted power for helicopters),height, speed and configuration at each of thereference noise measurement positions. Thenoise measurements are adjusted to referenceconditions and the mean value and 90 per centconfidence interval obtained in accordancewith Annex 16, Volume 1. As an alternative,aircraft may be flown over a range of noisecorrelating parameters (µ). In the case ofaeroplanes the correlating parameter may beengine power setting and/or helical tip Machnumber for propeller driven aeroplanes. Forhelicopters the parameter may be powersetting, advancing blade tip Mach number,speed or any other agreed parameter. At leastsix flights are needed to establish the noiselevel versus the relevant correlating parameterrelationship covering the range relevant to boththe prototype and derived aircraft for each ofthe reference noise measurement sites.Provided that the 90 per cent confidenceinterval limit of not greater than ±1.5 EPNdB(or ±1.5 dBA as appropriate) is satisfied, ascalculated in Appendix 1 of this manual, thenoise certification levels may be obtained byentering the curve of noise level versuscorrelating parameter (µ) at the appropriatereference µ.

6.3.2 In some areas an extrapolation ofthe data field may be approved but care mustbe taken to ensure that the relativecontributions of the component noise sourcesto the effective perceived noise level, soundexposure level or A-weighted noise level asappropriate, remains essentially unchanged andthat a simple extrapolation of noise/correlatingparameter curves can be made.

6.3.3 For propeller driven aeroplanes achange in propeller and/or powerplant maynecessitate further flight tests to establish arevised noise-power-distance relationship.

jskaleck



42

6.4 TEST ENVIRONMENT CORRECTIONS

6.4.1 The atmospheric conditionsspecified in Annex 16, Volume 1, section2.2.2(b), (c), (d) and (e) of Appendix 2 requirethe measurement of ambient air temperatureand relative humidity profiles during noisecertification tests, to ensure that thetemperatures, relative humidities andcorresponding atmospheric sound absorptioncoefficients do not deviate from the specifiedlimits over the whole noise path betweenground and aeroplane. Ordinarily, profilemeasurements are recorded by balloon,instrumented aeroplane, or other similarmethod during flight testing, in order to ensurethat the criteria are met.

6.4.2 At the discretion of the certificatingauthority atmospheric profile measurements ofambient air temperature and relative humiditymay be made by instruments mounted on thetest aeroplane, and may be consideredsufficient to determine compliance with thecriteria specified in section 2.2.2(b), (c), (d)and (e) of Annex 16, Volume 1, Appendix 2.

6.5 INERTIAL NAVIGATION SYSTEMS FOR

AEROPLANE FLIGHT PATH MEASUREMENT

6.5.1 Criteria for the measurement ofaeroplane height and lateral position relative tothe intended track are described in section 2.3of Appendix 2 of Annex 16 Volume 1. Thissection indicates that the method used shouldbe independent of normal flightinstrumentation. Since the development of thisrequirement, other tracking systems (inertialnavigation systems (INS) and microwavesystems) which have a high degree of accuracyhave been installed in aeroplanes andconsequently have been accepted by severalcertificating authorities for use during noisecertification. However, it is important that anyinherent drift in the system is regularlydetermined and the system calibrated. For thispurpose ground based cameras can be used todetermine the position of an aeroplane relativeto them both laterally and in terms of height.The calibration should be undertakensufficiently frequently to retain the accuracyspecification of the system.

6.5.2 The accuracy of such systems mustbe acceptable to the certificating authority.

6.6 COMPUTATION OF EPNL BY THE

INTEGRATED METHOD OF ADJUSTMENT

6.6.1 Section 9.1 of Annex 16, Appendix2 provides for the use of the "simplified" or"integrated" method for adjusting measurednoise data to reference day conditions. The"integrated" procedure may be applied tomeasured data at the flyover, lateral, andapproach noise measurement points. The"integrated" adjustment method consists ofapplying all data adjustments to each measuredset of sound pressure levels obtained at 0.5 sintervals to identify equivalent referenceaverage sound pressure levels which are usedto compute EPNL's consistent with valueswhich would be obtained under referenceconditions. For complete acoustic consistencythe adjustment is only applicable if evaluatedfor identical pairs of noise emission angle (θ)relative to the flight path and noise elevationangle (ψ) relative to the ground for both themeasured (test) and corrected (reference) flightpaths. While this requirement may besatisfactorily approximated for the flyover andapproach noise measurements it can be shownthat it is not possible to retain identical pairs ofangles when lateral noise measurementadjustments are necessary. Therefore whenlateral noise measurement adjustments aremade by the "integrated" method, thegeometric conditions, of identical noiseemission angle should be maintained for testand reference flight paths while thecorresponding differences between test andreference elevation angles should beminimised. The slight difference that willoccur between test and reference elevationangle will have negligible effect on thecorrected EPNL value.

6.6.2 This section describes an integratedadjustment method that is applicable for usewhen the aeroplane is operated at constantconditions (flight path and power) during thenoise measurement period.

6.6.3 Test aircraft position

6.6.3.1 The "integrated" method foradjustment of measured noise-level data toreference conditions requires acoustic andaeroplane performance data at each 0.5 stime interval during the test flights. Thesedata include aeroplane position relative toa three-dimensional (X, Y, Z) co-ordinatesystem, 1/3-octave-band sound pressurelevels SPL(i,k), and time (tk) at themidpoint of each averaging time periodrelative to a reference time. Additionallyaeroplane performance parameters, themeasurement microphone locations, and

jskaleck



43

temperature and humidity is required foreach flyover.

6.6.3.2 The aircraft height Z ismeasured above the reference X-Y plane,generally taken to be the ground plane,with the measurement microphone 1.2 mabove this reference plane. The averagetest flight path is assumed to be a straightline (except when power reduction is usedduring the flyover measurement) and thetime-correlated aeroplane-position data areused to determine time of overhead (toh),the test overhead height (hTo)* , and thetest minimum distance (dTm) from the testflight path to the microphone location(K(XTM, YTM, ZTM)).

6.6.3.3 Using the test data directly orby geometric analysis of the relationbetween the average straight line flightpath and the minimum distance line fromKT to RT (XRT, YRT, ZRT) as shown inFigure 10 the minimum distance becomes:

dTm

XRT

XTM

YRT

YTM

ZRT

ZTM

= −

+ −

+ −

2 2 21

2

Equation (1)

6.6.4 Sound-propagation times andsound-emission angles

6.6.4.1 The test sound propagationtime ( ∆tpk) is identified with the datarecord time (tk), the noise emission time(tek), the aeroplane position (Ak) at time(tak), and the averaging time (tAv)through the relationships:

tk

tak

tAv

= − 12

Equation (2)

tek

tk

tpk

= − ∆

Equation (3)

∆ tpk

KT

Qek

cT

=

Equation (4)

where cT is the speed of sound for theaverage absolute temperature of the airbetween the surface (Ts) and the height of

*For emphasis the subscript "T" is used here for testconditions. Annex 16 uses un-subscripted symbolsfor test conditions.

the aeroplane (TA) (see Paragraph 6.7,where ( )T T TS A= + 2 ).

6.6.4.2 Using the geometricrelationships of Figure 11, the minimumdistance from Equation (1), the testdistance QekR, and defining the timedifference B equal to tTm - tk yields thefollowing expression for the test-flight-path sound-propagation times:

∆tTpk

cT

VT

BVT

cT

VT

dTm

BcT

VT

= −

×

+ −

+

1 2 2

2 2 2 2 21

2

Equation (5)

where VT is the average true air speed ofthe test aeroplane along the flight path.

6.6.4.3 Similarly the test sound-emission angle is defined as:

θek

dTm

dTpk

= −

sin 1 , or

θek

dTm

tTpk

cT

= −

sin 1 ∆

Equation (6)

6.6.5 Aircraft reference flight path

6.6.5.1 The geometry of the referenceflight path is essentially similar to thatshown in Figure 10, however, thefollowing differences exist. The referenceflight path is directly over the runwaycenterline (ie. YDEV=0). For the take offand approach flyovers, the measurementstation is on the runway centerline (ie.YRr=YrM); for lateral noisemeasurements, (YRr-YrM) equals thereference lateral displacement of themeasurement station.

Note (1): The subscript r is used to denotereference conditions.

Note (2): The reference microphonelocation (Kr) for flyover or lateral noisemeasurements is usually at the same co-ordinates as for the test location (KT), ie.( ) ( )X Y Z X Y ZTM TM TM rM rM rM, , , ,= .

6.6.5.2 The reference flight path maybe geometrically specified relative to the

jskaleck



44

reference microphone location (Kr) byusing the measurement station lateraldistance, the height overhead (hro) and theflight path inclination angle (γr). Thesevalues are equated to the minimumdistance (drm) from Kr by the following:

drm

hro r

YRr

YrM

= + −

2 2 21

2cos γ ,or

drm

XRr

XrM

YRr

YrM

ZRr

ZrM

= −

+ −

+ −

2 2 21

2

Equation (7)

6.6.5.3 The basic acousticassumption relating the test and referenceflight conditions is that the threedimensional acoustic emission angles (θekand θerk) for each test record time (tk) andthe corresponding reference time (trk) areequal. Using Equation (6) and thisequality the test sound pressure levels,SPLT(i,k), for each of the i-th frequencybands, are adjusted for spherical spreadingand atmospheric absorption over theacoustic path lengths by the equation:

( ) ( )SPLr

i rk SPLT

i k drpk

dTpk

ai

drpk

ai

dTpi

, , log= −

−

−

20

0

Equation (8a)

where ai0 and ai are the reference and testday sound attenuation coefficientsrespectively,

or, when the test and reference flight pathminimum distances are used, by theequation:

( ) ( )SPLr

i rk SPLT

i k drm

dTm

ai

drm

ai

dTm ek

, , log= −

−

−

20

0 cosecθ

Equation (8b)

6.6.6 Time interval computation

6.6.6.1 In addition to the aboveadjustments of the test data for sphericalspreading and atmospheric absorption it isnecessary to make an adjustment for thechange in the time increment trk used inthe computation of EPNL. Since the timeincrements are not equal to the 500 ms testmeasurement time increments when

adjusted by the "integrated" method,successive aeroplane position referencetimes (trk and tr(k+1)) occur after the timereference (trek) at the sound emissionpoint (Figure 11). The average timeincrement to be used in the EPNLcomputation is:

δtrk

trk

tr k

= +−

∆ ∆1

2

Equation (9)

where the reference time interval ( ∆trk)between data records is:

∆trk

tr k

trk

=+

−

1

and using the relationship between sampletimes, sound emission times, and soundpropagation times the reference intervalbecomes:

∆ ∆ ∆trk

tre k

trek

trp k

trpk

=+

−

++

−

1 1

Equation (10)

6.6.6.2 This time interval reflects thetime for the aeroplane to travel at test andreference speeds (VT and Vr) from onesound emission point to the next and alsothe effect of differences between test andreference minimum distances (drm anddTm) as well as sound speeds (cr and cT).These factors are expressed explicitly byarranging Equation (10) as follows:

∆ ∆ ∆

∆ ∆

trk

drm

dTm

VT

Vr

tTp k

tTpk

c cr

tTp k

tTpk

=

−

+−

+

+

−

0 51

1

.

T

Equation (11)

6.6.7 Adjusted effective perceived noiselevel

After the sound pressure levels have beenadjusted using Equation (8) the tonecorrections are calculated following section 4.3of Appendix 2, Annex 16 and using the noyweighting and the procedure for calculatingperceived noise level (Section 4.2, Appendix 2,Annex 16) the reference PNLT's are availablefor the times tr1 to trn which include the firstand last 10 dB-down times. These values andthe adjusted average time increment,

jskaleck



45

Equation (9), are combined to compute theadjusted EPNL as follows:

EPNL 1 T0

100.1PNLT

k

k 1

n t

rk=

=∑

10log δ

where the reference time (T0) is 10 s and thesummation is started by setting ( ) ( )∆ ∆t t

r r1 1 2 1− −= so

that ( )δt tr r1 1 1− = ∆ . The summation is terminated

by assuming ( )∆ ∆t trn r n= −1

giving

( )δt t trn rn r n= = −∆ ∆

1.

6.7 CALCULATION OF THE SPEED OF SOUND

For the purposes of noise certification thevalue of the speed of sound, c, shall be calculatedfrom the equation taken from ISO 9613-1: 1993(E):

( )c T T= 3432 0

12. metres / sec

(i.e. ( )c T T=11259 0

12. feet / sec )

where T0 29315= . °K and T and the absolute ambientair temperature.

jskaleck



46


jskaleck



47

SECTION 7 - MEASUREMENT AND ANALYSIS EQUIPMENT

Current technology noise recording andanalysis equipment employing digital techniqueswhich offer greater flexibility and precision thanbefore are not catered for in the specification ofequipment used in noise certification and describedin Annex 16, Volume 1, Appendix 2, Section 3.

This section describes new instrumentationstandards which have been approved by CAEPWorking Group 1 to replace the existing Section 3of Appendix 2 of Annex 16, Volume 1 and therebyenable the increasing use of digital instrumentation.This section will be removed from theEnvironmental Technical Manual when it isadopted into the Annex.

7.1 DEFINITIONS

For the purposes of this section the followingdefinitions apply:

7.1.1 Measurement system The combination of instruments used for

the measurement of sound pressure levels,including a sound calibrator, windscreen,microphone system, signal recording andconditioning devices, and one-third octaveband analysis system.

Note: Practical installations may include anumber of microphone systems, the outputsfrom which are recorded simultaneously by amulti-channel recording/analysis device viasignal conditioners as appropriate. For thepurpose of this section, each completemeasurement channel is considered to be ameasurement system to which the requirementsapply accordingly.

7.1.2 Microphone system:

The components of the measurementsystem which produce an electrical outputsignal in response to a sound pressure inputsignal, and which generally include amicrophone, a preamplifier, extension cables,and other devices as necessary.

7.1.3 Sound incidence angle

In degrees, an angle between the principalaxis of the microphone, as defined in IEC

61094-31 and IEC 61094-42, as amended and aline from the sound source to the centre of thediaphragm of the microphone.

Note: When the sound incidence angle is 0°,the sound is said to be received at themicrophone at "normal (perpendicular)incidence"; when the sound incidence angle is90°, the sound is said to be received at"grazing incidence".

7.1.4 Reference direction

In degrees, the direction of soundincidence specified by the manufacturer of themicrophone, relative to a sound incidenceangle of 0°, for which the free-field sensitivitylevel of the microphone system is withinspecified tolerance limits.

7.1.5 Free-field sensitivity of amicrophone system

In volts per pascal, for a sinusoidal planeprogressive sound wave of specified frequency,at a specified sound-incidence angle, thequotient of the root mean square voltage at theoutput of a microphone system and the rootmean square sound pressure that would exist atthe position of the microphone in its absence.

7.1.6 Free-field sensitivity level of amicrophone system

In decibels, twenty times the logarithm tothe base ten of the ratio of the free-fieldsensitivity of a microphone system and thereference sensitivity of one volt per pascal.

Note: The free-field sensitivity level of amicrophone system may be determined by

1 IEC 61094-3: 1995 entitled “Measurement microphones –

Part 3: Primary method for free-field calibration of laboratory

standard microphones by the reciprocity technique”

2 IEC 61094-4: 1995 entitled “Measurement microphones –

Part 4: Specifications for working standard microphones”

These IEC publications may be obtained from the Bureau

central de la Commission électrotechnique internationale, 1 rue

de Varembé, Geneva, Switzerland

jskaleck



48

subtracting the sound pressure level (indecibels re 20 mPa) of the sound incident onthe microphone from the voltage level (indecibels re 1 V) at the output of themicrophone system, and adding 93.98 dB tothe result.

7.1.7 Time-average band soundpressure level

In decibels, ten times the logarithm to thebase ten, of the ratio of the time mean squareof the instantaneous sound pressure during astated time interval and in a specified one-thirdoctave band, to the square of the referencesound pressure of 20µPa.

7.1.8 Level range

In decibels, an operating range determinedby the setting of the controls that are providedin a measurement system for the recording andone-third octave band analysis of a soundpressure signal. The upper boundary associatedwith any particular level range shall berounded to the nearest decibel.

7.1.9 Calibration sound pressure level

In decibels, the sound pressure levelproduced, under reference environmentalconditions, in the cavity of the coupler of thesound calibrator that is used to determine theoverall acoustical sensitivity of a measurementsystem.

7.1.10 Reference level range

In decibels, the level range for determiningthe acoustical sensitivity of the measurementsystem and containing the calibration soundpressure level.

7.1.11 Calibration check frequency

In hertz, the nominal frequency of thesinusoidal sound pressure signal produced bythe sound calibrator.

7.1.12 Level difference

In decibels, for any nominal one-thirdoctave midband frequency, the output signallevel measured on any level range minus thelevel of the corresponding electrical inputsignal.

7.1.13 Reference level difference

In decibels, for a stated frequency, thelevel difference measured on a level range foran electrical input signal corresponding to thecalibration sound pressure level, adjusted asappropriate, for the level range.

7.1.14 Level non-linearity

In decibels, the level difference measuredon any level range, at a stated one-third octavenominal midband frequency, minus thecorresponding reference level difference, allinput and output signals being relative to thesame reference quantity.

7.1.15 Linear operating range

In decibels, for a stated level range andfrequency, the range of levels of steadysinusoidal electrical signals applied to the inputof the entire measurement system, exclusive ofthe microphone but including the microphonepreamplifier and any other signal-conditioningelements that are considered to be part of themicrophone system, extending from a lower toan upper boundary, over which the level non-linearity is within specified tolerance limits.

Note: It is not necessary to includemicrophone extension cables as configured inthe field.

7.1.16 Windscreen insertion loss

In decibels, at a stated nominal one-thirdoctave midband frequency, and for a statedsound incidence angle on the insertedmicrophone, the indicated sound pressure levelwithout the windscreen installed around themicrophone minus the sound pressure levelwith the windscreen installed.

7.2 REFERENCE ENVIRONMENTAL

CONDITIONS

7.2.1 The reference environmentalconditions for specifying the performance of ameasurement system are:

- air temperature 23°C

- static air pressure 101.325 kPa

- relative humidity 50 %

7.3 GENERAL

Note Measurements of aircraft noise that utiliseinstruments that conform to the specifications of

jskaleck

jskaleck



49

this section yield one-third octave band soundpressure levels as a function of time, for thecalculation of effective perceived noise level asdescribed in Annex 16, Volume 1, Appendix 2,Section 4.

7.3.1 The measurement system shallconsist of equipment approved by thecertificating authority and equivalent to thefollowing:

a) a windscreen (see 7.4);

b) a microphone system (see 7.5);

c) a recording and reproducing system tostore the measured aircraft noisesignals for subsequent analysis (see7.6);

d) a one-third octave band analysissystem (see 7.7); and

e) calibration systems to maintain theacoustical sensitivity of the abovesystems within specified tolerancelimits (see 7.8).

7.3.2 For any component of themeasurement system that converts an analogsignal to digital form, such conversion shall beperformed so that the levels of any possiblealiases or artefacts of the digitisation processwill be less than the upper boundary of thelinear operating range by at least 50 dB at anyfrequency less than 12.5 kHz. The samplingrate shall be at least 28 kHz. An anti-aliasingfilter shall be included before the digitisationprocess.

7.4 WINDSCREEN

7.4.1 In the absence of wind and forsinusoidal sounds at grazing incidence, theinsertion loss caused by the windscreen of astated type installed around the microphoneshall not exceed ±1.5 dB at nominal one-thirdoctave midband frequencies from 50 Hz to 10kHz inclusive.

7.5 MICROPHONE SYSTEM

7.5.1 The microphone system shallconform to the specifications in 7.5.2 to 7.5.4.Various microphone systems may be approvedby the certificating authority on the basis ofdemonstrated equivalent overallelectroacoustical performance. Where two ormore microphone systems of the same type areused, demonstration that at least one systemconforms to the specifications in full issufficient to demonstrate conformance.

Note: This demonstration of equivalentperformance does not eliminate the need tocalibrate and check each system as defined in7.9.

7.5.2 The microphone shall be mountedwith the sensing element 1.2 m (4 ft) above thelocal ground surface and shall be oriented forgrazing incidence, i.e., with the sensingelement substantially in the plane defined bythe predicted reference flight path of theaircraft and the measuring station. Themicrophone mounting arrangement shallminimise the interference of the supports withthe sound to be measured. Figure 12 illustratessound incidence angles on a microphone.

7.5.3 The free-field sensitivity level of themicrophone and preamplifier in the referencedirection, at frequencies over at least the rangeof one-third-octave nominal midbandfrequencies from 50 Hz to 5 kHz inclusive,shall be within ±1.0 dB of that at thecalibration check frequency, and within ±2.0dB for nominal midband frequencies of 6.3kHz, 8 kHz and 10 kHz.

7.5.4 For sinusoidal sound waves at eachone-third octave nominal midband frequencyover the range from 50 Hz to 10 kHz inclusive,the free-field sensitivity levels of themicrophone system at sound incidence anglesof 30°, 60°, 90°, 120° and 150°, shall not differfrom the free-field sensitivity level at a soundincidence angle of 0° (“normal incidence”) bymore than the values shown in Table 7-1. Thefree-field sensitivity level differences at soundincidence angles between any two adjacentsound incidence angles in Table 7-1 shall notexceed the tolerance limit for the greater angle.

jskaleck



50

7.6 RECORDING AND REPRODUCING

SYSTEMS

7.6.1 A recording and reproducingsystem, such as a digital or analogue magnetictape recorder, a computer-based system orother permanent data storage device, shall beused to store sound pressure signals forsubsequent analysis. The sound produced bythe aircraft shall be recorded in such a way thata record of the complete acoustical signal isretained. The recording and reproducingsystems shall conform to the specifications in7.6.2 to 7.6.9 at the recording speeds and/ordata sampling rates used for the noisecertification tests. Conformance shall bedemonstrated for the frequency bandwidths andrecording channels selected for the tests.

7.6.2 The recording and reproducingsystems shall be calibrated as described in 7.9.

Note: For aircraft noise signals for which thehigh frequency spectral levels decrease rapidlywith increasing frequency, appropriate pre-emphasis and complementary de-emphasisnetworks may be included in the measurementsystem. If pre-emphasis is included, over therange of nominal one-third octave midbandfrequencies from 800 Hz to 10 kHz inclusive,the electrical gain provided by the pre-emphasis network shall not exceed 20 dBrelative to the gain at 800 Hz.

7.6.3 For steady sinusoidal electricalsignals applied to the input of the entiremeasurement system exclusive of themicrophone system, but including themicrophone preamplifier, and any other signal-conditioning elements that are considered to bepart of the microphone system, at a selectedsignal level within 5 dB of that correspondingto the calibration sound pressure level on thereference level range, the time average signallevel indicated by the readout device at anyone-third octave nominal midband frequencyfrom 50 Hz to10 kHz inclusive shall be within±1.5 dB of that at the calibration checkfrequency. The frequency response of ameasurement system, which includescomponents that convert analogue signals todigital form, shall be within ±0.3 dB of theresponse at 10 kHz over the frequency rangefrom 10 kHz to 11.2 kHz.

Note: It is not necessary to includemicrophone extension cables as configured inthe field.

7.6.4 For analogue tape recordings, theamplitude fluctuations of a 1 kHz sinusoidalsignal recorded within 5 dB of the levelcorresponding to the calibration sound pressurelevel shall not vary by more than ±0.5 dBthroughout any reel of the type of magnetictape utilised. Conformance to this requirementshall be demonstrated using a device which has

Nominal midbandfrequency kHz

Maximum difference between the free-field sensitivity level of a microphone system at normal incidence andthe free-field sensitivity level at specified sound incidence angles

dB______________________________________________________________________________________

Sound Incidence angledegrees

30 60 90 120 150

0.05 to 1.62.02.5

3.154.05.0

6.38.0

10.0

0.50.50.5

0.50.50.5

1.01.52.0

0.50.50.5

1.01.01.5

2.02.53.5

1.01.01.0

1.52.02.5

3.04.05.5

1.01.01.5

2.02.53.0

4.05.56.5

1.01.01.5

2.02.53.0

4.05.57.5

Table 7-1 Microphone Directional Response Requirements

jskaleck



51

time-averaging properties equivalent to thoseof the spectrum analyser.

7.6.5 For all appropriate level ranges andfor steady sinusoidal electrical signals appliedto the input of the measurement systemexclusive of the microphone system, butincluding the microphone preamplifier, andany other signal-conditioning elements that areconsidered to be part of the microphonesystem, at one-third-octave nominal midbandfrequencies of 50 Hz, 1 kHz and 10 kHz, andthe calibration check frequency, if it is not oneof these frequencies, the level non-linearityshall not exceed ±0.5 dB for a linear operatingrange of at least 50 dB below the upperboundary of the level range.

Note 1: Level linearity of measurement systemcomponents should be tested according to themethods described in IEC 612653 as amended.

Note 2: It is not necessary to includemicrophone extension cables as configured inthe field.

7.6.6 On the reference level range, thelevel corresponding to the calibration soundpressure level shall be at least 5 dB, but nomore than 30 dB less than the upper boundaryof the level range.

7.6.7 The linear operating ranges onadjacent level ranges shall overlap by at least50 dB minus the change in attenuationintroduced by a change in the level rangecontrols.

Note: It is possible for a measurement systemto have level range controls that permitattenuation changes of either 10 dB or 1 dB,for example. With 10 dB steps, the minimumoverlap required would be 40 dB, and with 1dB steps the minimum overlap would be 49 dB.

3 IEC 61265: 1995 entitled “Instruments for measurement of

aircraft noise – Performance requirements for systems to

measure one-third-octave band sound pressure levels in noise

certification of transport-category aeroplanes”. This IEC

publications may be obtained from the Bureau central de la

Commission électrotechnique internationale, 1 rue de Varembé,

Geneva, Switzerland

7.6.8 Provision shall be made for anoverload indication to occur during anoverload condition on any relevant level range.

7.6.9 Attenuators included in themeasurement system to permit range changesshall operate in known intervals of decibelsteps.

7.7 ANALYSIS SYSTEMS

7.7.1 The analysis system shall conformto the specifications in 7.7.2 to 7.7.7 for thefrequency bandwidths, channel configurationsand gain settings used for analysis.

7.7.2 The output of the analysis systemshall consist of one-third octave band soundpressure levels as a function of time, obtainedby processing the noise signals (preferablyrecorded) through an analysis system with thefollowing characteristics:

a) a set of 24 one-third octave bandfilters, or their equivalent, havingnominal midband frequencies from 50Hz to 10 kHz inclusive;

b) response and averaging properties inwhich, in principle, the output fromany one-third octave filter band issquared, averaged and displayed orstored as time averaged soundpressure levels;

c) the interval between successivesound-pressure level samples shall be500 ms ±5 ms for spectral analysiswith or without SLOW timeweighting;

d) for those analysis systems that do notprocess the sound-pressure signalsduring the period of time required forreadout and / or resetting of theanalyser, the loss of data shall notexceed a duration of 5 ms; and

e) the analysis system shall operate inreal time from 50 Hz to at least 12kHz inclusive. This requirementapplies to all operating channels of amulti-channel spectral analysissystem.

7.7.3 The one-third octave band analysissystem shall at least conform to the class 2electrical performance requirements of IEC

jskaleck



52

612604 as amended, over the range of one-third octave nominal midband frequencies from50 Hz to 10 kHz inclusive.

Note: Tests of the one-third octave bandanalysis system should be made according tothe methods described in IEC 612604 or by anequivalent procedure approved by thecertificating authority, for relative attenuation,anti-aliasing filters, real time operation, levellinearity, and filter integrated response(effective bandwidth).

7.7.4 When SLOW time averaging isperformed in the analyser, the response of theone-third octave band analysis system to asudden onset or interruption of a constantsinusoidal signal at the respective one-thirdoctave nominal midband frequency shall bemeasured at sampling instants 0.5, 1, 1.5 and2 s after the onset and 0.5 and 1 s afterinterruption. The rising response shall be -4 ± 1 dB at 0.5 s, -1.75 ± 0.75 dB at 1 s, -1 ± 0.5 dB at 1.5 s and -0.5 ± 0.5 dB at 2 srelative to the steady-state level. The fallingresponse shall be such that the sum of theoutput signal levels, relative to the initialsteady-state level, and the corresponding risingresponse reading is -6.5 ± 1 dB, at both 0.5 and1 s. At subsequent times the sum of the risingand falling responses shall be -7.5 dB or less.This equates to an exponential averagingprocess (SLOW weighting) with a nominal 1 stime constant (i.e. 2 s averaging time).

7.7.5 When the one-third octave bandsound pressure levels are determined from theoutput of the analyser without SLOW timeweighting, SLOW time weighting shall besimulated in the subsequent processing.Simulated SLOW weighted sound pressurelevels can be obtained using a continuousexponential averaging process by the followingequation:

( ) ( ) ( )[ ] ( ) ( )L i kSL i k L i kS, log . .

. , . ,= +

−10 0 60653 10 0 39347 10

0 1 1 0 1

4 IEC 61260: 1995 entitled “ Electroacoustics – Octave-band

and fractional-octave-band filters”. This IEC publications may

be obtained from the Bureau central de la Commission

électrotechnique internationale, 1 rue de Varembé, Geneva,

Switzerland

where Ls(i,k) is the simulated SLOW weightedsound pressure level and L(i,k) is the as-measured 0.5 s time average sound pressurelevel determined from the output of theanalyser for the k-th instant of time and the i-thone-third octave band. For k =1, the SLOWweighted sound pressure Ls[i,(k-1=0)] on theright hand side should be set to 0 dB.

An approximation of the continuousexponential averaging is represented by thefollowing equation for a four sample averagingprocess for k ≥ 4:

( ) ( ) ( )[ ] ( ) ( )[ ]

( ) ( )[ ] ( ) [ ]L i ks

L i k L i k

L i k L i k, log

. .

. .

. , . ,

. , . ,=

+ +

+

− −

−10

013 10 0 21 10

0 27 10 0 39 10

0 1 3 0 1 2

0 1 1 0 1

where Ls(i,k) is the simulated SLOW weightedsound pressure level and L(i,k) is the asmeasured 0.5 s time average sound pressurelevel determined from the output of theanalyser for the k-th instant of time and the i-thone-third octave band.

The sum of the weighting factors is 1.0 in thetwo equations. Sound pressure levelscalculated by means of either equation arevalid for the sixth and subsequent 0.5 s datasamples, or for times greater than 2.5 s afterinitiation of data analysis.

Note: The coefficients in the two equationswere calculated for use in determiningequivalent SLOW weighted sound pressurelevels from samples of 0.5 s time averagesound pressure levels. The equations shouldnot be used with data samples where theaveraging time differs from 0.5 s.

7.7.6 The instant in time by which aSLOW time weighted sound pressure level ischaracterised shall be 0.75 s earlier than theactual readout time.

Note: The definition of this instant in time isrequired to correlate the recorded noise withthe aircraft position when the noise wasemitted and takes into account the averagingperiod of the SLOW weighting. For each ½second data record this instant in time mayalso be identified as 1.25 seconds after thestart of the associated 2 second averagingperiod.

7.7.7 The resolution of the sound pressurelevels, both displayed and stored, shall be 0.1dB or better.

jskaleck



53

7.8 CALIBRATION SYSTEMS

7.8.1 The acoustical sensitivity of themeasurement system shall be determined usinga sound calibrator generating a known soundpressure level at a known frequency. Thesound calibrator shall at least conform to theclass 1L requirements of IEC 609425 asamended.

7.9 CALIBRATION AND CHECKING OF

SYSTEM

7.9.1 Calibration and checking of themeasurement system and its constituentcomponents shall be carried out to thesatisfaction of the certificating authorities bythe methods specified in 7.9.2 to 7.9.10. Thecalibration adjustments, including those forenvironmental effects on sound calibratoroutput level, shall be reported to thecertificating authority and applied to themeasured one-third-octave sound pressurelevels determined from the output of theanalyzer. Data collected during an overloadindication are invalid and shall not be used. Ifthe overload condition occurred duringrecording, the associated test data shall beconsidered as invalid, whereas if the overloadoccurred during analysis the analysis shall berepeated with reduced sensitivity to eliminatethe overload.

7.9.2 The free field frequency response ofthe microphone system may be determined byuse of an electrostatic actuator in combinationwith manufacturer’s data or by tests in ananechoic free-field facility. The correction forfrequency response shall be determined within90 days of each test series. The correction fornon-uniform frequency response of themicrophone system shall be reported to thecertificating authority and applied to themeasured one-third octave band sound pressurelevels determined from the output of theanalyser.

7.9.3 When the angles of incidence ofsound emitted from the aircraft are within ±30°of grazing incidence at the microphone (see

5 IEC 60942: 1997 entitled “ Electroacoustics - Sound

calibrators”. This IEC publications may be obtained from the

Bureau central de la Commission électrotechnique

internationale, 1 rue de Varembé, Geneva, Switzerland

Figure 12), a single set of free-field correctionsbased on grazing incidence is consideredsufficient for correction of directional responseeffects. For other cases, the angle of incidencefor each ½ second sample shall be determinedand applied for the correction of incidenceeffects.

7.9.4 For analogue magnetic taperecorders, each reel of magnetic tape shallcarry at least 30 seconds of pink random orpseudo-random noise at its beginning and end.Data obtained from analogue tape-recordedsignals shall be accepted as reliable only iflevel differences in the 10 kHz one-third-octave-band are not more than 0.75 dB for thesignals recorded at the beginning and end.

7.9.5 The frequency response of the entiremeasurement system while deployed in thefield during the test series, exclusive of themicrophone, shall be determined at a levelwithin 5 dB of the level corresponding to thecalibration sound pressure level on the levelrange used during the tests for each one-thirdoctave nominal midband frequency from50 Hzto10 kHz inclusive, utilising pink random orpseudo-random noise. The output of the noisegenerator shall be determined by a methodtraceable to a national standards laboratorywithin 6 months of each test series andtolerable changes in the relative output fromthe previous calibration at each one-thirdoctave band shall be not more than 0.2 dB.The correction for frequency response shall bereported to the certificating authority andapplied to the measured one-third octave soundpressure levels determined from the output ofthe analyser.

7.9.6 The performance of switchedattenuators in the equipment used during noisecertification measurements and calibrationshall be checked within six months of each testseries to ensure that the maximum error doesnot exceed 0.1 dB.

7.9.7 The sound pressure level producedin the cavity of the coupler of the soundcalibrator shall be calculated for the testenvironmental conditions using themanufacturer’s supplied information on theinfluence of atmospheric air pressure andtemperature. The sound pressure level shall beused to establish the acoustical sensitivity ofthe measurement system. The output of thesound calibrator shall be determined by amethod traceable to a national standardslaboratory within 6 months of each test series

jskaleck



54

and tolerable changes in output from theprevious calibration shall be not more than 0.2dB.

7.9.8 Sufficient sound pressure levelcalibrations shall be made during each test dayto ensure that the acoustical sensitivity of themeasurement system is known at the prevailingenvironmental conditions corresponding witheach test series. The measurement system shallbe considered satisfactory if the difference isnot greater than 0.5 dB between the acousticalsensitivity levels recorded immediately beforeand immediately after each test series on agiven day. The 0.5 dB limit applies after anyatmospheric pressure corrections have beendetermined for the calibrator output level. Thearithmetic mean of the before and aftermeasurements shall be used to represent theacoustical sensitivity level of the measurementsystem for that test series. The calibrationcorrections shall be reported to the certificatingauthority and applied to the measured one-thirdoctave band sound pressure levels determinedfrom the output of the analyser.

7.9.9 Each recording medium, such as areel, cartridge, cassette, or diskette, shall carrya sound pressure level calibration of at least 10seconds duration at its beginning and end.

7.9.10 The free-field insertion loss of thewindscreen for each one-third octave nominalmidband frequency from 50 Hz to 10 kHzinclusive shall be determined with sinusoidalsound signals at appropriate incidence angleson the inserted microphone. For a windscreenwhich is undamaged and uncontaminated theinsertion loss may be taken frommanufacturer’s data. In addition the insertionloss of the windscreen may be determined by amethod traceable to a national standardslaboratory within 6 months of each test seriesand tolerable changes in the insertion loss fromthe previous calibration at each one-third-octave frequency band shall be not more than0.4 dB. The correction for the free-fieldinsertion loss of the windscreen shall bereported to the certificating authority andapplied to the measured one-third octave soundpressure levels determined from the output ofthe analyser.

7.10 ADJUSTMENTS FOR AMBIENT NOISE

7.10.1 The ambient noise, including bothacoustical background and electrical noise ofthe measurement system, shall be recorded forat least 10s at the measurement points with the

system gain set at the levels used for theaircraft noise measurements, at appropriatetimes during each test day. The ambient noiseshall be representative of the acousticalbackground that exits during the flyover testrun. The recorded aircraft noise data shall beaccepted only if the ambient noise levels whenanalysed in the same way and quoted in PNL(see Paragraph 4.1.3(a) of Appendix 2, Annex16, Volume 1) are at least 20 dB below themaximum PNL of the aircraft.

7.10.2 Aircraft sound pressure levelswithin the 10 dB-down points (see Paragraph4.5.1 of Appendix 2, Annex 16, Volume 1)shall exceed the mean ambient noise levelsdetermined above by at least 3 dB in each one-third octave band or be adjusted using themethod described in Appendix 3.

jskaleck



55

SECTION 8 - CONTROL OF NOISE CERTIFICATION COMPUTER PROGRAM SOFTWARE ANDDOCUMENTATION RELATED TO STATIC-TO-FLIGHT PROJECTION PROCESSES

8.1 GENERAL

8.1.1 Procedures for computer programsoftware control shall be developed, approvedby the certificating authority, and maintainedand adhered to by each applicant utilising"static-to-flight equivalencies (SFE's)".

8.1.2 The procedures shall consist of fourkey elements which, when implemented by thenoise certification applicant, shall result indocumentation which properly describes andvalidates the applicable SFE noise certificationcomputer program and data output.Throughout the development of a givenaeroplane type, adherence to these procedureswill enable critical computer programs to betracked in order to verify that the initialsoftware design has not been changed withoutsubstantiation.

8.1.3 The four key elements ofconfiguration index, control procedures, designdescription and verification process aredescribed below.

8.2 SOFTWARE CONTROL PLAN

8.2.1 Configuration index

A configuration index shall be establishedfor each unique SFE software system. It willinclude all applicable elements of the softwaresystem and provide historic tracking ofdocuments and software under control. Whereappropriate, the index may be maintained in ageneral data base.

8.2.2 Software control plan

8.2.2.1 A procedure for SFE softwarechange management shall be establishedthat includes the baseline designidentification, a software change controlsystem and a method of reviewing andauditing software changes and maintaininga status accounting of changes.

8.2.2.2 Control of software changesshall be maintained by establishingbaselines within the verification process(see 8.2.4) and documenting modificationsto the baseline case that result fromprogram coding changes. Review andauditing procedures will be established

within the verification process that allowthe validity of the program coding changesfor the "modified" configuration to beassessed relative to the "baseline"configuration.

8.2.2.3 The configuration index shallbe updated to reflect, historically, thechanges made to the software system.

8.2.3 Design description

A technical description of the methodsused to accomplish the SFE certification shallbe provided; including an overview and adescription of the software system design toaccomplish the technical requirements. Thesoftware design description should include theprogram structure, usage of subroutines,program flow control and data flow.

8.2.4 Verification process

The validation procedure for the SFEsoftware system, or modifications to it, shallinclude a process to verify that the calculationsdescribed in the documentation are beingproperly performed by the software. Theprocess may include hand calculationscompared to computer output, stepwisegraphical displays, software audits, diagnosticsubroutines that generate output of all relevantvariables associated with the modifications, orother methods to establish confidence in theintegrity of the software. The process resultsshall be monitored and tracked relative tosoftware calculation changes.

8.3 APPLICABILITY

Although the software control plan isapplicable to all SFE-specific computer programsoftware and documentation established through thespecific procedures and processes of eachapplicant, it may not be necessary to review andaudit ancillary software (such as, but not limited to,subroutines dealing with atmospheric absorptionrates, noy calculations, tone corrections) for eachmain program source code change.

jskaleck



56


jskaleck



57

SECTION 9 - GUIDELINES FOR NOISE CERTIFICATION OF TILTROTOR AIRCRAFT

Guidelines have been developed for the noisecertification of tiltrotor aircraft. These arepresented as Attachment YY to ICAO Annex 16,Volume 1. To help in the understanding of theseguidelines and to assist in their applicationbackground information is presented in Appendix6.

jskaleck



58


jskaleck



59

REFERENCES

1. Jet mixing noise: comparison of measurement and theory.B.J. Tester and V.M. Szewczyk, AIAA Paper 79-0570.

2. 1974 Proceedings of the Second Inter-Agency Symposium on University Research in Transportation Noise,Volume 1, Pages 50-58, On Noise Produced by Subsonic Jets.J. Laufer, R.E. Kaplan and W.T. Chu.

3. Airframe Noise Prediction Method.M.R. Fink, USA DOT Report FAA-RD-77-29 dated March 1977.

4. SAE ARP 866A: 1975 Standard Values of Atmospheric Absorption as a Function of Temperature andHumidity.

5. SAE ARP 876D: 1993 Gas Turbine Jet Exhaust Noise Prediction.

6. SAE AIR 1672B: 1983 Practical Methods to Obtain Free Field Sound Pressure Levels from AcousticalMeasurements over Ground Surfaces.

7. SAE AIR 1846-1984: Measurement of Noise from Gas Turbine Engines During Static Operation.

8. SAE AIR 1905-1985: Gas Turbine Co-axial Exhaust Flow Noise - Methods of Prediction Considered forInclusion in SAE ARP 876.

9. ESDU Item 80038, Amendment A: The Correction of Measured Noise Spectra for the effects of GroundReflection.

NOTE 1: ESDU Data items may be obtained from ESDU International Ltd., 251-259 Regent Street, London,W1R 7AD, United Kingdom.

NOTE 2: SAE AIR's and ARP's may be obtained from the Society of Automotive Engineers, Inc., 400Commonwealth Drive, Warrendale, PA 15096, United States of America.

jskaleck



60


jskaleck



61

Figure 1 Flight Path Intercept Procedures

jskaleck



62

jskaleck



63

jskaleck



64

jskaleck



65

jskaleck



66

Figure 7Generalised projection of static engine data to aeroplane flight conditions

jskaleck



67

jskaleck



68

jskaleck



69

jskaleck



70

0° Normal Incidence

60° 120°

Sensing Element

90° Grazing Incidence

±30°

180°

Figure 12 : Illustration of sound incidence angles on a microphone

jskaleck



71

Figure 13 Typical Test and Reference Procedures

jskaleck



72

jskaleck



73

ANNEX 16, CHAPTER 8 “ZERO ATTENUATION ADJUSTMENT WINDOW”

FIGURE 16

jskaleck



74


jskaleck



Appendix 175

Appendix 1

CALCULATION OF CONFIDENCE INTERVALS

1. INTRODUCTION

The use of Noise-Power-Distance (NPD) curves requires that confidence intervals be determined usinga more general formulation than is used for a cluster of data points. For this more general case confidenceintervals may need to be calculated about a regression line for:

(a) flight test data,

(b) a combination of flight test and static test data,

(c) analytical results,

or a combination thereof.

The latter two are of particular significance for noise certifications of an aircraft model range and requirespecial care when pooling the different sources of sampling variability.

Sections 2 to 5 provide an insight into the theory of confidence interval evaluation. The application ofthis theory and some worked examples are presented in Section 6. A suggested bibliography is given in Section7 for those wishing to gain a greater understanding.

2. CONFIDENCE INTERVAL FOR THE MEAN OF FLIGHT TEST DATA

2.1 Confidence interval for the sample estimate of the mean of clustered measurements

If n measurements of effective perceived noise levels y y yn1 2, ,...., are obtained under approximately

the same conditions and it can be assumed that they constitute a random sample from a normal population withtrue population mean, µ , and true standard deviation, σ , then the following statistics can be derived:-

y = estimate of the mean = 1

1ny i

i

i n

( )=

=

∑

,

s = estimate of the standard deviation

=

( )y y

n

i

i

i n

−

−=

=

∑ 2

1

1 .

From these and the Student's t-distribution, the confidence interval, CI , for the estimate of the mean,y , can be determined, as:

( )CI y ts

n= ±

−12α ζ,

where ( )t1

2− α ζ,

denotes the ( )1 2− α percentile of the single-sided Student's t-test with ζ degrees

of freedom ( for a clustered data set ζ = −n 1 ) and where α is defined such that ( )100 1 − α percent is the

desired confidence level for the confidence interval. That is it denotes the probability with which the intervalwill contain the unknown mean, µ . For noise certification purposes 90% confidence intervals are generallydesired and, thus t. ,95 ζ is used. See Table 1-2 situated at the end of this appendix for a listing of values of t. ,95 ζ

for different values of ζ .

jskaleck



Appendix 176

2.2 Confidence interval for mean Line obtained by regression

If n measurements of effective perceived noise levels y y yn1 2, ,...., are obtained under significantly

varying values of engine-related parameter x x xn1 2, ,...., respectively, then a polynomial can be fitted to the

data by the method of least squares. The following polynomial regression model for the mean effective perceivednoise level, µ , is assumed to apply:

µ = + + + +B B x B x B xkk

0 1 22 .....

and the estimate of the mean line through the data of the effective perceived noise level is given by:

y b b x b x b xkk= + + + +0 1 2

2 .....

Each regression coefficient Bi is estimated by bi from the sample data using the method of least

squares in a process summarised below.

Every observation ( x yi i, ) satisfies the equations

y B B x B x B xi i i k i

k

i= + + + + +0 1 2

2..... ε

= + + + + +b b x b x b x ei i k i

k

i0 1 2

2.....

where εi and ei are the random error and residual associated with the effective perceived noise level.

The random error, εi , is assumed to be a random sample from a normal population with mean zero and standard

deviation σ . The residual, ei , is the difference between the measured value and the estimate of the value

using the estimates of the regression coefficients and xi . Its root mean square value, s , is the sample estimate

for σ . These equations are often referred to as the normal equations.

The n data points of measurements ( x yi i, ) are processed as follows:

Each elemental vector, x i , and its transpose ′xi , are formed such that

( )x x x xi i i ik= 1 2 . . , a row vector,

and ′ =

x

x

x

x

i

i

i

ik

1

2

.

.

, a column vector.

A matrix X is formed from all the elemental vectors x i ni for = 1,...., . ′X is the transpose of

X .

We define a matrix A such that A X X= ′ and a matrix A

−1 to be the inverse of A.

Also y y y yn= ( . . . )1 2

and b b b bk= ( . . . )0 1

jskaleck



Appendix 177

with b determined as the solution of the normal equations:

y X b=and ′ = ′ =X y X X b Ab ,

to give b A X y= ′−1 .

The 90% confidence interval, CI90 , for the mean value of the effective perceived noise level estimated

with the associated value of the engine-related parameter, x0 , is then defined as

( ) ( )CI y x t s v x90 0 95 0= ± . ,ζ , where ( )v x x A x0 0

1

0= ′− .

Thus ( )CI y x t s x A x90 0 95 01

0= ± ′−. ,ζ

where ( )x x x x k0 0 0

201= . . ,

′x 0 is the transpose of x0 ,

y x( )0 is the estimate of the mean value of the effective perceived noise level at the

associated value of the engine related parameter,

t. ,95 ζ is obtained for ζ degrees of freedom. For the general case of a multiple regression

analysis involving K independent variables (i.e. K + 1 coefficients) ζ is defined as ζ = − −n K 1 (for the

specific case of a polynomial regression analysis, for which k is the order of curve fit, we have k variables

independent of the dependent variable, and so ζ = − −n k 1),

and

( )( )s

y y x

n K

i ii

i n

=−

− −=

=

∑2

1

1 , the estimate of σ , the true standard deviation.

3. CONFIDENCE INTERVAL FOR STATIC TEST DERIVED NPD CURVES

When static test data is used in family certifications, NPD curves are formed by the linear combinationof baseline flight regressions, baseline projected static regressions, and derivative projected static regressions inthe form:

EPNL EPNL EPNL EPNLDF BF BS DS= − +

or using the notation adopted above:

y x y x y x y xDF BF BS DS

( ) ( ) ( ) ( )0 0 0 0= − +

where subscript DF denotes derivative flight, BF denotes baseline flight, BS denotes baseline static,and DS denotes derivative static.

Confidence intervals for the derivative flight NPD curves are obtained by pooling the three data sets(each with their own polynomial regression). The confidence interval for the mean derived effective perceivednoise level at engine-related parameter x0 , i.e., for µDF x( )0 , is given by:-

( )CI x y x t v xDF DF90 0 0 0= ± ′( ) ( )

jskaleck



Appendix 178

where ( ) ( ) ( )v x s v x s v x s v xDF BF BF BS BS DS DS( ) ( ) ( ) ( )0 0

2

0

2

0

2= + +

with s s s v x v x v xBF BS DS BF BS DS, , , , and ( ) ( ) ( )0 0 0 computed as explained in Section

2.2 for the respective data sets indicated by the subscripts BF, BS, and DS, and

( ) ( ) ( )( ) ( ) ( )

′ =+ +

+ +t

s v x t s v x t s v x t

s v x s v x s v x

BF BF BF BS BS BS DS DS DS

BF BF BS BS DS DS

( ) ( ) ( )

( ) ( ) ( )

0

2

0

2

0

2

0

2

0

2

0

2

where t t tBF BS DS, and are the t. ,95 ζ values each evaluated with the respective degrees of

freedom ζ ζ ζBF BS DS, and as they arise in the corresponding regressions.

4. CONFIDENCE INTERVAL FOR ANALYTICALLY DERIVED NPD CURVES

Analysis may be used to determine the effect of changes in noise source components on certificatedlevels. This is accomplished by analytically determining the effect of hardware change on the noise component itgenerates. The resultant delta is applied to the original configuration and new noise levels are computed. Thechanges may occur on the baseline configuration or on subsequent derivative configurations. The confidence

intervals for this case are computed using the appropriate method from above. If $∆ represents the analyticallydetermined change and if it is assumed that it may deviate from the true unknown ∆ by some random amountd , i.e.

$∆ ∆= + d ,

where d is assumed to be normally distributed with mean zero and known variance τ2 ,

then the confidence interval for µ( )x0 + ∆ is given by

( )y x t v x( ) $ ( )0 0+ ± ′ ′∆

where ′ = +v x v x( ) ( )0 02 2τ and ′t is as above without change.

5. ADEQUACY OF THE MODEL

5.1 Choice of engine-related parameter

Every effort should be made to determine the most appropriate engine-related parameter x, which may bea combination of various simpler parameters.

5.2 Choice of regression model

It is not recommended in any case that polynomials of greater complexity than a simple quadratic beused for certification purposes, unless there is a clear basis for such a model.

Standard texts on multiple regression should be consulted and the data available should be examined toshow the adequacy of the model chosen.

jskaleck



Appendix 179

6. WORKED EXAMPLE OF THE DETERMINATION OF 90% CONFIDENCE INTERVALS FROM THE

POOLING OF THREE DATA SETS

This section presents an example of the derivation of the 90% confidence intervals arising from thepooling of three data sets. Worked examples and guidance material are presented for the calculation ofconfidence intervals for a clustered data set and for first order (ie. straight line) and second order (ie. quadratic)regression curves. In addition it is shown how the confidence interval shall be established for the poolingtogether of several data sets.

Consider the theoretical evaluation of the certification noise levels for an aircraft retro-fitted withsilenced engines. The approach noise level for the datum aircraft was derived from a clustered data set of noiselevels measured at nominally reference conditions, to which were added source noise corrections derived from aquadratic least squares curve fit through a series of data points made at different engine thrusts. In order toevaluate the noise levels for the aircraft fitted with acoustically treated engines a further source noise curve(assumed to be a straight least squares regression line) was established from a series of measurements of thesilenced aircraft. Each of the three data bases is assumed to be made up of data unique to each base.

The clustered data set consists of six EPNL levels for the nominal datum hardwall condition. Theselevels have been derived from measurements which have been fully corrected to the hardwall approach referencecondition.

The two curves which determine the acoustic changes are the regression curves (in the example givenboth a quadratic and straight line least squares curve fit) for the plots of EPNL against normalised thrust for thehardwall and silenced conditions. These are presented in figure below. The dotted lines plotted about each linerepresent the boundaries of 90% confidence.

CORRECTED NET THRUST PER ENGINE

90.0

91.0

92.0

93.0

94.0

95.0

96.0

97.0

98.0

99.0

100.0

1000 1200 1400 1600 1800 2000 2200

Hardwall

Silenced (y=89.93+0.001843x)

Figure 1.1

Each of the two curves is made up of from the full set of data points obtained for each condition duringa series of back to back tests. The least squares fits therefore have associated with them all the uncertaintiescontained within each data set. It is considered that the number of data points in each of the three sets is largeenough to constitute a statistical sample.

6.1 Confidence interval for a clustered data set

The confidence interval of the clustered data set is defined as follows:

Let EPNLi be the individual values of EPNL

n = number of data pointst = Student's t-distribution for (n-1) degrees of freedom (the number of degrees of

jskaleck



Appendix 180

freedom associated with a clustered data set).

Then the Confidence Interval CI EPNL ts

n= ±

where s , the estimate of the standard deviation, is defined as

( )s

EPNL EPNL

n

ii

i n

=−

−=

=

∑ 2

1

1

and EPNL

EPNL

n

i

i

i n

= =

=

∑1 .

Let us suppose that our clustered set of EPNL values consists of the following:

Run Number EPNL

1 95.82 94.83 95.74 95.15 95.66 95.3

Then number of data points (n) = 6 ,

degrees of freedom (n −1) = 5,

Student's t-distribution for 5 degrees of freedom = 2.015 (See Table 1-2),

EPNL

EPNL

n

ii

i n

= ==

=

∑1 95 38. ,

( )s

EPNL EPNL

n

ii

i n

=−

−==

=

∑ 2

1

10 3869. ,

and Confidence Interval

CI EPNL ts

n= ± = ± = ±95 38 2 015

0 3869

695 38 0 3183. .

.. .

6.2 Confidence interval for a first order regression curve

Let us suppose that the regression curve for one of the source noise data sets (for the silenced case)can best be represented by a least squares straight line fit ie. a first order polynomial.

The equation for this regression line is of the general form:

Y a bX= +

where Y represents the dependent variable EPNL ,

jskaleck



Appendix 181

and X represents the independent variable normalised thrust FN

δ ,(in this case).

Although for higher order polynomial least squares curves a regression line's coefficients (ie. thesolutions to the "normal equations") are best established through computer matrix solutions, the twocoefficients for a straight line fit, a and b , can be determined from the following two simple formulae for themeasured values of X and Y , X i and Yi :

bCovariance

Variance

S

Sxy

x

= =2

2

where S

X Y

n

X Y

nxy

i ii

i n

ii

i n

ii

i n

2 1 1 12

= −=

=

=

=

=

=

∑ ∑ ∑

and S

X

n

X

nx

ii

i n

ii

i n

2

2

1 1

2

= −

=

=

=

=

∑ ∑ ,

a

Y b X

n

ii

i n

ii

i n

=−

=

=

=

=

∑ ∑1 1

The 90% confidence interval about this regression line for X x= 0 is then defined by:

CI Y ts x A x90 01

0= ± ′−

where t = Student's t-distribution for 90% confidence corresponding to ( )n k− − 1 degrees of

freedom (where k is the order of the polynomial regression line and n is the number of data points),

( )x x0 01= and xx0

0

1′ =

,

A−1

is the inverse of A A X X where = ′ ,

with X and X′ defined as in paragraph 2.2 from the elemental vectors formed from the measured

values of independent variable X i ,

( )s

Y

n k

ii

i n

=− −

=

=

∑ ∆ 2

1

1

where ( )∆ Y i = the difference between the measured value of Yi at its associated value of X i ,and

the value of Y derived from the least squares fit straight line for X X i= ,and n and k are defined as

above.

Let us suppose that our data set consists of the following set of six EPNL values together with theirassociated values of engine related parameter (Note that it would be usual to have more than six data pointsmaking up a source noise curve but in order to limit the size of the matrices in this example their number has beenrestricted):

jskaleck



Appendix 182

RunNumber

FNδ

EPNL

1 1395 92.32 1505 92.93 1655 93.24 1730 92.95 1810 93.46 1850 93.2

Table 1-1

By plotting this data (See Figure 1.1) it can be seen by examination that a linear relationship between

EPNL (the dependent variable Y ) and FN

δ (the independent variable X ) is suggested with the following

general form:

Y a bX= +

The coefficients a and b of the linear equation are defined as above and may be calculated asfollows:

X Y XY X 2

1395 92.3 128759 19460251505 92.9 139815 22650251655 93.2 154246 27390251730 92.9 160717 29929001810 93.4 169054 32761001850 93.2 172420 3422500

X∑ Y∑ XY∑ X 2∑9945 557.9 925010 16641575

bCovariance

Variance

S

Sxy

x

= =2

2 where

( )( )S

X Y

n

X Y

nxy

i ii

i n

ii

i n

ii

i n

2 1 1 12

925010

6

9945 557 94846= − = − ==

=

=

=

=

=

∑ ∑ ∑ ..

36

and S

X

n

X

nx

ii

i n

ii

i n

2

2

1 1

2

216641575

6

9945

626289 6= −

= −

==

=

=

=

∑ ∑.

to give b = =48 46

26289 60 001843

.

..

and( )( )

a

Y b X

n

ii

i n

ii

i n

=−

=−

=−

=

=

=

∑ ∑1 1 557 9 0001843 9945

689 93

. ..

jskaleck



Appendix 183

The 90% confidence interval about this regression line which is defined as:

CI Y ts x A x90 01

0= ± ′−

is calculated as follows.

From the single set of measured independent variables tabulated in Table 1-1 let us form the matrix, X, from the elemental row vectors such that:

X =

1 1395

1 1505

1 1655

1 1730

1 1810

1 1850

,

and X′ , the transpose of X , where

X ′ =

1 1 1 1 1 1

1395 1505 1655 1730 1810 1850 .

We now form the matrix A , defined such that A X X= ′ , and so

A =

6 9945

9945 16641575 and its inverse A

−1 such that

A− =−

− −

1

175836 001051

001051 63396 6

. .

. .

E .

NB. The manipulation of matrices, their multiplication and inversion, are best performed by computersvia standard routines. Such routines are possible using standard functions contained within many commonlyused spreadsheets.

Suppose for example we now wish to find the 90% confidence interval about the regression line for a

value of FN

δ (i.e. x0) of 1600. We form the row vector x0 such that:

( )x0 1 1600= and its transpose , a column vector x0

1

1600′ =

.

From our calculation of A−1

we have:

( )x A01 1 1600

17 5836 001051

001051 6 3396 6− =

−− −

E

. .

. .

( )= − −0 7709 36453 4. . E

jskaleck



Appendix 184

and so ( )x A x01

0 0 7709 36453 41

1600− ′ = − −

. . E

= 0 1876. .

Our equation for confidence interval also requires that we evaluate the value of standard deviation forthe measured data set. From Table 1-1 and our regression equation for the least squares best fit straight line

(from which we calculate the predicted value of EPNL at each of the 6 measured values of FN

δ ) we proceed

as follows:

Run Number FNδ

EPNL(Measured)

EPNL(Predicted)

( )∆ EPNL2

1 1395 92.3 92.50 0.039792 1505 92.9 92.70 0.039113 1655 93.2 92.98 0.048964 1730 92.9 93.12 0.047085 1810 93.4 93.26 0.018386 1850 93.2 93.34 0.01909

( )s

y

n k

ii

i n

=− −

=− −

==

=

∑ ∆ 2

1

1

0 21241

6 1 102304

.. for n = 6 and k = 1.

and so taking the value of Student's t from Table 1-2 for ( )n k− − 1 degrees of freedom (i.e. 4) to be

2.132, we have the confidence interval about the regression line at FN

δ = 1600 defined as follows:

( )( )CI s A x901

0 92 98 2132 02304 01876 92 98 02128= EPNL t x 0± ′ = ± = ±− . . . . . .

In order to establish the lines of 90% confidence intervals about a regression line the values of CI90

for a range of values of independent variable(s) should be calculated, through which a line can be drawn. Theselines are shown as the dotted lines on Figure 1.1

6.3 Confidence interval for a second order regression curve

The confidence intervals about a second order regression curve are derived in a similar manner to thosefor a straight line detailed in Section 6.1. It is not felt that a detailed example of their calculation would beappropriate. However the following points should be borne in mind.

The coefficients of the least squares regression quadratic line are best determined via computer matrixsolutions. Regression analysis functions are a common feature of many proprietary software packages.

The matrices x x X0 0, , ′

and X′ formed during the computation of the confidence interval

according to the formula:

CI Y t s x A x90 01

0= ± ′−

are formed from 1 x 3 and 3 x 1 row and column vectors respectively, made up from the values of independentvariable X according to the following general form:

jskaleck



Appendix 185

( )x x x= 1 2 and x x

x

′ =

1

2

.

The number of degrees of freedom associated with a multiple regression analysis involving K variables

independent of the dependent variable (ie. with ( )K +1 coefficients, including the constant term) is defined as

( )n K− − 1 . For a second order regression curve we have two independent variables and so the number of

degrees of freedom is ( )n − 3 .

6.4 Confidence interval for the pooled data set

The confidence interval associated with the pooling of three data sets is defined as follows:

CI Y T Zii

i

= ± ′=

=

∑ 2

1

3

where ZCI

ti

i

i

= , with CIi = confidence interval for the i'th data set and ti = value of Student's t

for the i'th data set,

and T

Z t

Zi

i ii

i

ii

i= =

=

=

=

∑

∑

2

1

3

2

1

3 .

jskaleck



Appendix 186

The different stages in the calculation of the confidence interval at our reference thrust ofFN

δ = 1600 for the pooling of our three data sets is summarised below:

Description Function Datum Hardwall Silenced

Reference ThrustFN

δ 1600 1600

90% Confidence Intervalabout the mean CI90 0.3183 0.4817 0.2128

Number of data points n 6 23 6

Degree of curve fit k 0 2 1

Number of independentvariables K 0 2 1

Number of degrees offreedom n K− − 1 5 20 4

Student's t t 2.015 1.725 2.132

Z CIt

90 0.1580 0.2792 0.09981

Z 2CI

t90

2

2.4953E-2 7.7979E-2 9.9625E-3

Z t2CI

t t90

2

5.0280E-2 0.1345 2.1240E-2

Z 2∑ 0.1129

( )Z t2∑ 0.2060

T( )Z t

Z

2

2∑

∑1.8248

Z2∑ 0.3360

CI T Z 2∑ 0.6131

jskaleck



Appendix 187

7. BIBLIOGRAPHY

1. Hafner, New York, 1971. Kendal, M.G. and Stuart, A., The Advanced Theory of Statistics, Volumes 1and 2, 3

2. Yule, G.U. and Kendall, M.G., An Introduction to the Theory of Statistics, 14th ed., Griffin, New York,1950.

3. Walpole, R.E. and Myers R.H., Probability and Statistics for Engineers and Scientists. MacMillan,New York, 1972.

4. Cochran, W.G., Approximate Significance Levels of the Behrens-Fisher Test, Biometrics, 10 191-195,1964.

5. Rose, D.M. and Scholz, F.W., Statistical Analysis of Cumulative Shipper-Receiver Data, NUREG/CR-2819, Division of Facility Operations, Office of Nuclear Regulatory Research, U.S. Nuclear RegulatoryCommission, Washington, D.C. 20555, NRC FIN B1076, 1983.

6. Snedecor, G.W. and Cochran, W.G., Statisitical Methods, 6th ed., The Iowa State University Press,Arnes, Iowa, 1968.

7. Wonnacott, T.H. and Wonnacott, R.J., Introductory Statistics, 5th ed., John Wiley & Sons, 1990

jskaleck



Appendix 188

Degrees of Freedom(ζ)

t. ,95 ζ

1 6.3142 2.9203 2.3534 2.1325 2.0156 1.9437 1.8958 1.8609 1.83310 1.81212 1.78214 1.76116 1.74618 1.73420 1.72524 1.71130 1.69760 1.671

>60 1.645

Values in the Student's t-distribution to give a probability of 0.95 that the population mean value, µ , is

such that:

µ ζ≤ +y ts

n. ,95 , and thus a probability of 90% that

y ts

ny t

s

n− ≤ ≤ +. , . ,95 95ζ ζµ .

Student's t-DISTRIBUTION (FOR 90% CONFIDENCE INTERVAL)FOR VARIOUS DEGREES OF FREEDOM

TABLE 1-2

jskaleck



Appendix 289

Appendix 2

IDENTIFICATION OF SPECTRAL IRREGULARITIES

1.0 INTRODUCTION

Spectral irregularities which are not produced by aircraft noise sources may cause tone corrections to begenerated when the procedures of Annex 16, Volume 1 paragraph 4.3 of Appendix 1 and 2, are used. Thesespectral irregularities may be caused by:

a) the reflected sound energy from the ground plane beneath the microphone mounted at 1.2 m above it,interfering with the direct sound energy from the aircraft. The re-enforcing and destructive effects ofthis interference is strongest at lower frequencies, typically 100 Hz to 200 Hz and diminishes withincreasing frequency. The local peaks in the 1/3 octave spectra of such signals are termedpseudotones. Above 800 Hz this interference effect is usually insufficient to generate a tonecorrection when the Annex 16, Volume 1 tone correction procedures is used;

b) small perturbations in the propagation of aircraft noise when analysed with 1/3 octave bandwidthfilters; or

c) the data processing adjustments and corrections such as the background noise correction methodand the adjustment for atmospheric attenuation. In the case of the latter, the atmosphericattenuation coefficients (α) given in ARP866A ascribe α values at 4 kHz to the centre frequency ofthe 1/3 octave band whereas at 5 kHz the value of α is ascribed to the lower pass frequency of the1/3 octave. This difference is sufficient in some cases to generate a tone correction.

The inclusion of a tone correction factor in the computation of EPNL accounts for the subjectiveresponse to the presence of pronounced spectral irregularities. Tones generated by aircraft noise sources arethose for which the application of tone correction factors are appropriate. Tone correction factors which resultfrom spectral irregularities, i.e. false tones produced by any of the above causes may be disregarded. ThisAppendix describes methods which have been approved for detecting and removing the effects of such spectralirregularities. However, approval of the use of any of these methods remains with the certificating authority.

2.0 METHODS FOR IDENTIFYING FALSE TONES

2.1 Frequency tracking

Frequency tracking of flyover noise data is useful for the frequency tracking of spectral irregularities.The observed frequency of aeroplane noise sources decrease continuously during the flyover due to Dopplerfrequency shift, fDOPP where:-

ff

MDOPP =

−1 cosλ

where f is the frequency of the noise at sourceM is the Mach number of the aeroplaneλ is the angle between the flight path in the direction of flight and a line connecting the source

and observer at the time of emission.

Reflection related effects in the spectra, i.e. pseudotones, decrease in frequency prior to, and increase infrequency after, passing the closest point of aeroplane passing overhead, or the lateral point. Spectralirregularities caused by perturbations during the propagation of the noise from the aeroplane to the microphonetend to be random in nature in contrast to the Doppler effect. These differing characteristics can be used toseparate source tones from false tones.

jskaleck



Appendix 290

2.2 Narrow band analysis

Narrow band analysis with filter bandwidths narrower than those of 1/3 octaves is useful for identifyingfalse tones. For example when the analysis is produced such that the spectral noise levels at an instance arepresented in terms of image intensity on a line, the overall flyover analysis clearly indicates the Doppler shiftedaeroplane tones and those due to reflection as described in paragraph 2.1.

2.3 Microphone mounting height

Comparison of 1/3 octave spectra of measurements taken using the 1.2 m high microphone andcorresponding data obtained from a neighbouring microphone mounted flush on a hard reflecting surface, aconfiguration similar to that described in paragraph 4.4 of Appendix 6 of ICAO Annex 16, volume 1, or at a heightsubstantially greater than 1.2 m, such as 10 m, may be used to identify false tones. Changes to the microphoneheight alters the interference spectra irregularities from the frequency range of data from the 1.2 m highmicrophone and when a comparison is made between the two data sets collected at the same time, noise sourcetones can be separated from any false tones which may be present.

2.4 Inspection of noise time histories

Spectral irregularities which arise following data correction or adjustment (as described in (b) above) willoccur in the frequency range of between 1 kHz to 10 kHz and the resulting false tone corrections will normallyvary in magnitude between 0.2 to 0.6 dB. Time histories of perceived noise levels, PNL, and tone correctedperceived noise levels, PNLT, which exhibit constant level differences are often indicative of the presence offalse tone corrections. Supplementary narrow band analysis is useful to demonstrate that such tone correctionsare not due to aeroplane generated noise.

3.0 TREATMENT OF FALSE TONES

When spectral irregularities give rise to false tones which are identified by, for example, the methodsdescribed in Section 2 of this Appendix, their value, when computed in Step 9 of the tone correction calculation(described in Section 4.3 of Appendix 2 of Annex 16, Volume 1) may be set to zero.

jskaleck



Appendix 391

Appendix 3

A PROCEDURE FOR REMOVING THE EFFECTS OF AMBIENT NOISELEVELS FROM AEROPLANE NOISE DATA

1.0 INTRODUCTION

1.1 The following information is provided as guidance material for certificating authorities on themethod of removing the effect of ambient noise on aeroplane recorded noise.

1.2 This is not the only procedure which may be used and changes under certain instances may bemade to it, but approval for its use in its current or modified form remains with the certificating authority.

2.0 Correction procedure

2.1 Aeroplane sound pressure levels within the 10 dB down points should exceed the mean ambientnoise levels determined in section 7.5.6 of the manual by at least 3 dB in each one-third octave band or becorrected by the following or similar method.

1) The identification of the pre-detection and post-detection noise are made, i.e.:

a) one which adds to the recorded noise data on an energy basis, such as that from extraneousacoustic background noise signals or electrical noise from the microphone pre- amplifierand tape recorder is termed pre-detection noise; and

b) one which is non-additive but masks the aeroplane noise signal, such as would beproduced by the presence of a lower level 'window' of the signal analyser, is termed post-detection noise.

2) Over the frequency range of the pre-detection noise, the background noise is subtracted from theanalysed noise on an energy basis.

a) at frequencies of 630 Hz and below, if the analysed level is within 3 dB of the backgroundpre-detection noise level ('masked' band), the corrected aeroplane noise is set equal to thepre-detection background level. If the analysed level is less than the background level, nochanges are made to this level; and

b) at frequencies above 630 Hz, if the analysed level is within 3 dB or less than the pre-detection noise level, these levels are also identified as 'masked' and are corrected as inSteps 4), 5) and 6).

3) The remaining bands which fall inside the frequency range of the post-detection background noiseare uncorrected unless they are within 3 dB of the identified post-detection noise, these bands are thus identifiedas 'masked' bands.

4) The 'as measured' spectrum is normalised to reference day conditions (25 C, 70% RH) and a distancefrom source of 60 m.

5) For the 'masked' high frequency bands at 60 m a linear extrapolation from the next lower frequencyunmasked band of 0 dB/one-third octave, or a greater slope if derived from measured data, is applied.

6) The normalised spectrum is then converted back to 'as measured' distances and atmosphericconditions.

Note: If 'masked' pre- or post-detection bands are surrounded on both sides by 'unmasked' bands, nocorrections are applied.

jskaleck



Appendix 392

2.2 The following procedure is applicable to measured data where more than seven consecutive one-third octave bands are masked during a period of the noise time history.

Corrections by extrapolating in the time domain for frequencies above 630 Hz are made by taking intoaccount propagation distance (spherical divergence and atmospheric attenuation) relative to the first(approaching) or last (receding) unmasked sound pressure level measurement in the one-third octave bandrequiring correction.

Source directional characteristics in each masked frequency band are assumed to be either:

a) directional, as supported by the applicant's test data or in the absence of such data; or

b) omnidirectional.

jskaleck



Appendix 493

Appendix 4

REFERENCE TABLES AND FIGURES USED IN THE MANUAL CALCULATION OF EFFECTIVE PERCEIVEDNOISE LEVEL

This Appendix contains material useful in the manual calculation of Effective Perceived Noise Level. Suchmanual calculations are often required to verify the accuracy of computer programs used for calculatingnoise certification levels.

jskaleck



Appendix 494

jskaleck



Appendix 495

jskaleck



Appendix 496

jskaleck



Appendix 497

jskaleck



Appendix 498

jskaleck



Appendix 599

Appendix 5

WORKED EXAMPLE OF CALCULATION OF REFERENCE FLYOVER HEIGHT AND REFERENCECONDITIONS FOR SOURCE NOISE ADJUSTMENTS FOR CERTIFICATION OF LIGHT PROPELLER

DRIVEN AEROPLANES TO ICAO ANNEX 16, VOLUME 1, CHAPTER 10

1. INTRODUCTION

The reference flyover height for an aeroplane certificated to Chapter 10 of Annex 16 is defined for apoint 2500 m from start of roll beneath a reference flight path determined according to the take-off referenceprocedure described in Paragraph 10.5.2 of the Annex. An expression for the reference flyover height in terms ofcommonly approved performance data and an example of how such an expression may be worked are presentedin this Appendix. The relationship between this reference height and the conditions to which source noisecorrections are to be made is also explained.

2. TAKE-OFF REFERENCE PROCEDURE

The take-of reference procedure for an aeroplane certificated to Chapter 10 is defined under sea level,ISA conditions, at maximum take-off mass for which noise certification is requested, in Paragraph 10.5.2 of theAnnex. The procedure is described in terms of two phases.

The first phase commences at "brakes release" and continues to the point where the aircraft has reacheda height of 15 m (50 ft) above the runway (the point of interception of a vertical line passing through this pointwith a horizontal plane 15 m below is often referred to as "reference zero").

The second phase commences at the end of the first phase and assumes the aeroplane is in normal climbconfiguration with landing gear up and flap setting normal for "second segment" climb.

Note that in this respect the reference "acoustic" flight path ignores the "first segment" part of the flightpath, during which the aircraft accelerates to normal climb speed and, where appropriate, landing gear and flapsare retracted.

3. EXPRESSION FOR REFERENCE HEIGHT

The reference flyover height is defined according to the take-off reference flight path at a point 2500 mfrom start of roll for an aeroplane taking-off from a paved, level runway under the following conditions:

− sea level atmospheric pressure of 1013.25 hPa;− ambient air temperature of 15°C, i.e. ISA− relative humidity of 70 per cent; and− zero wind. This height can be defined in terms of the approved take-off and climb performance figures for the

conditions described above as follows:

( ) ( )( )H D RC VR y= − +−2500 15151tan sin Equation 1

where: HR is the reference height in metres;

D15 is the sea level, ISA take-off distance in metres to a height of 15 m at the maximumcertificated take-off mass and maximum certificated take-off power;

jskaleck



Appendix 5100

RC is the sea level, ISA best rate of climb (m/s) at the maximum certificated take-off mass andthe maximum power and rpm that can be continuously delivered by the engine(s) during thissecond phase; and

Vy is the best rate of climb speed (m/s) corresponding to RC.

The performance data in many flight manuals is often presented in terms of non SI units. Typically thetake-off distance (expressed in feet) is given to a height of 50 ft, the rate of climb is expressed in feet

per minute and the air speed in knots. In such instances the expression for reference flyover height, HR

ft, becomes:

( ) ( )( )H D RC VR y= − +−8203 1014 50501tan sin . Equation 2

where: D50 is the sea level, ISA take-off distance in feet to a height of 50 ft;

RC is the sea level, ISA best rate of climb (ft/m); and

Vy is the best rate of climb speed (kt).

The performance figures can normally found in the performance section of an aircraft's flight manual orpilot's handbook. Note that for certain categories of aircraft a safety factor may be applied to the take-off andclimb performance parameters presented in the flight manual. In the case of multi-engined aircraft it may beassumed that one engine is inoperative during part of Phase 1 and during Phase 2. For the purpose of calculatingthe "acoustic" reference flight path the take-off distance and rate of climb should be determined for all enginesoperating using gross, i.e. unfactored, data.

In addition Vy, the best rate of climb speed, used in the expression above is defined as the true air speed(TAS). However in the flight manual speed is normally presented in terms of indicated airspeed (IAS). Thisshould be corrected to the calibrated airspeed (CAS) by applying the relevant position error and instrumentcorrections for the airspeed indicator. These corrections can also be found in the manual. For an ISA day at sealevel the TAS is then equal to the CAS.

4. REFERENCE CONDITIONS FOR SOURCE NOISE ADJUSTMENTS

Paragraphs 5.2c and 5.2d of Appendix 6 of the Annex 16, Volume 1 describe how corrections fordifferences in source noise between test and reference conditions shall be made.

The reference helical tip Mach number and engine power are defined for the reference conditions abovethe measurement point, i.e. the reference atmospheric conditions at the reference height, HR.

The reference temperature at this height is calculated under ISA conditions, i.e. for an ambient sea leveltemperature of 15°C and assuming a standard temperature lapse rate of 1.98°C per 1000 ft. The referencetemperature, TR °C, can be defined as:

( )T HR R= −15 198 1000. Equation 3

The reference atmospheric pressure, PR hPa, is similarly calculated at the reference height for a standardsea level pressure of 1013.25 hPa assuming a standard pressure lapse rate:

( )[ ]P HR R= − −101325 1 6 7862 10 65 325

. ..

x Equation 4

jskaleck



Appendix 5101

5. WORKED EXAMPLE

A worked example is presented for the calculation of reference flyover height and the associatedreference atmospheric conditions.

4.1 Example of reference flyover height calculation

In Figure 5-1 extracts are presented from the performance section of a flight manual for a typicallight , single engined propeller driven aeroplane.

The introduction contains a statement to the effect that the information is derived from"measured flight test data" and includes "no additional factors".

The take-off distance to 50 ft at the reference conditions of Chapter 10 can be read from thetable of take-off distances presented for a paved runway at the maximum certificated take-off weight of1920 lb. Thus D50 is 1370 ft.

The rate of climb at the reference conditions can similarly be read from the rate of climb table.Thus RC is 1000 ft/m.

The climb speed associated with the rate of climb figures is given as 80 kIAS. Thecorresponding true air speed at the reference conditions of Chapter 10 is equal to the indicated airspeedcorrected according to the airspeed calibration table at the appropriate flap setting of 0°. Thus Vy is 81kTAS.

Entering these parameters into the expression for reference height (ft) given in Equation 2gives:

( ) ( )( )HR = − × +−8203 1370 1000 1014 81 501tan sin .

and so HR = 888 ft .

4.2 Example of calculation of reference atmospheric conditions

The reference temperature at the reference height, HR, is given by Equation 3:

( )TR = −15 198 888 1000.

and so TR = °132. C .

The reference pressure at this height is given by Equation 4:

( )[ ]PR = − −1013 25 1 67862 10 88865 325

. ..

x x

and so PR = 981 hPa .

jskaleck



Appendix 5102

SECTION 5

PERFORMANCE

1. INTRODUCTION

The data processed in this section enables flight planning to be carried out for flightsbetween airfields with various altitudes, temperatures and field lengths. Theinformation is derived from measured flight test data using CAA approved methodsand factors to cover all the conditions shown. The data assumes average pilot skilland an aircraft engine and propeller in good condition.

No additional factors are included and it is the pilots responsibility to apply safetyfactors which must not be less than those.....

6. AIRSPEED CALIBRATION

0°Flap

KIASKCAS

--

6061

7071

8081

9091

100101

110111

120121

130131

180181

15°Flap

KIASKCAS

5051

6061

7071

8081

8586

--

--

--

--

--

35°Flap

KIASKCAS

5050

6059

7069

8079

8584

--

--

--

--

--

Figure 5-1. Example of Flight Manual Performance Section

TAKE-OFF DISTANCE - PAVED RUNWAY (1)CONDITIONSFlaps - 15°....................Rotation speed - 53 KIASPower - Full throttle......Speed at 50 ft - 65 KIASWeight - 1920 lbs

AIRFIELD ISA -20°C ISA -10°C ISAHEIGHTFT

GrndRoll

Totalto50 ft

GrndRoll

Totalto50 ft

GrndRoll

Totalto50 ft

SeaLevel

530 1230 565 1290 600 1370

5000 1045 2835 1065 2435 1090 258010000 1465 3335 1490 3390 1510 3435AIRFIELD ISA +10°C ISA +20°C ISA +30°CHEIGHTFT

GrndRoll

Totalto50 ft

GrndRoll

Totalto50 ft

GrndRoll

Totalto50 ft

Sea Level 700 1580 750 1715 840 19005000 1170 2670 1295 2840 1290 290510000 1575 3560 1610 3695 1670 3790

RATE OF CLIMBCONDITIONSFlaps UPFull throttleWeight - 1920 lbsSpeed - 80 KIAS

PRESSURE RATE OF CLIMB FEET/MINUTEALTITUDEFT

ISA-20°C ISA ISA+10°C ISA+20°C

Sea Level 1035 1000 915 8251000 980 945 860 7702000 925 890 805 7203000 870 830 750 6654000 815 775 695 6105000 765 720 640 5606000 700 665 585 5057000 635 605 560 4508000 570 550 475 3959000 495 480 410 33510000 415 405 335 270

jskaleck



Appendix 6103

Appendix 6

NOISE DATA CORRECTIONS FOR TESTS AT HIGH ALTITUDE TEST SITES

1. INTRODUCTION

Jet noise generation is somewhat suppressed at higher altitudes due to the difference in the engine jetvelocity and jet velocity shear effects resulting from the change in air density. Use of a high altitude test site forthe noise test of an aeroplane model that is primarily jet noise dominated should include making the followingcorrections. These jet source noise corrections are in addition to the standard pistonphone barometric pressurecorrection of about 0.1 dB/100 m (0.3 dB /1000 ft) which is normally used for test sites not approximately at sealevel, and applies to tests conducted at sites at or above 366 m (1200 ft) mean sea level (MSL).

2. JET NOISE SOURCE CORRECTION

2.1 Flight test site locations at or above 366 m (1200 ft) MSL, but not above 1219 m (4000 ft)MSL, may be approved provided the following criteria (Figure 6-1) are met and source noise corrections(paragraph 2.3) are applied. Alternative criteria or corrections require the approval of the certificatingauthority.

2.2 Criteria

Jet source noise altitude corrections from paragraph 2.3 are required for each one-half secondspectrum when using the integrated procedure and at the PNLTM spectrum when using the simplifiedprocedure (see paragraph 9.3 and 9.4 of Appendix 2 of Annex 16, Volume 1), and are to be applied inaccordance with the following criteria:

Figure 6-1

2.3 Correction Procedure

An acceptable jet source noise correction is as follows:

a) Correct each one-half second spectrum (or PNLTM one-half second spectrum, asappropriate) in accordance with the criteria of paragraph 2.2 using the following equation:

( ) ( ) ( )[ ] [ ] [ ]∆SPL = 10log d d c c k u u F FR T T R R T+ +50 10 1 2log log

where: Subscript T denotes conditions at the actual aeroplane test altitude above MSL understandard atmospheric conditions, i.e. ISA+10°C and 70% relative humidity;

jskaleck



Appendix 6104

Subscript R denotes conditions at the aeroplane reference altitude above MSL (i.e.aeroplane test altitude above MSL minus the test site altitude) under standardatmospheric conditions, i.e. ISA+10°C and 70% relative humidity;

dR is the density for standard atmosphere at the aeroplane reference altitude in kg/m3

(lb/ft3;

dT is the density for standard atmosphere at the aeroplane test altitude in kg/m3 (lb/ft3;

cR is the speed of sound corresponding to the absolute temperature for standardatmosphere at aeroplane reference altitude in m/s (ft/s);

cT is the speed of sound corresponding to the absolute temperature for standardatmosphere at aeroplane test altitude in m/s (ft/s);

k = 8, unless an otherwise empirically derived value is substantiated;

u = (ve - va) is the equivalent relative jet velocity in m/s (ft/s)

where: ve is the equivalent jet velocity as defined in SAE ARP876D, Appendix C(January 1994) and obtained from the engine cycle deck in m/s (ft/s); and

va is the aircraft velocity in m/s (ft/s)

uR is the equivalent relative jet velocity in m/s (ft/s) where ve is determined at N1CTEST

for standard atmosphere at the aeroplane reference altitude;

uT is the equivalent relative jet velocity in m/s (ft/s) where ve is determined at N1CTEST

for standard atmosphere at the aeroplane test altitude;

N1C is the corrected engine rpm ( )N T1 2θ ;

F1 is a factor corresponding to the percentage of applied correction related toacoustic angle in Figure 6-1 (values range from 0.00 to 1.00); and

F2 is a factor corresponding to the percentage of applied correction related to the one-third octave band in Figure 6-1 (values range from 0.00 to 1.00).

b) For each one-third octave band SPL, arithmetically add the altitude jet noise correction in a)above to the measured SPL's to obtain the altitude source jet noise corrected SPL's forparagraph 4.1.3a of Appendix 2 of Annex 16, Volume 1.

c) The above altitude correction is to be applied to all measured test data including approachconditions (unless it can be substantiated that the jet noise during approach does notcontribute significantly to the total aircraft noise).

jskaleck


CAEP Working Groups Environmental Technical Manual on the use ofRecommended Revision 7 Procedures in the Noise Certification of Aircraft

105

Appendix 7

TECHNICAL BACKGROUND INFORMATION ON THE GUIDELINES FOR THE NOISE CERTIFICATION OFTILTROTOR AIRCRAFT

1. INTRODUCTION

The Guidelines for the Noise Certification of Tiltrotor Aircraft presented in Attachment YY of ICAO Annex 16,Volume 1, have been developed by the ICAO CAEP Tiltrotor Task Group specifically for the noise certification ofthe Bell/Agusta 609, the first example of a civil tiltrotor aircraft. Nevertheless it is intended that these guidelinesbe used as the basis for noise certification of subsequent tiltrotor aircraft. The following explanatory material isintended to give an insight as to how the guidelines have been developed, particularly with regard to theirapplication to the Bell/Agusta 609. It is hoped that the information may serve as a useful guide to thedevelopment of the guidelines for use with other tiltrotor aircraft and their possible eventual adoption into theAnnex as a Standard.

2. GENERAL TECHNICAL INFORMATION

2.1 Aeroplane mode, helicopter mode

The term “aeroplane mode” is used for the situation where the rotors are orientated with their axis ofrotation substantially horizontal (i.e. engine nacelle angle near 0 degrees on the “down stops”, see below). Theterm “helicopter mode” is used where the rotors are orientated with their axis of rotation substantially vertical(nacelle angle around 90 degrees). In the guideline the latter condition is referred to as the “VTOL/Conversionmode”, which is the term used in the airworthiness standards in development for the Bell/Agusta 609. VTOLstands for Vertical Take-Off and Landing.

2.2 Nacelle angle

The “nacelle angle” is defined as the angle between the rotor shaft centreline and the longitudinal axisof the aircraft fuselage. The nacelle is normally perpendicular to the plane of rotation of the rotor.

2.3 Gates

In the design of the Bell/Agusta 609 there are a number of preferred nacelle angle positions called“gates”. These are default positions that will normally be used for normal operation of the aircraft. The nacelleangle is controlled by a self-centring switch. When the nacelle angle is 0 degrees (aeroplane mode) and the pilothits the switch upwards, the nacelles will automatically turn to a position of approximately 60 degrees, where itwill stop. Hitting the switch once more will make the nacelle turn to a position of approximately 75 degrees.Above 75 degrees the nacelle angle can be set to any angle up to approximately 95 degrees by holding the switchup (or down to go back). The “gate”-concept is expected to be typical for all future tiltrotors, although thenumber and position of the gates may vary. The gates play an important role in the airworthiness requirements,where they are defined as “authorised fixed operation points in the VTOL/conversion mode”. When the aircraftis flying in the aeroplane mode, the nacelle angle will be in line with the longitudinal axis of the aircraft. In thiscase the angle is fixed using the so-called “down stop”

2.4 Rotor RPM

The design of the Bell/Agusta 609 and most likely future designs of tiltrotors will have at least twopossible Rotor RPM’s: one for the helicopter mode and another (lower) RPM for the aeroplane cruising mode.The lower RPM can only be used when the nacelles are on the down stop. Before leaving the down stop, theRPM must be set to the higher value to be able to hover.

2.5 Form and extent of the guideline

jskaleck



106

It is considered that at this moment there is not enough experience with tiltrotor aircraft to justifyadoption of firm standards. Therefore guidance material has been developed as “green pages” for the Annex,much like the guidelines for noise certification of Propeller driven STOL aeroplanes that are already in greenpages.

It was deemed desirable to give the same level of detail as is found in comparable chapters of the Annexand for instance include information on date of applicability to promote a uniform application of the guidelines.

After careful deliberations it has been concluded that the current standards of Chapter 8 were a goodbasis for the guidelines and that the differences between the guideline and Chapter 8 should be minimised:

• The noise from tiltrotors will be most prominent during departure and approach. In these situationstiltrotors will normally operate in or near the “helicopter mode”.

• In the development of the guidelines the noise of the Bell XV15 tiltrotor aircraft (which serves as aprototype for the Bell/Agusta 609) has been observed. It was concluded that the character of thenoise of this aircraft was much like that of a normal helicopter.

• In horizontal flyover, the “helicopter mode” will normally be the noisiest configuration.

• The proposed guidelines are confined to tiltrotors that can only take off vertically, excluding thosehaving STOL-characteristics. They will operate much like normal helicopters, with relatively steeptake-off and approach paths.

• The level of available noise abatement technology for tiltrotors is considered to be the same as forhelicopters.

• Tiltrotors operations will often mix with helicopter operations from the same heliport. Thereforethere will be a desire to compare the noise from tilt-rotors and helicopters.

2.6 Transition phase noise

One of the items of strongest interest is the transition from one nacelle angle to the other. It could bethat this would be associated with particular noise generation mechanisms. For example, when one considers thetiltrotor transition from aeroplane mode to helicopter mode while decelerating there is a phase in which thecomponent of the speed vector perpendicular to the rotor changes from “top to bottom” to “bottom to top”. Itwould be conceivable that sometime during the transition phase blade vortices would be ingested or another nonstationary effect would create additional noise.

A number of overflights of the Bell XV15 were listened to, one of which was especially set up to studythe noise during transition. In this run the tiltrotor (XV15) passed overhead at 500 ft while transitioning fromaeroplane to helicopter mode. No special phenomena were noticed during this flight. Also during the other runs,in which there were demonstrations of hover, hover turns, sideward flight, take-off, level flyovers at variousspeeds/nacelle angles combinations and approaches at 6 and 9 degrees, no particularities were heard, other thannormal Blade Vortex Interaction noise during both the 6 and 9 degree approaches. During the procedures to setthe aircraft up for the various runs several transitions were made from helicopter to aeroplane mode and back,which were heard from different positions relative to the aircraft. No particular noise was heard.

Based on this experience and the arguments stated below it has been decided not to attempt to define aspecial test point aimed at catching transition noise of tiltrotors.

The arguments for this are:

• Experienced observers from industry claim they never noticed any particular noise phenomenaassociated with the transitional phase. This was backed up by the specific observations of the BellXV15 referred to above.

jskaleck


CAEP Working Groups Environmental Technical Manual on the use ofRecommended Revision 7 Procedures in the Noise Certification of Aircraft

107

• The conversion rate is relatively slow, which means that during the whole conversion process theflow field also changes very slowly.

• If there would be a transitional noise it would probably be related to some form of Blade VortexInteraction. This phenomenon is covered under the approach procedure and it might be hard tojustify adding a measurement point to get some additional information.

• Defining a reproducible and practicable procedure to catch the transition noise which nobody hasever noticed phenomenon is virtually impossible.

• If in future there would be a design that would have clear transitional noise characteristics the effectcould be studied and if deemed necessary an amendment to the guideline could be proposed.

3. COMMENTS TO INDIVIDUAL SECTIONS OF THE PROPOSED GUIDELINE

Following the order of appearance in the proposed guideline the following comments are offered:

3.1 Definition (re Note 1 of Attachment YY)

The proposed definition was proposed by ICCAIA. It focuses on the fundamental difference betweentiltrotors and other aircraft.

3.2 Applicability (re note 2 and section 1.1 of attachment YY)

As mentioned before, an applicability section has been to promote uniform application of the standards.The reference to derived versions has the effect that no measurements are required on aircraft that are quieterthan their parents, due to the definition of derived versions in Annex 16. The date chosen is the date when thissection of the guideline was discussed.

3.3 Noise Evaluation Measure (re section 2)

In view of the commonality with helicopters, the same units as used for Chapter 8 are proposed. It isproposed not to create a new appendix for tiltrotors, since the current appendix 2 is considered to be appropriate.For land use planning purposes it is proposed to have additional data made available. Which data should beprovided is left to be determined between the authority and the applicant, since the needs of different authoritiesin this respect may differ. At this moment, the intention of this section of the guideline is to only require data thatcan be gathered through additional analysis of the data already measured for certification purposes. It is hopedto have SAE investigate further the data requirements for Land Use Planning. Since the information needed forLand Use Planning can be of such detail that it is commercially sensitive, it is not the intent to make it available tothe public.

3.4 Noise Measurement Reference Points (re section 3)

In view of the desired commonality with helicopters, the same Noise measurement reference points asused for Chapter 8 are proposed.

3.5 Maximum Noise Levels and trade-offs. (re section 4 and 5)

For the reasons already mentioned under “form” it is considered that the current Chapter 8 limits andtrade-offs to be a good starting point for use in the guideline. In helicopter mode both the lift technology andoperating environment are similar to that of a helicopter. If technology requires higher levels or makes possiblelower levels this should be considered by the individual authority when using the guidelines in a particular case.For the flyover case there is only a limit specified for the helicopter mode, since this is normally the noisiestconfiguration, and also the most likely configuration to be used when flying the circuit pattern.

3.6 Noise Certification reference procedures (re section 6)

jskaleck



108

The capability to change the nacelle angle and the two different, possibly more RPMs require someadditions to the current Chapter 8 helicopter reference procedures.

3.7 RPM

In the guideline the RPM required is linked to the corresponding flight condition. This means that fortake-off, approach and flyover in the helicopter mode the higher RPM will have to be used, while for the flyoverin aeroplane mode the lower RPM has to be used.

3.8 Nacelle angle

In take-off the choice of the nacelle angle is left to the applicant. This is in line with the philosophyelsewhere in the Annex, where the choice of the configuration is left to the applicant. It is also in line with therequirement in Chapter 8 to use Vy, since the applicant will normally choose the nacelle angle that is close to thenacelle angle that corresponds to the overall best rate of climb. (Note that for each nacelle angle there is a speedthat gives the best rate of climb, which is normally not the same numerical value for different nacelle angles.There will be one nacelle angle that gives the highest overall rate of climb, but this is usually not an angle thatcorresponds to a “gate”.)

For the case of Fly-over in helicopter mode the definition of the nacelle angle to be used was one of themore difficult problems. Initially it was proposed to use a nacelle angle of 90 degrees, comparable to a helicopter.This was however unsatisfactory because a tiltrotor will normally not fly at this angle at the high speed requiredfor noise certification. Normally the rotor will be tilted to get more forward thrust without tilting the fuselageforward by selecting a nacelle angle of approximately 80 degrees. It was agreed that this unique capability of thetiltrotor should be incorporated in the reference procedure. On the other hand the requirement should prevent theapplicant from choosing a nacelle angle that would be to close to zero degrees since this would giveunrealistically low noise figures. (Note that tilting the rotor will reduce the advancing blade tip mach number).After long deliberations a satisfactory solution was found: For a tiltrotor there will normally be a nacelle anglebelow which hover is no longer possible and for which flight with zero airspeed is not permitted. It was decidedto fix the nacelle angle for the flyover in helicopter mode to the gate closest to that angle.

In the fly-over in aeroplane mode the nacelle angle is defined as on the down stop, the position that willnormally be used for cruise and high speeds. Two conditions are measured. One is with the high RPM and thesame speed as used in the helicopter mode flyover. This condition is intended to give the possibility to makecomparisons between the helicopter mode and aeroplane mode flyover. The other condition is with the cruiseRPM and speed Vmcp or Vmo, which is intended to represent a worst case cruise condition.

For the approach reference configuration the nacelle angle for maximum approach noise should beused. This is in line with the philosophy in Chapter 8 and other parts of the Annex that require the noisiestconfiguration for approach. This will normally require testing several different nacelle angles in order todetermine which is noisiest.

In the tiltrotor aircraft design the flap angle varies with airspeed the pilot may manually set flaps or mayuse autoflap control in order to reduce the pilot workload. In this latter case the flap angle for noise certificationwill be the flap angle that is normal for the approach configuration and approach condition flown. For a designwith pilot-controlled flap angle the applicant should use the flap angle designated for approach and will have toprove that the noisiest configuration is used for noise certification.

3.9 Test procedures (re section 7)

The test procedures are the same as in Chapter 8. Note that this means that as a minimum all data aretaken and evaluated at 1.2 m., including data taken for Land Use Planning purposes. This is proposed to maintaincommonality with Chapter 8 numbers and to reduce costs for the applicant. If for Land Use Planning or otherpurposes it were desired to have data at other microphone positions (i.e. at ground plane) this would of coursebe allowed but would have to be agreed between applicant and certificating authority.

jskaleck



Appendix 8109

Appendix 8

REASSESSMENT CRITERIA FOR THE RE-CERTIFICATION OF AN AEROPLANEFROM ICAO ANNEX 16, VOLUME 1, CHAPTER 3 TO A MORE STRINGENT STANDARD

1. INTRODUCTION

Re-certification is defined as the “Certification of an aircraft, with or without revision to noise levels, to aStandard different to that which it had been originally certified”.

This Appendix is restricted to the assessment of existing approved noise levels associated withapplications for the re-certification of an aeroplane from ICAO Annex 16, Volume 1, Chapter 3 to a more stringentstandard. It may be developed in the future to include guidelines for the re-certification of aircraft speciallymodified for the purpose of achieving compliance with a new standard and the re-certification of helicopters andlight propeller driven aeroplanes to a different standard from that to which they were originally certificated.

In the case of an aircraft being re-certificated from the standards of ICAO Annex 16, Volume 1, Chapter 3to a more stringent standard that may be directly applicable to new type designs only, noise re-certificationshould be granted on the basis that the evidence used to determine compliance is as satisfactory as the evidenceexpected of a new type design. In this respect the date used by a certificating authority to determine the re-certification basis should be the date of acceptance of the first application for re-certification.

2. REASSESSMENT CRITERIA

Noise levels already approved to Chapter 3 and submitted in support of applications for re-certificationof existing aircraft should be assessed against the criteria presented below. These criteria have been developedto ensure satisfactory compliance with the new standard. The criteria consist of a list of simple questionsconcerning the manner in which the original Chapter 3 data was obtained and subsequently processed. Thequestions are the result of a comparison of the various amendments and revisions to the Annex andEnvironmental Technical Manual to which an aircraft’s existing Chapter 3 noise levels may have been approved.

2.1 For aeroplanes which were approved in accordance with Amendment 5, or higher, of ICAOAnnex 16, Volume 1, a reassessment is not required. The aeroplane’s existing, approved Chapter 3 noiselevels shall be used to determine compliance with the new standard.

2.2 For aeroplanes which were approved in accordance with Amendment 4, or lower, of ICAOAnnex 16, Volume 1, the applicant shall be required to show that the existing approved Chapter 3 noiselevels are equivalent to those approved to Amendment 5 by answering the following questions(paragraph references refer to either Amendment 5 of ICAO Annex 16, Volume 1 or WGAR/6 of thisManual:

For all aeroplanes:a) Was full take-off power used throughout the reference flight path in the determination of

the lateral noise level? (Annex 16, paragraph 3.6.2.1c refers);b) Was the “average engine” rather than the “minimum engine” thrust or power used in

the calculation of the take-off reference flight path? (Annex 16, paragraphs 3.6.2.1a and3.6.2.1g refer);

c) Was the “simplified” method of adjustment defined in Appendix 2 of the Annex used and,if so, was -7.5 used as the factor for the calculation of the noise propagation path durationcorrection term? (Annex 16, Appendix 2, paragraph 9.3.3.2 refers)

d) Was the take-off reference speed between V2+10 kt and V2+20 kt? (Annex 16, paragraph3.6.2.1d refers); and

e) Was the four ½ s linear average approximation to exponential averaging used and, if so,were the 100% weighting factors used? (Annex 16, Appendix 2, paragraph 3.4.5, Note 1refers).

jskaleck



Appendix 8110

For jet aeroplanes only:f) Were the noise measurements conducted at a test site below 366 m (1200 ft) and, if not,

was a jet source noise correction applied? (Technical Manual, Appendix 6 refers);g) Do the engines have bypass ratios of more than 2 and, if not, was the peak lateral noise

established by undertaking a number of flights over a range of heights? (TechnicalManual, paragraph 2.1.3.2b refers);

h) In the event that “family” certification methods were used were the 90% confidenceintervals for the pooling together of flight and static engine test data establishedaccording to the Technical Manual guidance? (Technical Manual, Appendix 1 refers); and

i) Do the engines have bypass ratios of 2 or less and, if not, in the event that “family”certification methods were used, did all associated static engine tests involve the use of aturbulence control screen (TCS) or inflow control device (ICD)? (Technical Manual,paragraph 2.3.3.4.1 refers).

For propeller driven aeroplanes only:j) Were symmetrical microphones used at every position along the lateral array for the

determination of the peak lateral noise level?(Annex 16, paragraph 3.3.2.2);k) Was the approach noise level demonstrated at the noisiest configuration? (Annex 16,

paragraph 3.6.3.1e refers); andl) Was the target airspeed flown during the flight tests appropriate to the actual test mass

of the aeroplane? (Technical Manual, paragraph 3.1.2a).

If the applicant is able to answer in the affirmative, to the satisfaction of the certificating authority, allthe questions that may be relevant then reassessment is not required. The aeroplane’s existing,approved Chapter 3 noise levels shall be used to determine compliance with the new standard.Otherwise the certificating authority may require such re-analysis or re-testing that may be necessary inorder to bring the data up to the required standard.

jskaleck


environmental technical manual on the use of procedures in the

Documents