# AC 36-4B: Noise Certification Handbook - Type: Advisory circular (guidance, not mandatory unless a rule or AD requires it) - Status: Historical - Version in force: [AC 36-4D](https://openairdocs.org/ac/36-4), approved October 12, 2017. This is a historical version - Approved: March 23, 1988 - Last reviewed: January 15, 2003 - Family: AC 36-4 - Cancelled by: [AC 36-4C](https://openairdocs.org/ac/36-4C) - 14 CFR parts: 36 - Pages: 66 - PDF: https://openairdocs.org/files/sha256/33/54/3354b2d5e459a600de8b8a7f6c5b83c41ee3392ac0072cd30cff0c2d053a3f51 - FAA DRS: https://drs.faa.gov/browse/excelExternalWindow/206BE9CEC52B4915862569B60073BFD3.0001 - DRS record last changed: September 3, 2026 - Page: https://openairdocs.org/ac/36-4B ## Versions - [AC 36-4D](https://openairdocs.org/ac/36-4) (Current), approved Oct 12, 2017, 277 pages - [AC 36-4C](https://openairdocs.org/ac/36-4C) (Historical), approved Jul 15, 2003, 211 pages - [AC 36-4B](https://openairdocs.org/ac/36-4B) (Historical), approved Mar 23, 1988, 66 pages ## Text Each page's text from the PDF; open the PDF at a page with #page=N. ### Page 1 U.S. Department of Transportation Federal Aviation Administration Advisory Circular **NOISE CERTIFICATION** HANDBOOK ,- March 23, 1988 AC 36-48 Initiated by: AEE-3 ### Page 2 U.S. Department of Transportation Federal Aviation Administration Subject: NOISE CERTIFICATION HANDBOOK Advisory Circular Date: 3/23/88 Initiated by: AEE-3 FOREWORD AC No: 36-4B Change: The aim of this technical manual is to promote uniformity of implementation of the noise certification requirements of Part 36 of the Federal Aviation Regulations (FAR) by presenting test, analysis, and documentation procedures for subsonic turbojet airplanes that have been determined by the FAA to be technically acceptable for demonstrating compliance with that regulation. Where appropriate, FAA policy governing such certifications is reviewed. Norman H. Plummer Director of Environment and Energy ### Page 3 CONTENTS CHAPTER 1. GENER.AI.. • • • • • • • • • • 111 • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • 1. Purpose • •••••.•••••••••••••••••••••••••••••••••••••••••••• 2. Cancellation ...•.•.••••••••••••••••••••••••••••••••••.•••• 3. Background •••••••••••••••••••••••••••••••••••••••••••••••• 4. Objective .•.•••..•.........••.•.••.••••••••••..••••••••••• 5. Noise Demonstrations •.•••••••••.•••••••••••••••••••••••••• a. Flight Tests ...••..•••••.•..••...•••.•••••.•••.••.••.. b. Supplemental Noise Demonstration Methods •••••••••••••• c. Changes to Noise Certification Levels for Derived Versions •••••••••••••••••••••••••••••••••• 6. Witnessing of Tests ••••••••••••••••••••••••••••••••••••••• 7. [Rese"ed] ... .................................•........... CHAPTER 2. NOISE DEMONSTRATION FLIGHT TEST UNDER PART 36 APPENDICES A .AND B ••••••••••••••••••••••••••••••••••••• 8. Test Conditions ••••••••••••••••••••••••••••••••••••••••••• a. Test Site Characteristics ••••••••••••••••••••••••••••• b. Atmospheric Conditions •••••••••••••••••••••••••••••••• 9. Flight Procedures •••.••••.•••••••••••••••••.•.•••••••••••• a. Selection of a Flight Test Noise Demonstration Method•••••••••••••••••••••••••••••••••• b. Nominal Conditions for Development of a Reference Noise-Power-Distance Data Base (NPD) •••• a•••••••••••••••••••••••••••••••••• c. Measurements at Non-reference Points •••••••••••••••••• d. Test Day Flight Procedure~ •••••••••••••••••••••••••••• e. Target Test Conditions •••••••••••••••••••••••••••••••• 10. Measurements •• •••••••••••••••••••••••••••••••••••••••••••• a. Sound Measurements .. a••••••••••••••••••••••••••••••••• b. Airplane Position and Performance ••••••••••••••••••••• 11. Sound Measurement System Specifications ••••••••••••••••••• 12. Analys~s System .•........ ••••••••••••••••••••••••••••••••• Page 1 1 1 1 1 2 2 2 2 2 3 (and 4) 5 5 5 5 6 6 7 7 10 11 11 12 13 14 14 iii ### Page 4 AC 36-4B 13. Sound Measurement Equipment Calibrations and System Checks. . • . • • • . . • . . . • • • • . . . . . . • • • • • . • . • . • • • • • • • . 15 14. a. b. Frequency Response••••••••a••••••••••••••••••••••••••• Amp 11 tude . .•..••••.....•...••.••• • • • • • • • • • • • • • • • • • • • • • Computation of Effective Perceived Noise Level (EPNL) •••••••••••••••••••••••••••••••.•.•••••••••••• a. b. c. d. e. Data Processing ............................ ••••••••••• Test Day EPNL •••••••••••••••••••• • • •. • • • • • • • • • • • • • • • • • Normalized ~PNL • ..•••••••••••••••....•••••••• • • • ••• • • • Duration Correction for Simplified Adjustment Method •• ••••••••••••••.••••••••••••••••••• • Reference EPNL Evaluation ••••••••••••••••••••••••••••• 15 15 15 16 17 17 17 18 CHAPTER 3. SUPPLEMENTAL NOISE DEMONSTRATION METHODS iv UNDER PART 36 APPENDICES A AND B••••••••••••••••••••••• 21 15. Alternative Methods ••••••••••••••••••••••••••••••••••••••• 21 16. Technical Considerations •••••••••••••••••••••••••••••••••• 22 a. Examples of Demonstration Applications •••••••••••••••• 22 b. Noise Source Characteristics •••••••••••••••••••••••••• 22 c. Noise Increment Derivation Using Common Demonstration Method •••••••••••••••••••••••••••••••••• 22 d. Supplemental Test Data Compatibility •••••••••••••••••• 24 e. Component Noise Sources ••••••••••••••••••••••••••••••• 24 f. Noise Increments Caused by Nacelle Acoustic Treatment Changes. . • • • . • • • • • • • • • • . • • • • • • • . • • • … 24 17. Static Engine Tests ••••••••••••••••••••••••••••••••••••••• 24 a. b. c. d. e. f. g. h. i. j. Limits on the Projection of Static to Flight Data•••••••••••••••••••••••••••••o•••••••••• 25 Test Site Requirements •••••••••••••••••••••••••••••••• 25 Test Hardware. • . • • • • • • • • • . • . . . . . . . . . . . . . . . . . . . . . . • . • • • 25 Inflow Control Devices •••••••••••••••••••••••••••••••• 25 ICD Calibration.. . . . . . • . . . . . . . . . . . • . . . . • • • • • … 26 Acoustic Shadowing .•..•••••.••••••••••••••••••••.••••• 26 Microphone Locations .....•...•..•..••••••••••••••••••• 28 Engine Power Test Conditions •••••••••••••••••••••••••• 28 Data Analysis System Compatibility •••••••••••••••••••• 28 Data Acquisition, Analysis and Normalization •••••••••• 28 3/23/88 ### Page 5 18. Projection of Static Engine Data to Airplane Flight Conditions •••••••••••••••••••••••••••••••••••••.••• 29 a. Engine Installation Effects ••••••••••••••••••••••••••• 29 b. Normalization to Reference Conditions ••••••••••••••••• 32 c. Engine Component Noise Source Separation •••••••••••••• 32 d. Noise Source Position Effects ••••••••••••••••••••••••• 33 e. Engine Flight Conditions •••••••••••••••••••••••••••••• 34 f. Noise Source Motion Effects ••••••••••••••••••••••••••• 34 g. Airframe Noise Component •••••••••••••••••••••••••••••• 35 h. Extrapolation to Airplane Flight Path ••••••••••••••••• 36 i. Total Noise Spectra •••••....•••••••••••••.•••••••••••• 36 j. EPNL Computations •••••.••••••••••••••••.•••••••••••••• 37 k. Noise Increment Computations •••••••••••••••••••••••••• 37 19. Supplemental Flight Tests ••••••••••••••••••••••••••••••••• 37 20. Supplemental Analyses ..•••••..••.•••.•................•••• 37 a. Noise Increments Derived from an Approved NPD ••••••••• 38 b. Analytical Modeling of Noise Components ••••••••••••••• 38 21. Supplemental Component Tests •••••••••••••••••••••••••••••• 39 (and 40) 22. Methods for Demonstrating No Acoustical Change •••••••••••• 39 (and 40) CHAPTER 4. DOCUMENTATION REQUIREMENTS ••••••••••••••••••••••••••••• 41 23. General . ..........................••...•.•.•......• … 41 24. Noise Compliance Demonstration Plan ••••••••••••••••••••••• 41 a. Introduction ..............•.•.• … 41 b. Test Item ...••...•..................•...•• … 41 c. Type of Noise Demonstration Method •••••••••••••••••••• 41 d. Test Description ............................•.• … 41 e. Data Acquisition Systems•••••••••••••••••••••••••••••• 41 f. Data Analyses and Normalization Procedures •••••••••••• 41 g. Equi valencies … 41 25. Noise Certification Compliance Report ••••••••••••••••••••• 42 a. Introduction ... ..........................••• , ••.. • ... • 42 b. Airplane, Engine, and Nacelle Descriptions •••••••••••• 42 c. Airplane Performance at Reference Conditions .•........... ...........•....•..... -.. • .. • • .. 42 d. Certification Noise Levels •••••••••••••••••••••••••••• 42 e. Noise Certification Test Results •••••••••••••••••••••• 42 V ### Page 6 AC 36-4B 26. DER Witness Informal Report ••••••••••••••••••••••••••••••• 43 27. Noise Control Act Findings •••••••••••••••••••••••••••••••• 43 28. Airplane Flight Manuals ••••••••••••••••••••••••••••••••••• 44 a. General Requirements for Limitation ••••••••••••••••••• 44 b. AFM Maximum Certificated Weight Limits •••••••••••••••• 45 c. Multiple Noise Certifications ••••••••••••••••••••••••• 45 d. Landing Flap Restriction •••••••••••••••••••••••••••••• 45 e. Optional Engine Thrust Ratings (Derate and Reduced Thrust) ••.••••••••••.••.•.•...••••.•.••••• 46 f. Noise Level Information ••••••••••••••••••••••••••••••• 46 29. !CAO Certifications ••••••••••••••••••••••••••••••••••••••• 46 APPENDIX 1. NOISE DATA CORRECTIONS FOR TESTS AT HIGH ALTITUDE TEST SITES •••••••••••••••••••••••••• 1 1. 2. Introduction ••••••••••••• 8 •••••••••••••••••••••••••••••••• Jet Noise Source Correction ••••••••••••••••••••••••••••••• APPENDIX 2. GUIDANCE MATERIAL ON METHODS TO ACCOUNT FOR THE EFFECTS OF BACKGROUND NOISE 1 1 ON AIRPLANE RECORDED NOISE DATA •••••••••••••••••••••• 1 1. 2. 3. Introduction ••.•..••••• e.e•••••••••••••••••••••••••••••••• Background Noise ••.•.••••••••••••••••••••••••••••••••••••• Correction Procedures •••••••••••• ~························ APPENDIX 3. GUIDANCE MATERIAL ON METHODS TO IDENTIFY PSEUDOTONES/FICTITIOUS TONES IN AIRPLANE RECORDED NOISE DATA•••••••••••••••••••v•••••••••••••• 1. Introduction ••••••••••••.••••••••••••••••••••••••••••••••• 2. Spectral Irregularities ••••••••••••••••••••••••••••••••••• 3. Methods for Identification of Pseudotones/Fictitious Tones ••••••••••••••••••••••••••• APPENDIX 4. COMPUTATION OF EFFECTIVE PERCEIVED NOISE 1 1 3 1 1 1 2 3/23/88 LEVEL (EPNL) ••••••••••••••••••••••••••••••••••••••••• 1 (and 2) APPENDIX 5. GUIDANCE MATERIAL ON METHODS TO CALCULATE CONFIDENCE INTERVALS ••••••••••••••••••••••••••••••••• 1 1. Introduction ••••••••.•••.•••••..••••....•••••.•.•••••••••• 1 2. Confidence Interval for Clustered Data •••••••••••••••••••• 1 3. Confidence Interval for Regression Curve •••••••••••••••••• 2 4. Confidence Interval for Pooled Data Sets •••••••••••••••••• 5 vi ~' ### Page 7 **CHAPTER 1. GENERAL** 1. PURPOSE. The aim of this advisory circular (AC) is to promote uniformity of implementation of the noise certification requirements of Part 36 of the Federal Aviation Regulations (FAR) by presenting technically acceptable test, analysis and documentation procedures for subsonic turbojet airplanes that may be used by applicants for demonstrating compliance with Part 36. Where appropriate, FAA policy governing such certifications is reviewed. 2. CANCELLATION. Advisory Circular 36-4A, Noise Certification Test and Analysis Procedures, dated January 21, 1986, is cancelled. 3. BACKGROUND. Part 36 contains noise certification standards for turbojets, large transport category airplanes, and propeller-driven small airplanes. These regulations require that standard procedures be used unless an equivalent procedure is approved by the FAA. However, airplane noise measurement and analysis technology has developed considerably since Part 36 was adopted in 1969. Through aircraft noise certification, a body of technically sophisticated measurement and analysis methods has been developed by certificate applicants. As a result of this experience, alternative compliance methods have evolved and have been approved by the FAA. This AC describes those procedures. Applicants should always obtain prior approval from the responsible FAA certificating official before initiating any noise certification action. It should also be noted that the methods presented in this AC are not the only ways to demonstrate compliance with Part 36 nor the only ones that may be approved in the future. 4. OBJECTIVE. The objective of noise certification is the demonstration of compliance with Part 36 for one or more specific airplane versions. The demonstration method used to determine noise levels will depend upon a number of factors. For example, Part 36 requires an applicant for a new type design to conduct a noise demonstration flight test. However, depending upon the extent of the data already in existence for the type design, the applicant for a change to an existing type design may submit for FAA approval a proposal either to conduct a flight test or to conduct supplementary tests and/or analyses. When supplementary data are used, they serve as the basis to adjust flight noise levels already approved for an earlier version. Dialogue between the applicant and the certificating authority at the beginning of an airplane certification program is necessary to identify anticipated changes in airplane type design that can impact noise certification. A noise compliance demonstration plan outlining the planned tests and/or analyses should be submitted as soon as possible to the FAA by the applicant. This plan should contain the method by which the applicant proposes to show compliance with the noise certification requirements. Because of the unique nature of each noise demonstration, the test plan is reviewed with the certificating authority before each test and, if necessary, modified to reflect any unique aspects or current developments related to the immediate noise demonstration test. With this flexibility, a single test plan, submitted to encompass more than the immediate test, is often advantageous. Guidance on preparation of noise compliance demonstration plans and certification compliance data reports is provided in Chapter 4 of this AC. Chap 1 Par 1 1 ### Page 8 **5. NOISE DEMONSTRATIONS.** a. Flight Tests. This AC presents guidelines on methods to acquire noise data from a sufficient number of flights of a particular airplane to derive the three required Effective Perceived Noise Level (EPNL) levels. Guidelines are also provided on acquisition and normalization methods such that a generalized noise data base (defined by noise-power-distance relationships) may be developed for an airplane type. Suggested procedures for determining the certificated Part 36 noise levels derived from this generalized noise data base are also provided. This generalized data base is directly applicable to changes in type design which retain the same basic noise source characteristics. b. Supplemental Noise Demonstration Methods. For those applicants seeking approval for changes in type design, guidance is also provided on suggested procedures for determining noise level changes relative to a previously approved version of the type. These procedures rely on supplementary testing and/or analysis techniques in conjunction with approved noise data already available for the type. The use of such methods for demonstrating noise compliance is optional, subject to FAA approval, and does not preclude an applicant from using a flight test for each noise certification demonstration. c. Changes To Noise Certification Levels For Derivative Versions. Many of the equivalent procedures given in this AC relate to derivative versions, where the procedure used yields the information needed to obtain the noise certification levels of the derivative version by adjustment of the noise levels of the "flight datum" aircraft (i.e., the most appropriate aircraft for which the noise levels were measured during an approved Part 36 flight test demonstration). The physical differences between the "flight datum" aircraft and the derivative version can take many forms, for example, an increased takeoff weight, an increased engine thrust, changed powerplant or propeller types, etc. Some of these changes will alter the distance between the aircraft and the noise certification reference points, others the noise source characteristics. Procedures used in the determination of the noise certification levels of the derived versions will, therefore, depend upon the change to the aircraft being considered. However, where several similar changes are being made, for example, introduction of engines from different manufacturers, the procedures used to obtain the noise certification levels of each derivative aircraft should be followed in identical fashion. 6. WITNESSING OF TESTS. All flight and other tests conducted in support of noise certification need to be witnessed by FAA personnel or by an Acoustic Designated Engineering Representative (DER) appointed under Order 8110.37. Under that Order, Acoustic DER's may: 2 Witness and approve noise certification tests conducted in accordance with an approved test program, when specifically authorized to do so by the FAA. Chap I Par 5 ### Page 9 Approve noise analysis techniques and computer programs and certify the noise values reduced by these programs that were measured and evaluated as prescribed in Subparts Band/or F of Part 36 or by an equivalent procedure previously approved for that noise test series by the Office of Environment and Energy. Prior FAA approval is required for the re-delegation by Acoustic DER's of the authority to witness tests. In addition, Acoust:ic DER's may not determine whether a type design cbange is an acoustic change under Part 21. Acoustical DER's also may not approve: Test plans or equivalent procedures, Operating limitations or other Aircraft Flight Manual information, or Certificated aircraft noise levels. 7. [RESERVED] Chap 1 Par 6 3 (and 4) ### Page 10 3/23/88 CHAPTER 2. NOISE DEMONSTRATION FLIGHT TEST UNDER PART 36 APPENDICES A AND B AC 36-4B 8. TEST CONDITIONS. The airplane should be operated such that noise values corresponding to the takeoff, sideline, and approach reference conditions defined in Part 36 can be derived within the limitations of Appendix 1. To meet the test environmental reQuirements prescribed in Part 36, the following test site and atmospheric conditions have been approved as providing an acceptable test environment. a. Test Site Characteristics. The noise measurement stations are placed such that the reference locations specified in Part 36 can be adequately simulated. Part 36 prescribes reference noise measurement stations relative to runway brake release or threshold point (e.g., 6500 meters from brake release for takeoff measurements); however, actual test measurement station locations may differ for the following reasons: (1) To minimize the impact of ambient sound on the measurement of the airplane sound pressure levels. Prior to testing, a comparison of expected airplane sound pressure levels with the expected test-site ambient sound will aid in choosing test microphone locations. It is generally preferable to relocate measurement stations rather than contaminate airplane noise measurements with ambient sound from sources such as nearby road or air traffic. (2) To increase the airplane signal-to-noise ratio to ensure adequate measurement of the airplane's acoustic signature. In such cases, the takeoff measurement station and/or the airplane flight trajectory may be relocated so that the test day sound propagation path length is less than the reference path length to ensure acquisition of acceptable data. In the selection of an adequate signal-to-noise ratio, consideration should be given to limitations on corrections for the influence of anbient sound, as well as limitations on the total correction from test to reference conditions. (3) When obstructions near the noise measurement station(s) will influence sound measurements. Takeoff and approach measurement stations may be relocated as necessary to avoid undesirable obstructions. Sideline measurement stations (distances) may be relocated by distances which are of the same order of magnitude as the airplane lateral deviations (or offsets) relative to the nominal flight path that occur during flight testing. NOTE: Flight path intercepts adopted as alternatives to full takeoffs and full stop landings may be adjusted to attain the same.performance as would be observed at the reference distances. b. Atmospheric Conditions. (1) Temperature and Relative Humidity Limits. When atmospheric conditions of ambient temperature and relative humidity result in atmospheric absorption in the 8 kHz one-third octave band at any point aJong the sound propagation path exceeding 12 dB/100 m, then instrumentation used to determine relative humidity snould consist of an ambient air temperature and dew point Chap 2 Par 8 5 ### Page 11 temperature measurement system accurate to within+ 0.5°C. When such instrumentation is used, the 8 kHz atmospheric absorption may be extended to 14 dB/100 m and, for approaches only, to 16 dB/100 m. Relative humidity is required to be between 10 and 95 percent. (2) Wind Limits. Testing may be conducted when: the wind speed does not exceed an average value, over the sound measurement period, of 12 kts and a maximum value of 15 kts; the crosswind component does not exceed an average value over the sound measurement period, of 7 kts and a maximum value of 10 kts. A 30 second moving average is used to define average wind speed. Part 36 requires the wind measurements to be made 10 meters above the ground and in the vicinity of the measuring station. **9. FLIGHT PROCEDURES.** a. Selection of a Flight Test Noise Demonstration Method. An accurate, reliable definition of the airplane noise characteristics is the primary objective of the demonstration test. To achieve this goal, at least six sound measurements are required by section A36.5(e) and section 8(a)of Part 36 to ensure that a mean noise level for each of the three certification reference conditions can be defined with a 90 percent confidence interval not exceeding + 1.5 EPNdB. This requirement can be achieved by conducting a series of test flights at or near reference power. After adjusting each measured noise level to Part 36 reference conditions, a certification noise value should be computed by averaging the adjusted levels (see paragraph 14). An alternate method is to acquire sufficient measurements (at least six) to allow the generalized noise characteristics of the airplane to be developed (a noise-power-distance data base (NPD)). (See paragraph 14(e)(2)). The three certification levels for any specific configuration are then determined by entering the developed data base at the certification reference distance and power and applying appropriate airplane speed corrections. (1) For a range of powers covering full and cutback powers, the airplane is flown past sideline and under-flight-path microphones in accordance with section C36.3 of Part 36. Sufficient noise measurements are made to enable noise-power curves at a given distance for both sideline and takeoff cases to be established. These curves are extended either by calculation or by the use of additional flight test data to cover a range of takeoff distances to form the generalized noise NPD data base for use in the noise certification of the "flight datum" and derived versions of the type. The 90 percent confidence intervals about the mean lines are constructed through the data. This sequence should be repeated for an under-flight-path microphone for a range of approach powers using the appropriate speed and airplane configuration. In compliance with section C36.9 of Part 36, the approach configuration that is most critical from a noise standpoint must be used. When in doubt, an applicant may be required to test and evaluate more than one configuration. (2) When approved data are available to define the engine spool down characteristics, it may be appropriate to substantiate noise levels for cutback power without the implementation of a cutback takeoff procedure by making noise measurements during reduced power operations. The average engine 6 Chap 2 Par 8 ### Page 12 spool down time should be representative of average conditions and should reflect a 1.0 second minimum altitude-recognition lag time to account for pilot response. For determination of the reference cutback noise levels, the cutback initiation should be selected to ensure stabilized cutback power conditions before the initial 10 Tone Corrected Perceived Noise Level (PNLT) down point. Flyover noise levels with power cutback may be established from the merging of PNLT versus time measurement~ obtained during constant power operations. As seen in Figure l(a), the 10 dB down PNLT noise time history may contain portions of both the full power and cutback power noise time histories. Provided these noise time histories, the average engine spool-down thrust characteristics and the airplane flight path during this period (Figure l(b)), which includes the transition from full to cutback power, are known, the flyover noise level may be completed. (3) An alternate method may be used to derive cutback power noise levels. In some cases, this method results in lower EPNL values, depending upon the characteristics of the forward radiated noise. Actual test runs may be conducted using a power cutback procedure that is initiated close to the overhead point. This technique is different from the procedure described in the previous paragraph in that the engine power is not stabilized during the entire 10 dB down time period. The concept involves delaying power cutback in order to gain extra altitude, but cutting back power before the maximum PNLT (PNLTM) exceeds the PNLTM that would exist using stabilized cutback power. Since this is a more difficult flight test procedure, care should be exercised in conducting the test runs to ensure consistency in order to meet the 90 percent confidence interval criteria. (4) The airplane's source noise characteristics should be measured over the operating range in sufficient detail to define the difference between test and reference day source noise. To aid in obtaining adequate data for these adjustments it may be necessary to identify those airplane/engine parameters, e.g., N1/ 0 2 for turbofans, that are the major controllers of the airplane's noise level and appropriately vary these parameters during the test. For propeller-driven airplanes, the engine parameters are propeller helical tip Mach number and shaft horsepower. b. Nominal Conditions for Development of a Reference Noise-Power-Distance Data Base (NPD). For the development of a generalized NPD, sound pressure level (SPL) data are normalized to the ambient environmental conditions specified in Part 36 and to nominal takeoff and approach aerodynamic configurations and performance. For example, the selected airplane speed should be representative of the range of speeds for the configurations and weights which are anticipated to be certified. In addition, takeoff and approach configurations may be different; requiring normalizing SPL data to different airplane speeds. Note that the adjustment procedure requires FAA approval when developing an NPD. (See paragraph 14(e) of this AC.) An example of a generalized NPD is shown in Figure 2. c. Measurements at Non-reference Points. In some instances test measurement points may differ from the reference measurement points in Part 36, Appendix C. Under these circumstances an applicant may request Chap 2 Par 9 7 ### Page 13 **8 FIGURE 1** COMPUTATION OF CUTBACK-TAKEOFF NOISE LEVEL FROM CONSTANT POWER TESTS I l(a) NOISE TIME HISTORY 1 · PNLT **POINT OF POWER** **CUTBACK** • • I B ALONE l(b) FLIGHT PATH OVER FLYOVER **MEASURING POINT WITH POWER** CUT-BACK 1 OdB **FULL TAKE-OFF** POWER FLICHT PATH ENGINE SPOOL·DOWN PERIOD ' \ **FLYOVER NOISE** **TIME HISTORY** WITH POWER \ \ / CUT-BAO< A ALONE TIME **CUT-BACK** POWER FLIGHT PATH i NOISE MEASURING POINT Chap 2 Par 9 ### Page 14 3/23/88 Chap 2 DIL-e .... _. IN.-• AC 36-4B FIGURE 2 **EXAMPLE OF GENERALIZED NOISE POWER DISTANCE DATA BASE** **TAKEOFF** • Ufl"lNCI AI … ID lllllfUC DISTANCE (FT) uoo MOO *° IIOO S,00 l2CIO l400 IIOO IIOO IDOO £Jf&IN£ fQIU (TtWJST OR lPH) APPROACH e 11lU9'lNCl AIIIIPIID • .. n. ALTrTUl>E **SIDELINE** e 410 111 MAXI … LIYEL • UFlUNCI· AIIIIPllD 9 -- ---- - - - ~-- --~~ ~~~ ### Page 15 approval of data that have been adjusted from the actual measurements to repre­ sent data that would have been measured at the reference points in reference conditions. (See paragraph 8(a).) Reasons for such a request may be: (1) to allow the use of a measurement location that is closer to the airplane flight path so as to improve data quality by obtaining a greater ratio of signal to background noise; (2) to enable the use of an existing, approved certification data base for an airplane type design in the certification of a derivative of that type when the derivative is to be certificated under reference conditions that differ from the original type certification reference conditions; (3) to avoid obstructions near the noise measurement station(s) which could influence sound measurements. By using a flight path intercept technique, takeoff and approach noise measurement stations may be located as necessary to avoid undesirable obstructions. Sideline measurement stations may be relocated by distances which are of the same order of magnitude as the airplane lateral deviations (or offsets) relative to the nominal flight paths that occur during flight testing. Approval has been granted to applicants for the use of data from non-reference noise measurement points provided that measured data are adjusted to reference conditions in accordance with the requirements and limitations of Part 36, Appendices A and B. d. Test Day Flight Procedures. Airplane weight and flight trajectory can be properly simulated by controlling those parameters most directly related to the airplane/engine noise source strength and directivity. Test day flight procedures should be chosen to provide an adequate definition of the airplane noise source characteristics. The selection of test day flight procedures will depend upon the methodology chosen to derive certification noise levels, i.e., either by repeating the reference conditions or from an NPD. (1) Test Day Procedures. Selection of test day flight procedures should be compatible with the methodology chosen to derive certification noise levels. If an applicant chooses to derive certification noise levels by averaging at least six conditions flown at or near reference power and distance, then corrections should account for differences in airplane source noise between test and reference airplane power settings, To qualify these corrections, testing at other operating conditions may be required to define in sufficient detail the airplane source noise levels. This testing may be in addition to the number of measurements at or near reference power necessary to achieve a 90 percent confidence interval not exceeding+ 1.5 EPNdB. The applicant may, however, choose to derive the certification noise level directly from a generalized NPD. The NPD is developed by conducting a sufficient number of sound measurements over a range of airplane operating conditions so that the 90 percent confidence interval about a mean line does not exceed+ 1.5 EPNdB. To describe a power line an applicant should conduct a minimum of six flights, each at a different power setting, at a common target altitude. To attain the required confidence interval, more than six flights may be necessary. Typically, the power-related parameter which governs noise generation is corrected low pressure rotor speed, N1/~, corrected net thrust, Fn/cfamb' or EPR, where: 10 Chap 2 Par 9 ### Page 16 N1 is the actual low pressure rotor speed, 0 is the ratio of absolute total temperature of the air at the height of EEe airplane to the absolute temperature of the air for an international standard atmosphere (ISA) at mean sea level (i.e., 288.15°K) EPR is engine pressure ratio. F is the actual engine net thrust per engine, and n ,:S is the ratio of absolute static pressure of the ambient air at th:mgeight of the airplane to ISA air pressure at mean sea level (i.e., 29.92 in HgA or 101.325 kPa). (2) Simulated Takeoffs and Approaches. Where shown by the applicant to be equivalent to the procedures specified in Part 36, simulated takeoffs and approaches, consisting of flight path intercepts may be used in lieu of starting and ending each flight from a full stop. For takeoff, flight path intercepts consist of intercepting and following the desired climb profile beginning at a target distance above ground level. Approach intercepts maintain the approach path over the measurement location without actually landing. In either case, the airplane is stabilized and following the target trajectory throughout the sound measurement period. (The sound measurement period includes levels within 10 PNdB below the maximum.) Test airplane weights, speeds, and aerodynamic configurations should be selected to allow achieving target test conditions for those performance parameters having the major effect on noise generation. Approved inflight operable Auxiliary Power Units (APU's) must be on during all approach test conditions. e. Target Test Conditions. Target test conditions can be established for each sound measurement. These target test conditions will define the selected aerodynamic configuration, airplane weight, flight procedure (such as takeoff flight path intercept), altitude, power, and airspeed at the closest point of approach to the measurement location, For example, a takeoff aerodynamic configuration may be defined by the climb gradient and flap setting. NOTE: The airplane angle of attack will remain approximately constant for all weights if the tests are conducted at a constant flap setting and the airplane speeds correspond to the reference speed, v2 + 10 kts for each weight. The aerodynamic configuration, e.g., flap setting, must remain constant during the sound measurement period. Speed corrections, however, are generally permitted up to 15 kts. Alternatively, for many airplanes the airplane attitude remains approximately constant for all weights if the tests are conducted at the V + 10 kts for the maximum takeoff weight. In the execution of the flight for purposes of obtaining sound measurements, the pilot should "set up" the airplane with the selected configuration in order to pass by the measurement location at the target altitude, power, and airspeed. 10. MEASUREMENTS. This section provides guidelines on how to obtain sound measurements and relevant airplane position and performance data necessary for normalizing the measured sound data to conditions differing from the actual test day conditions. All of these are synchronized. Chap 2 Par 9 11 ### Page 17 a. Sound Measurements. (1) All sound measurements are made with equipment meeting the specification in Part 36. The measurements are made such that, when processed, they will provide one-third octave band sound pressure levels as a function of time for the calculation of EPNL. (2) The location of each noise measuring station must be surrounded by relatively flat terrain having no excessive sound absorption charac­ teristics, such as tall (in excess of 3 inches) grass, shrubs, or wooded areas. For measurement sites located on the extended runway centerline, a closely mowed grass circle SO feet in diameter is acceptable. For sideline locations, the grass may be mowed in a SO-foot semicircle facing the line of flight. Any earthen or sandy surface should be reasonably tamped down. Plowed furrows, silt, or soft powdered surfaces are unacceptable. In the winter time, a SO-foot radius microphone area that is clear of snow may be considered acceptable by the onsite FAA witness when other weather restrictions are acceptable. (3) Microphones should be located at the certification reference height of 1. 2 m (4 ft) above the ground surface. In some cases, such as for pseudotone identification, it may be desirable to obtain additional data exhibiting free-field spectral shapes by selecting a microphone location to minimize the interference effects of ground reflections on the measured sound spectra (see Appendix 3 of this AC). Free field spectral shapes can be obtained by: (a) mounting the microphone high above the ground first interference frequency falls below the range of interest. 10 m (33 ft) is generally adequate. so that the A height of (b) mounting the microphone diaphragm in the plane of, or very close to, a large, acoustically hard (reflective) ground-plane surface so that the direct and reflected waves arrive in phase causing a doubling of the sound pressure. Care must be taken to ensure that the hard surface does not have air pockets beneath the surface or edges that could influence the acoustic measurements. (4) Orientation of the microphone will depend on the microphone type. Most commonly used are grazing incidence microphones which are designed so that nearly a flat frequency response is maintained at grazing incidence to the microphone; the microphone is oriented to maintain near grazing incidence throughout the airplane flyover. Perpendicular incidence microphones are designed so that maximum sensitivity occurs when they are oriented so that the sound impinges upon the diaphragm in a perpendicular direction when the airplane is at the closest point of approach. Data obtained using perpendicular incidence microphone are acceptable provided that the data are corrected for directivity and frequency response effects and the maximum Noy value for the corrected data is located in the one-third octave band centered at 5000 Hz or less. 12 Chap 2 Par 10 ### Page 18 (5) If the airplane sound pressure levels within the sound measurement period do not exceed the background sound pressure levels by at least 10 dB in each one-third octave band, they may be adjusted to eliminate the additive influence of the background sound. When the airplane sound is less than 10 dB, but more than 3 dB above the background, the corrected airplane sound levels are obtained by logarithmically subtracting the background sound levels. For those situations where the airplane levels are less than 5 dB above the background, this influence may be avoided by obtaining measurements at distances closer to the airplane and then adjusting the measurement to the desired distance or by adjusting the measured sound pressure levels using technically supported analytical shaping methods (see Appendix 2 of this AC). NOTE: If the measurements are not adjusted to delete background noise, the resultant EPNL values will be higher than if measured in the absence of background noise. (6) Maximum sideline noise may be determined by positioning several measurement stations parallel to the airplane flight track as prescribed in Part 36. An alternative to this method may be to use two measurement stations, symmetrically positioned about the nominal flight track, and fly the airplane at several altitudes and thus simulate the data that would have been acquired during a "climb-out" using a series of measurement stations. Experience suggests that the maximum noise level will occur at an elevation angle near 34 degrees (approximately 1000 ft altitude for airplanes with 1,476 feet [450 meter] sideline measuring stations) and the levels will be rather insensitive within several degrees of this value. When the two-station method is used, the sideline noise for each run is the arithmetic average of the corrected EPNL from each of the two lateral measurement points. The confidence level should be calculated using the method in Appendix 5 (paragraph 2 of this AC). b. Airplane Position and Performance. (1) The airplane height and lateral position relative to a fixed location (the nominal flight track along the extended centerline of the runway) should be determined by a method independent of normal flight deck instrumentation. Methods found to be acceptable include radar tracking, theodolite triangulation, laser trajectography, microwave positioning, and photographic scaling techniques (when the simplified method is used). Care must be taken when the photographic scaling technique is used to ensure that the airplane position can be reliably related to the sound measurement positions. In the past, inaccuracies in space posicioning have been introduced by using hand-held cameras to take single photographs of aircraft supposedly overhead. The aircraft are often off-center from the reference flight path. Firm mounting on a fixed tripod can eliminate some of the sources of possible error The simplified method for correcting EPNL should be used when continuous tracking is not available. (2) The airplane position along the flight path can be related to the sound pressure levels recorded at the measurement locations by using synchronized signals. Airplane position should be recorded during the sound measurement period. Chap 2 Par 10 13 ### Page 19 (3) Measurements of airplane/engine performance parameters for calculation of test day performance should be recorded throughout the sound measurement period. These measurements should be time-related to the airplane position along the flight path. Examples of parameters for measurement are airplane attitude, engine power setting, climb angle, and airplane speed. Flap position and gear posHion should be fixed and noted for each test condition. 11. SOUND MEASUREMENT SYSTEM SPECIFICATIONS. The noise certification test should use a quality sound measurement system. Output should be measured in the form of one-third octave band sound pressure levels covering the frequency range from 45 Hz to 11,200 Hz for each one-half second over the sound measurement period. Included in the measurement system should be components that perform the following functions: transform sound pressure waves to electrical signals (microphones), condition the signals, store signals for later analysis, and retrieve and analyze the signals in terms of one-third octave band levels as a function of time. a. Corrections to compensate for instrumentation non-uniform frequency response characteristics should be determined and applied to the sound pressure level data. These should include response characteristics of (a) microphone free field and directivity, (b) windscreen insertion loss and directivity, (c) preemphasis, (d) signal conditioning equipment, (e) data storage and retrieval system, and (f) analysis system. All corrections should be identified, documented, and FAA approved. b. When selecting sound measurement system components, compliance with requirements may be determined from manufacturers' specifications, user calibrations, or a combination of both. When multiple systems of the same type are used, demonstration of compliance of at least one system is satisfactory. However, this does not preclude the need to perform the system calibrations and checks described in Part 36 and paragraph 13 of this AC. Analysis systems that meet the specifications defined in the following paragraphs have been approved. 12. ANALYSIS SYSTEM. The analysis system may be either analog or digital and provide a root mean square (RMS) sound pressure level in one-half second increments in each of the 24 one-third octave bands having geometric mean frequencies from 50 Hz to 10 kHz. The averaging properties of the integrator may correspond to one of the following: a. For instruments with exponential averaging characteristics, the requirements for Type 1 sound level meter with "slow" response as defined in International Electrotechnical Commission (IEC) Publication No. 651. b. For true integrating instruments compliance with the "slow" response as defined in !EC 651 may be demonstrated by combining successive readouts from the data processor, such as -3, 0, +1 half-second samples or -2, 0, +2 half-second samples. This combination of readouts is required since the effective "slow" response averaging time is greater than the sampling interval. When a true integration system is used, an acceptable alternative 14 Chap 2 Par 10 ### Page 20 to combining "successive" readouts is to set the integration time equal to the sampling interval. The use of this type of data reduction system has required the use of the integrated EPNL method. 13. SOUND MEASUREMENT EQUIPMENT CALIBRATIONS AND SYSTEM CHECKS. Part 36 requires calibration of the sound measurement system. This section describes approved procedures to calibrate and check the complete measurement system. These procedures provide guidance which ensures proper equipment performance over the frequency range from 45 Hz to 11.2 kHz. a. Frequency Response. (1) The free-field frequency response characteristics of each micro­ phone are determined by a pressure response calibration (which may be obtained from an electrostatic calibrator) in combination with manufacturer provided corrections or by a free-field calibration in an anechoic facility. The directivity response characteristics of the microphone need to be defined and cover a sufficient angular range to encompass the sound measurement period. (2) Within 5 days of test the electrical response characteristics of each measurement system, excluding the microphone and windscreen, should be determined at a level within 10 dB of the full-scale reading used during the test, using one of the following input signals: sine waves at the center frequency of each one-third octave band, or pink noise. Sufficient determinations are made to ensure that the overall calibration of the system is known for each test. The calibration signal generator should have been checked within 6 months of the test series with instruments which meet standards set by the National Bureau of Standards. b. Amplitude (1) An acoustic calibration of the system, including the microphone, should be performed in the field at least once every 5 days. The acoustic calibration should contain a reference signal of known amplitude and frequency in order to provide correlation with the calibrated sound pressure level. A pistonphone operating at a nominal 124 dB and 250 Hz is generally used. In addition, immediately before and after each day's testing, a recorded calibration of the system can be made in the field by injecting a 250 Hz sine wave signal into each microphone preamplifier. (2) Each magnetic tape reel should contain a reference signal of known amplitude and frequency in order to provide correlation with the calibrated sound pressure level. The reference signal (which represents a known sound pressure level) is generated when the acoustic calibrator is in the proper position over the microphone or an electrical signal of known amplitude is inserted into the tape recorder. 14. COMPUTATION OF EFFECTIVE PERCEIVED NOISE LEVEL (EPNL). SPL data used in compliance demonstrations should be corrected to the airplane and environ­ mental reference conditions prescribed in Part 36. Methods to determine a corrected EPNL at each measurement location for evaluating compliance with Part 36 noise level requirements are provided in this section. Chap 2 Par 12 15 ### Page 21 a. Data Processing. Before computing an EPNL, the measured SPL, meteorological, and airplane position and performance data acquired as described in paragraph 10 should be reviewed to ensure there are no anomalous data. However, it should be noted that Part 36 requires that no data may be omitted without specific FAA approval. Measured data should be processed in accordance with the guidelines provided below. (1) Sound Pressure Level Data (SPL). Sound pressure levels for each of the one-third octave bands covering the frequency range from 45 Hz to 11200 Hz should be provided for each one-half second over the sound measurement period by analyzing measured signals with equipment that is consistent with the performance specifications defined in paragraph 11. Corrections for instrument responses, as described in paragraph 13 (such as frequency response), should be taken into account during the data processing. Measured SPL data may be adjusted to account for the presence of the following undesirable data contamination. (NOTE: If the measurements are not adjusted to delete the presence of contamination the resultant levels will be higher than if measured in the absence of such contamination.) (a) Extraneous Sound Sources. Contamination of the measured airplane-generated signal may occur occasionally due to the presence of extraneous sound sources. Examples of such sound sources include birds chirping in the vicinity of the microphone or tape anomalies (tape anomalies can be caused by an infrequent, temporary loss of the FM carrier signal). The measured SPLs may be adjusted to reflect the SPL which would have been measured in the absence of the extraneous source, Where appropriate, spurious one-third octave band spectrum levels may be replaced by linear interpolation of adjacent, uncontaminated spectral levels. (b) Background Noise Corrections. The resultant airplane sound pressure level values, after adjustment, should correspond to the sound pressure levels which would have been measured under test conditions in the absence of any background noise. As discussed in paragraph lO(a)(S) of this AC, sound pressure levels recorded during the airplane flyover and which do not exceed the mean background noise sound pressure levels by at least 3 dB, or the analyzer floor by 1 dB, may be adjusted using a technically supported analytical shaping method. Acceptable methods are provided in Appendix 2. (c) Pseudotone/Fictitious Tone Identification. Spectral irregularities resulting from ground plane reflections, known as pseudotones, or any spectral irregularities other than those resulting from airplane sound, known as fictitious tones, may be excluded from the tone correction evaluation in the EPNL calculation. Acceptable methods for identifying pseudotones/ fictitious tones are provided in Appendix 3 of this AC. (2) Airplane Position and Performance Data. Representative airplane position and performance data for the sound measurement period should be computed for each measurement location. (3) Meteorological Data. The ambient air temperature and relative humidity values which are used for the computation of the atmospheric absorption of sound, are those which are representative of the atmospheric 16 Chap 2 Par 14 ### Page 22 conditions over the sound propagation path at the time of each sound measurement. An absorption value determined from the meteorological data acquired at 10 m (33 feet) provides an adequate representation of the atmospheric conditions, if the absorption coefficient in the 3150 Hz one-third octave band does not vary more than+ 0.5 dB/100 mover the sound propagation path corresponding to the maximum PNLT. When the atmospheric absorption coefficient in the 3150 Hz one-third octave band, over the sound propagation path relative to the PNLTM varies more than+ 0.5 dB/100 m, then "layered" sections of the atmosphere should be used to-compute equivalent weighted sound attenuations in each one-third octave band. The arithmetic average of values of temperature and humidity at the top and bottom of each layer are used to represent the sound path atmospheric absorption rate for each layer. (Also see paragraph 8(b)(l) of this AC.) b. Test Day EPNL. Using the SPL data, corrected for instrumentation and background noise and processed in a manner described in paragraph 14(a), an EPNL corresponding to test day conditions should be calculated per the method provided in Part 36, Appendix B. To reduce the influence of random spectral irregularities, one or more of the correction methods provided in Appendix 3 of this AC may be used to meet the requirements of Part 36, Appendix B. c. Normalized EPNL. Measured SPL data are normalized to the reference ambient conditions specified in Part 36 and to nominal airplane position and engine performance conditions. Nominal airplane position and engine performance data should be derived by using approved aerodynamic performance data at the reference ambient conditions. The measured SPL data should be corrected for the effects caused by differences between test day and nominal conditions. The corrections account for the effects of atmospheric absorption, sound propagation distance, airplane speed, and airplane source noise. A normalized EPNL should be calculated using either the "simplified" or "integrated" adjustment methods as described in Part 36, Appendix A (section A36.ll(f)) for takeoff. The integrated method has been extended to sideline and approach noise measurements for some applications. However, the "simplified" method should generally be used for adjusting sideline noise measurements. Some NPD procedures have required the use of the integrated adjustment method. When an array of sideline microphones located on one side of the airplane flight path is used, the maximum sideline noise level and the associated airplane altitude should be determined from a least squares polynomial regression of sideline noise as a function of airplane altitude. The regression curve, and thus the maximum sideline noise, is adjusted by an amount equal to one-half the difference between noise levels observed at the symmetric sideline locations. When the airplane is flown past two symmetrical microphones at several altitudes, the arithmetic average of the two microphone levels is used as the datum for each flight. d. Duration Correction for Simplified Adjustment Method. When calculating a normalized EPNL using the "simplified" method of adjustment, a duration correction may be applied as follows: A 2 = - 7. 5 log (DM/DR) Chap 2 Par 14 17 ----- -- -~---~~~- ### Page 23 where: DM is the minimum distance to the measured flight path for takeoff and approach. For sideline, DM is the measured sound propagation distance associated with maximum PNLT. DR is the minimum distance to the reference flight path for takeoff and approach. For sideline, DR is the reference sound propagation distance for maximum PNLT. e. Reference EPNL Evaluation. Reference EPNL values for each of the three certification reference measurement locations and their 90 percent confidence intervals can be produced from the test results and reported. Guidance on confidence interval computations is provided in Appendix 5 of this AC. No test result should be omitted in the evaluation of the reference EPNL unless specified by the certificating authorities. Applicants should meet the 90 percent confidence interval not to exceed+ 1.5 EPNdB as required by section A36.5(e)(2) of Part 36. (1) Average of Clustered Flight Conditions. For each measurement location, one method for demonstrating compliance with the 90 percent confidence interval requirements is to conduct a series of sound measurements at prescribed flight conditions, A minimum of six flight conditions must be conducted at or near reference power or at judiciously spaced intervals surrounding the reference power. Measured SPL data for each flight condition should be normalized to reference ambient conditions and to reference airplane position and performance accounting for the effects of atmospheric absorption and sound propagation distance. These normalized values of SPL should be used to compute EPNL. Values of EPNL should be further corrected to account for differences between test and reference airplane speed and source noise. Source noise level corrections may be derived from approved supplemental data acquired in addition to the test flights clustered around the reference power. The resulting corrected EPNL's for all flight conditions at each single measurement location are arithmetically averaged to derive a single reference EPNL value. (See paragraph 14(e)(2)(b) below when the data are to be used to derive a mean EPNL line for an NPD or for determining sideline noise levels.) The corrections may be derived solely from the test flights provided the test flight conditions were conducted at intervals surrounding the reference power such that the source noise level versus power setting is defined. In this case, the resulting EPNL's at a measurement location are obtained using the NPD methods in the following paragraph. (2) Noise-Power-Distance Data Base. (a) An alternate method of demonstrating compliance with the 90 percent confidence interval requirements is achieved by developing an NPD data base. The NPD data base should be developed by conducting a series of flights, each at a different airplane operating condition, while maintaining the same aerodynamic configuration. The measured sound pressure level data for each power condition should be normalized to the reference ambient conditions and specified airplane position and performance. The normalized sound pressure level data are used to compute values of EPNL. For each measurement location, the nominal airplane position and performance will be obtained for each selected aerodynamic configuration. The EPNL at each power 18 Chap 2 Par 14 ### Page 24 condition is then normalized to an array of minimum distances. All correction techniques in paragraph 14(c) should be applied except the source noise correction which is not applicable in the development of the NPD data base. (b) For each array of corrected EPNl, and power values, corresponding to a common airplane to ground minimum distance, a mean line and the 90 percent confidence intervals about the line should be computed and documented. This mean line as shown in Figure 5-1 in Appendix S, is obtained from a first, second, or third order least-squares polynomial regression analysis. (Note: the choice of the order of curve fit requires FAA approval.) Similarly constructed lines may be used for EPNL versus slant range or altitude. (c) The reference EPNL value should be derived by entering the NPD data base at the reference location and the aerodynamic performance conditions of power and minimum distance and by applying a duration correction to account for the difference between the nom:i.nal airplane speed used in development of the NPD and the reference airplane speed. To minimize errors introduced in the speed correction pro~ess, hold the variations in test speed around the reference values to less than 15 knots. (3) Procedures for the Determination of Changes in Noise Levels. Noise level changes determined by comparison of flight test data for different configurations of an airplane type have been used to establish certification noise levels of newly derived versions by comparison to the noise levels of the flight datum airplane. These noise changes should be added to or subtracted from the noise levels obtained from individual flights of the "flight datum" airplane. Chap 2 Par 14 19 (and 20) ### Page 25 3/23/88 CHAPTER 3. SUPPLEMENTAL NOISE DEMONSTRATION METHODS UNDER PART 36 APPENDICES A AND B AC 36-4B 15. ALTERNATIVE METHODS. This chapter provides guidance on alternative methods for deriving noise levels in support of Part 36 compliance demonstration for conventional turbofan-powered airplanes that result from a change in type design. The methods addressed are ground static engine testing, supplemental flight testing, analytical adjustment and component testing. Although emphasis is placed on deriving noise increments suitable for adjustment of a Noise-Power-Distance (NPD) data base derived according to the methods outlined in Chapter 2, the guidance provided is also applicable to adjustment of singular noise levels at any of the three Part 36 reference locations. Approval of equivalent procedures for the use of static test information depends critically upon the availability of an adequate approved data base acquired from the flight testing of the flight datum airplane. For each application, the FAA will review the technical validity of the existing certification noise levels/NPD to ensure the data base is sufficient for derivation of certification noise levels for airplanes resulting from a change in type design. a. The requirements for approval of acoustical changes are contained in FAR section 21.93(b), not in Part 36. As a result, once a flight test data base is approved for the baseline airplane, it is seldom necessary to employ full Part 36 noise tests to demonstrate compliance for an acoustical change. In many instances, approval has been based on compliance demonstration methods not requiring flight tests. In these instances noise level increments between two airplane versions were derived using supplemental test and/or analysis techniques. The results of these analyses were used to demonstrate either no acoustical change, or to establish certification noise levels for the changed airplane version. Certification noise levels for a derived version are determined by applying the appropriate noise increments to noise levels already approved for the airplane prior to the design change provided that the data are within the required confidence interval. This adjustment of previously approved noise levels may be made either at the specific Part 36 reference points or, more generally, to an approved NPD for the type design from which Part 36 noise levels may be derived. b. Noise increments for certain changes in type design can be established by a specific supplementary demonstration method; for example, noise increments due to airplane performance changes may be assessed using an approved NPD analysis. The available supplementary demonstration methods should be reviewed for the most appropriate method based on the guidelines in this AC. The use of these methods for demonstrating noise compliance is discretionary and does not preclude an applicant from conducting a noise demonstration flight test to show compliance. c, When it has been determined in advance that more than one aircraft configuration may be certificated from common flight test data bases, it is often useful to the applicant to develop and seek FAA approval of an overall certification plan for that family of aircraft. This is referred to as a "family plan." These plans characteristically involve one or more approved NPD data bases and supplementary demonstrations, such as engine static tests. Chap 3 Par 15 21 ### Page 26 16. TECHNICAL CONSIDERATIONS, Selection of a supplementary noise demonstration method for a specific change of type design depends upon a number of general technical considerations. The technical items common to each demonstration method are discussed in this section; technical items associated with a particular type of supplemental demonstration method are provided in the respective sections. a. Examples of Demonstration Applications. The nature of the change in type design may limit the selection of appropriate supplementary noise demonstration method(s). Examples of the application of supplementary methods are provided in Figure 3. The list is not comprehensive, and, for a specific type change, more than one method may be technically valid. b. Noise Source Characteristics. (1) For type changes which retain the same basic noise source characteristics as the approved version that has demonstrated, by flight test, compliance with FAR Part 36, certification noise levels may be derived directly from the approved NPD; for example, noise increments resulting from airplane aerodynamic performance differences can be obtained directly from the NPD by using analytical methods described in paragraph 18 of this AC. (2) The nature of some type changes may or may not constitute a change in the basic airplane noise source characteristics. In such cases, the approved NPD may be acceptable for deriving noise increments if supporting data or analyses are provided which demonstrate that the noise source characteristics remain unchanged. For a power rating extension, a certification noise level may be directly derived from an NPD, which has been extended using approved analytical techniques. (3) For type changes which may result in changes to the basic noise source characteristics, modifications to the approved NPD may be incorporated through use of one of the supplementary methods described in this AC. Noise level increments representative of the differences between the reference and changed version should be algebraically added to the approved NPD values. A new engine type or nacelle are examples of changes in type design which may change the basic noise source characteristics. (4) For some type changes which may result in altering the basic noise source characteristics, adjustments may be made directly to the approved flight levels. These adjustments reflect the noise difference between two nacelle and/or engine configurations and may be applied to the approved flight levels as one-third octave band adjustments or as EPNL adjustments. Application of this method follows the same procedures as defined for adjusting an NPD data base, except it is applicable only to a discrete airplane flight condition. c. Noise Increment Derivation Using Common Demonstration Method, To ensure derivation of the correct noise increment between two different airplane configurations, the supplemental test and analysis methods used for both the previously approved and changed version must be of the same type. 22 Chap 3 Par 16 ### Page 27 **FIGURE 3** **POSSIBLE APPLICATIONS OF SUPPLEMENTAL** **NOISE DEMONSTRATION METHODS** DEMONSTRATION BASIS ANAL.VSlS C0'1PONENT TESTING 't:~: LINl!li STATIC **TESTING** SUPPLEMENTAL ~ FLIGHT ~:t:iiiiu;::::~b TEST 1 NG ' 777777777?, Chap 3 Par 16 NOTE: **EXAMPLE OF CHANGES IN TYPE DESIGN** **AIRPLANE AERO-PERFORMANCE CHANGES** (e.g., Gross Weight, Flap, Fuselage, etc.) **MINOR ENGINE THRUST AND/OR RATING** **CHANGES** **MINOR MODIFICATIONS TO ENGINE** **MINOR MODIFICATIONS TO ENGINE** NACELLE [ ENGINE COMPONENTS (e.g., Jet, etc.) ACOUSTIC TREATMENT CHANGES Fan, **NEW ENGINE OF COMPARABLE TECHNOLOGY** **AND SOURCE NOISE CHARACTERISTICS** **ACOUSTIC TREATMENT CHANGES** **NEW ENGINE NACELLE** ENGINE COMPONENT CHANGES NOISE SUPPRESSION DEVICE (e.g., Exhaust Gas Mixer) ENGINE CGMPONENTS (e.g., Jet, Fan, etc.) **THIS LIST IS NOT COMPREHENSIVE AND** **FOR A SPECIFIC TYPE CHANGE, MORE** **THAN ONE DEMONSTRATION METHOD MAY BE TECHNICALLY VALID.** ### Page 28 This avoids discrepancies that can occur from merging differing methods, The demonstration method is designed to provide noise levels representative of those occurring with production hardware (e.g., engine and nacelle). Furthermore, the spectral characteristics of the basic and modified configurations need to be similar. d. Supplemental Test Data Compatibility. The application of a noise demonstration method depends upon the compatibility with data previously acquired, Items which should be considered include test hardware, test and analysis procedures, test site and configuration. e. Component Noise Sources. The airplane noise source components affected by the change should be considered in determination of the best method for the evaluation of changes to the total airplane source noise characteristics. Parameters controlling the component noise source generation mechanisms and directional characteristics should be identified in order to assess the noise sensitivity of each component source. For example, when the total airplane noise characteristics are dominated by one source, such as the low speed rotor, the use of component test results, in combination with a component noise source analysis, may be an appropriate method to demonstrate noise increments. f. Noise Increments Caused by Nacelle Acoustic Treatment Changes. Noise increments caused by nacelle acoustic treatment changes made to noise certified aircraft can be empirically derived by coupling data from static engine noise tests with data from certification flyover noise tests. The noise increments, in EPNdB, can be determined by applying the difference in measured spectra between the two nacelle treatments to measured one-third octave band sound pressure levels in flight at the same directivity angle. If the modified nacelle treatment produces levels lower than the existing nacelle treatment, noise generated by the airframe should be considered in determining the dominant noise source. 17. STATIC ENGINE TESTS. Data acquired from static tests of engines incorporating similar designs to those that were flight tested may be projected, where appropriate, to flight conditions and, after FAA approval, used to supplement an approved NPD for the purpose of demonstrating compliance with Part 36 in support of a change in type design. This section provides guidelines on static engine test data acquisition, analysis and normalization techniques, The information provided is used in conjunction with the technical considerations in paragraph 16 and the general guidelines for test site, measurement and analysis instrumentation and test procedures provided in Society of Automotive Engineers (SAE) Aerospace Recommended Practice (ARP) No. 1846, "Measurement of Noise from Gas Turbine Engines During Static Opera­ tion." The engine designs and the test and analysis techniques to be used should be presented in the test plan and submitted to the FAA for concurrence prior to conducting the test. It should be noted that test restrictions defined for flight testing in conformity with Part 36 are not necessarily appropriate for static testing. For example, the distances associated with atmospheric absorption for static tests are substantially less than those encountered in flight testing and thus may require different criteria. 24 Chap 3 Par 16 ### Page 29 a. Limits on the Projection of Static to Flight Data. The amount by which the measured noise levels of a derivative engine will differ from the reference engine is a function of several factors, including: (1) thermo­ dynamic changes to the engine cycle, including increases in thrust; (2) design changes to major components, e.g., the fan, compressor, turbine, exhaust system, etc.; and (3) changes to the nacelle. Additionally, day-to-day and test site-to-site variables can influence measured noise levels and, therefore, the test, measurement, and analysis procedures described in this AC are designed to account for these effects. While the FAA does not impose a fixed numerical limit on the amount of extrapolation from reference to derivative engines, it is important to consider whether there are practical limitations to proposed extrapolation techniques and at what point new flight test data may be required. b. Test Site Requirements. The test site should meet at least the criteria specified in SAE ARP 1846. Different test sites may be selected for testing the subject configurations provided the test sites approved by the FAA are acoustically similar. See paragraph 17(g) of this AC for criteria concerning data acquisition and analysis systems. Depending upon test objectives and unique test stand and test site characteristics, wind speed and direction limits should be considered. c. Test Hardware. A flight inlet adjusted for static testing by the addition of a bellmouth forward of the inlet may be used with each turbofan or turbojet static noise test. Production inlet acoustic lining and spinners are also to be installed during noise testing. d. Inflow Control Devices. The use of static engine test noise data for Part 36 noise certification under an approved "family plan 11 requires the use of an approved Inflow Control Device (ICD) for high bypass engines (BPR 2.0). For static tests conducted after May 1, 1986, the following requirements have been imposed on the ICD: (1) The specific ICD hardware must be inspected for conformance with an approved ICD design by an approved method. (2) The ICD must be acoustically calibrated for sound transmission loss by an approved method. (3) The ICD acoustic calibration must be used as a correction in the certification data analysis. In applying the calibration, an approved method must be used to determine that portion of the noise received at the forward quadrant microphones that is radiated through the ICD. The calibration is applied to that portion of the noise that is radiated through the ICD. (4) The ICD position relative to the engine inlet must be determined and found to be one of the calibrated positions. (5) The ICD cali.bration consists of an SPL correction for each one-third octave band at each forward quadrant microphone that will be used during static engine testing. Chap 3 Par 17 25 ### Page 30 (6) No more than one calibration is required for an ICD hardware design, provided that there is no deviation from the design for any one ICD serial number hardware set. The above calibration requirements are not required if the same ICD hardware (identical serial number) is used as was previously used in the static noise test of the flight engine configuration, and the fan tones for both engines remain in the same one-third octave bands. e. ICD Calibration. An acceptable ICD calibration method is as follows: (1) Assume the correction is zero for the one-third octave bands from 30 Hz through 800 Hz, unless the engine design would dictate that forward radiated tones or major inlet broadband noise occurs below 1000 Hz. In that case, an approved correction procedure is required. (2) Place an acoustic driver(s) on a simulated engine centerline in the plane of the engine inlet lip. Locate the calibration outer microphones on the forward quadrant azimuth at a radius between 50 and 150 feet that provides a good signal-to-ambient noise ratio and at each microphone angle to be used to analyze static engine noise data. Locate a near-field microphone on the centerline of and within 2 feet of the acoustic center of the acoustic driver(s). (3) Conduct a test without the ICD in place. The test must be for a minimum of 60 seconds duration following system stabilization. The test must be conducted at a constant input voltage to the acoustic driver(s). (4) Repeat item (3), alternately with and without the ICD in place. A minimum of three tests of each configuration is required. To be acceptable, the total variation of the 55° microphone on-line OASPL signal (averaged for a 1 minute duration) for all three test conditions of each configuration shall not exceed 0.5 dB. NOTE: Physically moving the ICD alternately in and out of place for this calibration may be eliminated if it is demonstrated that the ICD positioning does not affect the calibration results, (S) All measured data are to be corrected for sound pressure level variations as measured with the near-field microphone and for atmospheric absorption to 77°F and 70 percent RH conditions using the slant distance between the outer microphones and the acoustic driver(s). (6) The calibration for each one-third octave band at each microphone is the difference between the average of the corrected SPLs without the ICD in place and the average of the corrected SPLs with the ICD in place. (7) The tests must be conducted under wind and thermal conditions that preclude acoustic shadowing at the outer microphones and weather induced variations in the measured SPL data. f. Acoustic Shadowing. (1) If ground-plane microphones are used, the applicant should demonstrate that acoustic "shadowing" resulting from either thermal gradients 26 Chap 3 Par 17 ### Page 31 or wind does not significantly influence the measured sound pressure levels. When there is a wind in the opposite direction to the sound wave propagating from the engine, or when there is a substantial thermal gradient in the test area, refraction can influence near-ground plane microphone measurements to a larger degree than measurements at greater heights. Previous evidence, or data from supplemental tests, may be used to demonstrate that testing within the approved weather restrictions at a particular test site results in no significant evidence of shadowing. In lieu of this evidence, the supplemental noise demonstration tests should include an approved method to indicate the absence of shadowing effects. One such method that has been approved involves the use of three measurements located as follows: - average windspeed at engine centerline height (WCL) - air temperature at engine centerline height (TCL) - air temperature at the ground-plane microphone height (TMIC) (2) Ground plane microphones should be mounted within one-half microphone diameter or one-quarter wavelength of the highest frequency to be recorded, whichever is least. Generally, this spacing is 7 mm (0.25 inch). (3) The instruments for these measurements should be co-located at a distance from the engine similar to that of the microphones and placed close to the 90 degree noise measurement position without impeding the acoustic measurement. The limits that follow are in addition to any wind and temperature limits established by other criteria, such as the maximum crosswind at the engine and the maximum windspeed at the microphones if microphone windscreens are not used. For an acceptable data test point, the average windspeed and temperature should be within the following limits: - Engine centerline height at 16 ft. (assuming a 150 ft radius microphone array and smooth concrete pad): (TCL - TMIC)~ -7°F (-3.9°C), WCL ~ 11 km/h (6 kts) average during test time for each data point (typically 30 seconds), WCL!:: 15 km/h (8 kts) maximum during test time for each data point. - Engine centerline height at 21 ft. Chap 3 Par 17 (assuming a 150 ft radius microphone array and smooth concrete pad): (TCL - TMIC):l,. -7°F (-3.9°C), 27 ### Page 32 AC 36-4B WCL~19 km/h (10 kts) average during test time for each data point, WCL~23 km/h (12 kts) maximum during test time for each data point. 3/23/88 g. Microphone Locations. Microphones should be located at sufficient angular locations and at a height to provide adequate definition of the engine noise source characteristics. More specific guidance is given in SAE ARP 1846. The objective is to obtain measurements which can be corrected to a connnon reference condition. One method is to obtain or correct data to free field conditions (see paragraph 18(b) of this AC). Certification experience has been primarily limited to microphone installations near the ground or at centerline height. Because of the difficulties associated with extrapolating to flight conditions, the FAA suggests using near-ground plane microphone installations. However, the choice of microphone installations is dependent upon the specific test objectives and FAA approval of methods to be used for data normalization. It is currently required to use consistent microphone locations, heights, etc., for noise measurements of both the prior approved and changed aircraft version. h. Engine Power Test Conditions. A range of static engine operating conditions should be selected to correspond to the expected maximum range of in-flight engine operating conditions. A sufficient number of engine power settings should be tested to ensure that the 90 percent confidence intervals in flight projected EPNL can be established over the desired range. (See Appendix 5 of this AC.) i. Data Analysis System Compatibility. If more than one data analysis system is used for the acquisition of static data, compatibility of the two analysis systems is necessary. This can be verified by analyzing the same tape on both systems. The systems are compatible if the resulting differences in the projected in-flight EPNL values are no greater than 0.5 EPNdB. Evaluation should be conducted at flight conditions representative of those fo~ certification. j. Data Acquisition, Analysis, and Normalization. For each power setting designated in the test plan, the engine performance, meteorological, and sound pressure level data should be acquired and analyzed. Instrumentation and test procedures described in SAE ARP 1846 are acceptable. Sound measurements should be normalized to consistent conditions (generally free field) and consist of 24 one-third octave band RMS sound pressure levels between 45 and 11,200 Hz for each measurement (microphone) station. Before projecting the static engine data to flight conditions, the sound pressure level data should be corrected for the following effects: (1) Frequency response characteristics of the data acquisition and analysis system. (2) Spherical divergence over the sound propagation path. Chap 3 28 Par 17 ### Page 33 (3) Atmospheric absorption over the sound propagation path due to the difference in test and Part 36 reference ambient conditions of temperature and relative humidity. (4) Spectral distortions resulting from ground reflections (methods to account for spectral distortions are presented in SAE Aerospace Information Report (AIR) No. 1672B). (5) Contamination by background ambient or electrical system noise. 18. PROJECTION OF STATIC ENGINE DATA TO AIRPLANE FLIGHT CONDITIONS. The static engine sound pressure level data acquired at each angular location should be analyzed and adjusted to account for the effects identified in paragraph 17(h) of this AC. They should be then projected to the same airplane flight conditions used in the development of the approved NPD. As appropriate, the projection procedure includes the effects of source motion, number of engines and shielding, flight geometry and atmospheric propagation. To account for these effects, the measured static noise data should be analyzed to determine contributions from individual noise sources. After projecting the one-third octave band spectral data to flight conditions, EPNL values should be calculated for the revised NPD plot. Guidelines on the elements of an acceptable projection procedure are provided in this section. This process is also illustrated in Figures 4 and 5. It is not intended that the procedure illustrated in Figures 4 and 5 should be exclusive. There are several options, depending upon the nature of the powerplant noise sources and the relevance of individual noise sources to the EPNL of the airplane. The method presented does, however, specify the main features that should be considered in the computational procedure. It is also not necessary that the computations should always be carried out in the order specified. There are interrelations between the various steps in the procedure which depend on the particular form of the computation being followed. Hence the most efficient manner of structuring the computation cannot always be predetermined. NOTE: Static-to-flight projection procedures for installations or engine types that are not similar to those that have been previously approved may require flight validation. a. Engine Installation Effects. There are several engine installation effects which can modify the generated noise levels but which are not derivable from static tests. Additional noise sources, such as jet/flap or jet/wing interaction effects, may be introduced on a derivative version of the airplane which are not present on the flight datum airplane. The aircraft's noise directivity patterns (field shapes) may be modified by wing/nacelle or jet-by-jet shielding, tailplane and fuselage scattering or airframe reflection effects. However, general methods to adjust for these effects are not yet available. It is, therefore, important that, before the following procedures are approved for the derivative version of the airplane, the geometry of the airframe in the vicinity of the engines can be shown to be essentially identical to that of the flight datum airplane so that the radiated noise is unaffected. Chap 3 Par 18 29 ### Page 34 **30 AC 36-4B FIGURE 4** **EXAMPLE OF A METHOD FOR PROJECTING STATIC ENGINE** **DATA TO FLIGHT CONDITIONS JET .AIRFRAME** NOISE CAPPROAC ONLY) STATIC DATA e determined. using a rmre general fo:rnu.lation than used for a cluster of data points. For this zoore general case, confidence intervals should be calculated. al>out a regression curve for fli,tlt test data and for static test data. 'lbese confidence intervals mould be used 1n calcµlati.n, the confi<1ence intervals for the pooled data sets that constitute HPD maps. 2. cx::fiFIIECE I:tll'mVAL FOR 1niE MEAN OF CLUS'lliRED ~ Let L : data (EP.NL) H = :rn.m:>er of data points K = order of regression curve (k. = o for clustered data) v = degrees of freedom = H - K - 1 t = Student's t distribution for 90 percent confidence interval and v degrees of freedom for exanple: t = 2.015 for H = 6 (v = 5) t = 1. 6,ij-5 for N: infinity t. (mean) = ----- H H -2 ::IE: (L - L) 1 l r I H I"~ L.2 - I Lit L K 1 H I ( ~ LL )2· l s (stanHFII>mCE I:tmllWAL -~ A MEAN RmRESSIOH CURVE. '!be general exp:-ess1on for a regression curve ne.y be written as: EPNL: cio + a1X + azx2 +. • • • ~xk Where EPNL: noise level X = parameter such as WR, RPM. altitude, etc. a= coefficients for the regression curve k = polynomial order 0£ regression curve de-fine: AEPHl. = difference between regression curve and individual data point (X constant) N = number of data points v = degrees of freedom = N - k -1 t = students t distribution for 90 percent confidence interval EPNL0 , X0 = a particular point on the regression curve CI0 = confidence interval for a particular point For a linear (k = 1) regressioni the equation is as follows: Note: 2 N 2 ~ (AEPNL)2 (Xo - X)2 1 CI0 : ± t -+ N-K-1 N N ~ (X. - X)2 L N L X 1 L "4l.ere X : N k. is normally limited to K = 2. K = 3 may :be approved When there 1s a :i;nysical basis for that polynomial order of curve fit. Par 3 ### Page 63 -3123/88 AC 36-4B Appendix 5 For a higher order regression curve, a particular EPm..0 is: EPHLo = 8o + a1Xo + a2Xo2 + •• + ~Xol<. For higher order regression curves, the regression coefficients are usually calculated using conputer-matrix solutions, since there are many terms. Defining the vectors: x' : [ 1 X x2. 0 0 0 a row vector Xo = r 1 1 I X~ I I X I a colum vector I • I I I I . I L x~ .1 and #-A-1 = the var1ence-covar1ence matrix for the ~ values 'lben the 90 percent conf1dence interval £or a particular noise level, EPm..0 , on a regression curve of EPNL vs. X is: CI0 = ± t N-X:-1 Par 3 3 ### Page 64 AC 36-4B Appendix 5 4 ..I z ~ Ill ..I z ~ Ill ,oo " " 97 H to 94 .., 92 t1 to et "' llf7 N so Cl4 c,;, IS2 cs, IIO 79 110 108 10C5 104 102 100 tel H 14 92 to "' 156 114 IS2 GO n, 3/23/88 Linear Example 2 2., ~ho111i:1ndl) :z.;, 2.4 NI C N~on -o Quadratic Example 2 -Q Par 3 ### Page 65 Appendix 5 .!!. OONFIDEMCE ImERVAL FOR POOLED DATA SEIS. 'Ihe confidence interval for pooled data sets may be calculated as follows. For exanple, for three data sets (flight test data and two d.1.fferent static tests) - each with its own confidence interval: 1s: let zz1 = (Cli/ti)2 (square of the confidence interval without the Student's t term for each data set) then for tbree data sets, the confidence interval for the pooled data where t' = 5, REFERENCES Par 4 (a) Kendall, M. G. and Stu.art, A. 1 'Ihe AdVanced 'Iheory of Statistics, Volurres 1 and 2, Hafner, New York, 1973. (b) Yule, G. U. and Kendal 1, H. G., An Introduction to the 'Iheory of Statistics, 14th ed, Griffin, New YorK, 1950. (c) Walpole, R. E and Myers, R. R, Probability and Statistics for Engineers and Scientists, MacMillan, New YoeK, 1972, (d) Cochran, W. G., "Approximate Signi£icance Levels of the Behrens-Fisher Test," Biometrics, 20, 191-195, 1964. (e) Snedecor, G. W. and Cochran, W. G., Statistical Methods, 6th ed., Iowa State University Press, .Arres Iowa, 1968. ### Page 66 AC 36-4B Appendix 5 6 3/23/88 Table 5-1 STUDENT'S t DISTRIBUTION FOR 90 PERCENT CONFIDENCE INTERVAL Degree of freedom (V) t 1 6.314 2 2.920 3 2.353 4 2.132 5 2.015 6 1.943 7 1.895 8 1.860 9 1.833 10 1.812 12 1.782 14 1.761 16 1.746 18 1.734 20 1.725 24 1.711 30 1.697 60 1.671 00 1.645 Par 5 •U.S. GOVERNMENT' PIJNTlNC OlflCE, 1 t t 2 •3 12 ·, 17/5ss96 --- OpenAirDocs.org is an independent research project of [Contrail Labs](https://contraillabs.com/). It is not affiliated with the FAA or the NTSB. Source documents are U.S. government works in the public domain.