Helicopter Acoustics, part 2

NASA Conference Publication 2052

Part I1

Helicopter Acoustics

Proceedings of an International Specialists Symposium held at NASA Langley Research Center Hampton, Virginia May 22-24, 1978

TECH LIBRARY KAFB, MM

I llllll I I I I I 1 1 1 1 1 I l l 1 1 I l l 1 1 lllll I i l 1 1 llll I l l 1

0067346

NASA Conference Publication. 2052

Part I1

Helicopter Acoustics

Proceedings of an International Specialists Symposium sponsored by the American Helicopter Society, Washington, D . C . ; the U . S . Army Research Office, Durham, North Carolina; and the Langley Research Center, Hampton, Virginia, and held at Langley Research Center May 22-24, 197.8 National Aeronautics and Space Administration Scientific and Technical Information Offico PREFACE The papers included herein were presented at the International Specialists Symposium on Helicopter Acoustics which was held at the NASA Langley Research The symposium was jointly sponsored by the American Center on May 22-24, 1978.

Helicopter Society, the U . S . Army Research Office, and the NASA Langley Research Center. Exterior and interior noise problems were addressed both from the physics and engineering as well as the human factors points of view.

The objective of the symposium was to explore the role of technology in closing the gap between what the customers and the regulating agencies would like to have and what is currently available. In this regard, papers were pre- sented on noise regulation concepts, human factors and criteria, rotor noise generation and control, design, operations and testing for noise control, heli- copter noise prediction and research tools and measurements. There was active participation by attendees from a number of foreign countries.

The included papers are largely as submitted as camera-ready copy at the time of the symposium. Only minor editorial changes have been performed and a title page and abstract have been added. The assistance of the Scientific and Technical Information Programs Division of the NASA Langley Research Center in publishing these proceedings is gratefully acknowledged.

Use of manufacturers or identification of commercial products in this report does not constitute an official endorsement of such manufacturers or products, either expressed or implied, by the National Aeronautics and Space Administration.

iii CONTENTS P a r t I"

P F J F A C E . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii

REGULATIONS

1. HELICOPTER EXTERNAL NOISE REQUIREMENTS - FAA PERSPECTIVE . . . . . . . 1

C h a r l e s R. Foster

2. HELICOPTER NOISE REGULATIONS: AN INDUSTRY PERSPECTIVE . . . . . . . . 1 7

R. A. Wagner

3 . NOISE REQUIREMENTS FROM A MILITARY POINT OF VIEW . . . . . . . . . . . 33

Charles C. C r a w f o r d , Jr.

4. THE IMPACT OF URBAN OPERATIONS ON HELICOPTER' NOISE REQUIREMENTS . . . 45

S t a n l e y R. S p e c t o r ROTOR NOISE

5. PREDICTION AND REDUCTION OF ROTOR BROADBAND NOISE . . . . . . . . . . . 6 1

R i c h a r d E . H a y d e n and Krishna S. A r a v a m u d a n

6 . THEORETICAL MODELS OF HELICOPTER ROTOR NOISE . . . . . . . . . . . . . 89

D. L. H a w k i n g s

7 . NOISE DUE TO ROTOR-TURBULENCE INTERACTION . . . . . . . . . . . . . . 109

R. K. A m i e t

8. THEORY ON ACOUSTIC SOURCES . . . . . . . . . . . . . . . . . . . . . . 127

S. E. Wright

9. POTENTIAL ACOUSTIC BENEFITS OF CIRCULATION CONTROL ROTOR . . . . . . . 149

R o b e r t M. W i l l i a m s and Ian C. Cheeseman

10. HELICOPTER NOISE RESEARCH AT THE LANGLEY V/STOL TUNNEL . . . . . . . I81

Danny R. H o a d and George C. G r e e n e i l l . EXPLORATORY WIND-TUNNEL INVESTIGATION OF THE EFFECT OF THE MAIN

ROTOR WAKE ON TAIL ROTOR NOISE . . . . . . . . . . . . . . . . . . . 205

R o b e r t J. Pegg and P h i l l i p A. S h i d l e r

*

Papers 1 t o 1 9 are presented under separate cover.

V

WIND TUNNEL INVESTIGATIONS OF MODEL ROTOR NOISE AT LOW T I P SPEEDS . . . 2 2 1

1 2 .

K. S. A r a v a m u d a n , A. L e e , and W. L. H a r r i s 1 3 . HELICOPTER EXTERNAL NOISE PREDICTION AND CORRELATION WITH

FLIGHT TEST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263

B h a r a t P. G u p t a

14. FULL-SCALE TESTING OF AN OGEE T I P ROTOR . . . . . . . . . . . . . . . . 277

Wayne R. Mantay, R i c h a r d L. C a m p b e l l , and P h i l l i p A. S h i d l e r

15. HOVERING IMPULSIVE NOISE - SOME MEASURED AND CALCULATED RESULTS . . . . 309

D. A. B o x w e l l , Y. H. Yu, and F. H. S c h m i t z

1 6 . IMPROVED METHODS FOR CALCULATING THE THICKNESS NOISE . . . . . . . . . 323

Y o s h i y a N a k a m u r a and A k i r a Azuma THE IMPORTANCE OF QUADRUPOLE SOURCES I N PREDICTION OF TRANSONIC 1 7 .

T I P SPEED PROPELLER NOISE . . . . . . . . . . . . . . . . . . . . . . 339

D o n a l d B. H a n s o n and Martin R. Fink 18. BOUNDS ON THICKNESS AND LOADING NOISE OF ROTATING BLADES AND THE

FAVORABLE EFFECTS O F BLADE SWEEP ON NOISE REDUCTION . . . . . . . . . 373

F. Farassat, Paul A. N y s t r o m , and Thomas J. B r o w n

. . . . 387

1 9 . A STUDY OF THE NOISE RADIATION FROM FOUR HELICOPTER ROTOR BLADES A l b e r t L e e and Marianne Masher PART I1 HUMAN FACTORS AND CRITERIA

20 SUBJECTIVE EVALUATION O F HELICOPTER BLADE SLAP NOISE . . . . . . . . . 4 0 3

W. J. G a l l o w a y

RATING HELICOPTER NOISE . . . . . . . . . . . . . . . . . . . . . . . . . 419

2 1 .

John W. Leverton, B. J. S o u t h w o o d , and A. C. P i k e

2 2 . ANNOYANCE 'OF HELICOPTER IMPULSIVE NOISE . . . . . . . . . . . . . . . . 439

F. d ' A m b r a and A. D a m o n g e o t

2 3 . ANNOYANCE DUE TO SIMULATED BLADE-SLAP NOISE . . . . . . . . . . . . . . 4 6 3

C l e m a n s A. P o w e l l

2 4 . HUMAN RESPONSE TO AIRCRAFT-NOISE-INDUCED BUILDING VIBRATION . . . . . . 4 7 9

J i m m y M. C a w t h o r n , T h o m a s K. D e m p s e y , and R i c h a r d D e L o a c h v i 25. A METHOD FOR DETERMINING INTERNAL NOISE CRITERIA BASED ON PRACTICAL

SPEECH COMMUNICATION APPLIED TO HELICOPTERS . . . . . . . . . . . . . 493

Harry Sternfeld, Jr., and Linda Bukowski Doyle DESIGN AND OPERATIONS

26. THE EFFECTIVE ACOUSTIC ENVIRONMENT OF HELICOPTER CREWMEN . . . . . . . 513

Robert T. Camp, Jr., and Ben T. Mozo 27. THE EFFECT OF OPERATIONS ON THE GROUND NOISE FOOTPRINTS ASSOCIATED

WITH A LARGE MJLTIBLADED, NONBANGING HELICOPTER . . . . . . . . . . . 519

David A . Hilton, Herbert R. Henderson, Domenic J. Maglieri, and William B. Bigler I1 28. A STATIC ACOUSTIC SIGNATURE SYSTEM FOR THE ANALYSIS OF DYNAMIC

FLIGHT INFORMATION . . . . . . . . . . . . . . . . . . . . . . . . . 535

Daniel J. Ramer

29. AN ACTIVE NOISE REDUCTION SYSTEM FOR AIRCREW HELMENTS . . . . . . . . . 545

Peter D. Wheeler, David Rawlinson, Stephen F. Pelc, and Tony P. Dorey

30. DESIGN OF HELICOPTER ROTORS TO NOISE CONSTRAINTS . . . . . . . . . . . 551

Edward G. Schaeffer and Harry Sternfeld, Jr.

31. THE COST OF APPLYING CURRENT HELICOPTER EXTERNAL NOISE REDUCTION

METHODS WHILE MAINTAINING REALISTIC VEHICLE PERFORMANCE . . . . . . . 563

Michael A. Bowes INTERIOR NOISE

32. HELICOPTER CABIN NOISE - METHODS OF SOURCE AND PATH IDENTIFICATION

AND CHARACTERIZATION . . . . . . . . . . . . . . . . . . . . . . . . 583

Bruce S. Murray and John F. Wilby

33. A PRACTICAL APPROACH TO HELICOPTER INTERNAL NOISE PREDICTION . . . . . 595

Larry S. Levine and Jon J. DeFelice

34. HELICOPTER INTERNAL NOISE CONTROL - THREE CASE HISTORIES . . . . . . . 639

Bryan D. Edwards and Charlie R. Cox 35. AN ANALYTICAL METHOD FOR DESIGNING LOW NOISE HELICOPTER

TRANSMISSIONS.. . . . . . . . . . . . . . . . . . . . . . . . . . . 657

Robert B. Bossler, Jr., Michael A. Bowes, and Allen C . Royal

36. THE INFLUENCE OF THE NOISE ENVIRONMENT ON CREW COMMUNICATIONS . . . . . 679

John W. Leverton vii 37. HELICOPTER INTERNAL NOISE REDUCTION RESEARCH ANI) DEVELOPMENT

APPLICATION TO THE SA 360 AND SA 365 DAUPHIN . . . . . . . . . . . . 695

H. J. Marze and F. d ' A m b r a STATE OF THE ART

38. THE STATUS OF ROTOR NOISE TECHNOLOGY - ONE MAN'S O P I N I O N . . . . . . . 723

R i c h a r d P . White, Jr.

39. TRENDS I N LANGLEY HELICOPTER NOISE RESEARCH . . . . . . . . . . . . . 7 8 1

H. Hubbard, D o r n e n i c J. Maglieri, and D a v i d G. Stephens H a r v e y

40 AEROACOUSTIC RESEARCH - AN ARMY PERSPECTIVE . . . . . . . . . . . . . 7 9 7

H. A n d r e w Morse and F r e d r i c H. S c h m i t z SESSION REVIEWS

41. REGULATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 819

Charles J. H o c h

42. ROTOR NOISE PREDICTION . . . . . . . . . . . . . . . . . . . . . . . . 823

A. R. George

4 3 . MODEL AND FULL-SCALE TESTING OF ROTOR NOISE . . . . . . . . . . . . . 8 2 7

F. H. Schmitz

44. DESIGN AND OPERATIONS . . . . . . . . . . . . . . . . . . . . . . . . 833

E. R. Wood

45. I N T E R I O R N O I S E . . . . . . . . . . . . . . . . . . . . . . . . . . . . 839

R o n a l d G. Schlegel ~

46. HUMAN FACTORS AND CRITERIA . . . . . . . . . . . . . . . . . . . . . . 843

E. Gene L y m a n v i i i SUBJECTIVE EVALUATION OF HELICOPTER BLADE SLAP NOISE W i l l i a m J. Galloway Bolt Beranek and Newman I n c .

S U M M A R Y Several methods f o r a d j u s t i n g EPNL t o account f o r i t s underestimate of judged annoyance a r e a p p l i e d t o eight h e l i c o p t e r f l y o v e r n o i s e s i g n a t u r e s having various degrees of b l a d e slap.

A p r o p o s a l f r o m an IS0 working group f o r making such adjustments i s i n v e s t i g a t e d f o r these data as w e l l as two s e t s of data sub- m i t t e d by France t o t h e I C A O Committee on A i r c r a f t Noise, Working Group B. When a l l data a r e combined, t h e IS0 proposal i s l i t t l e b e t t e r t h a n s i m p l y adding an a r b i t r a r y f i x e d adjustment of 3 d e c i b e l s t o EPNL.

INTRODUCTION Means f o r measurement of t h e p h y s i c a l c h a r a c t e r i s t i c s of impulsive n o i s e produced b y h e l i c o p t e r blade s l a p and accounting f o r t h e underestimate of judged annoyance b y EPNL have been s t u d i e d by a number o f groups i n Europe and t h e U S A over t h e l a s t few years. Working Group 2 on A i r c r a f t Noise of IS0 TC43/ SC1, Acoustics/Noise, has considered a number of p o s s i b l e measurement and assessment procedures a t t h e request o f ICAO/CAN/ WGB f o r use i n i t s development of n o i s e c e r t i f i c a t i o n procedures f o r h e l i c o p t e r s . A draft proposal f o r an impulsive n o i s e c o r r e c t i o n procedure emerged from IS0 i n January 1978 and has been c i r c u l a t e d f o r comment. The procedure i s based on a d i g i t a l a n a l y s i s of t h e f l y o v e r s i g n a l .

The b a s i c psychoacoustical data used t o d e r i v e t h e IS0 proposal were o b t a i n e d f r o m a combination of steady s t a t e and s i m u l a t e d h e l i c o p t e r blade slap n o i s e s . T h i s paper d e s c r i b e s t h e i n v e s t i g a t i o n of t h e a b i l i t y of an analog a n a l y s i s of a number of simulated n o i s e s and e i g h t recorded n o i s e s i g n a t u r e s from a c t u a l h e l i c o p t e r s , as w e l l as t h e use o f t h e IS0 and o t h e r d i g i t a l l y based procedures, t o 'account f o r t h e r e s u l t s o f psychoacoustical judgments of these s i g n a l s . Comparisons of t h e a p p l i c a t i o n of t h e IS0 procedures t o t h e eight recorded h e l i c o p t e r n o i s e s i g n a l s and to two sets of French data on simulated h e l i c o p t e r s are made f o r t h e separate data sets and t o t h e aggregated data.

ABBREVIATIONS AND SYMBOLS Abbreviations IS0 I n t e r n a t i o n a l S t a n d a r d i z a t i o n Organization I C A O I n t e r n a t i o n a l C i v i l Aviation Organization EPNL e f f e c t i v e perceived n o i s e l e v e l PNLT tone-corrected perceived n o i s e l e v e l Rep. r e p e t i t i o n TSC T r a n s p o r t a t i o n Systems Center Symb o 1 s CF c r e s t f a c t o r , d e c i b e l s maximum v a l u e of C F over an e v e n t , d e c i b e l s CFM

-

l o g a y i t h q i c average over an event of c r e s t f a c t o r s C F O . 5 obtained f o r each 0 . 5 second o f t h e event, d e c i b e l s C I French impulse c o e f f i c i e n t I IS0 impulse f a c t o r derived from sampled v o l t a g e s

r product moment c o r r e l a t i o n c o e f f i c i e n t

v o l t a g e sampled a t i t h time increment 'i c a l c u l a t e d a d d i t i v e adjustment to EPNL, d e c i b e l s A C s u b j e c t i v e d i f f e r e n c e between judged and measured

As

EPNL, d e c i b e l s MEASURES O F IMPULSIVENESS The psychoacoustical study r e p o r t e d i n r e f e r e n c e 1 i n v e s t i g a t e d t h e use of s e v e r a l analog and d i g i t a l measures of impulsiveness t h a t had been proposed i n t h e IS0 working group.

Since t h e t i m e of that r e p o r t t h e IS0 proposal has emerged as a coalescence of twio p r o p o s a l s , one f r o m t h e National Physical Laboratory i n England and t h e o t h e r from France. The two computational procedures r e s u l t s i n impulse measures t h a t d i f f e r only b y a s u b t r a c t i v e constant of u n i t y when based on t h e same d i g i t a l sampling i n t e r v a l s and i n t e g r a t i o n times. I n t h i s paper both t h e IS0 proposal and t h e l a s t French proposal are used, s i n c e t h e t r a n s f e r f u n c t i o n s between t h e impulsiveness measures and t h e s u b j e c t i v e c o r r e c t i o n s t o EPNL a r e d i f f e r e n t . I n a d d i t i o n t o t h e d i g i t a l techniques, s e v e r a l analog measures of c r e s t f a c t o r were used i n t h e a n a l y s i s of t h e s u b j e c t i v e data i n r e f e r e n c e 1.

I n a l l t h e procedures t h e b a s i c concept i s t o d e r i v e an adjustment f a c t o r , f o r each 0 . 5 second i n t e r v a l o f t h e f l y o v e r , which i s added t o t h e measured value of tone-corrected perceived n o i s e l e v e l (PNLT) f o r t h a t 0 . 5 second i n t e r v a l , b e f o r e i n t e - g r a t i n g (summing) over t h e event t o o b t a i n EPNL. The d i f f e r e n c e s between t h e impulse adjustment procedures l i e i n how t h e a d j u s t - ment increment i s determined f o r each 0.5 second i n t e r v a l .

D i g i t a l Analyses The n o i s e s i g n a l v o l t a g e i s A-weighted, passed through a 2000 Hz low pass a n t i - a l i a s i n g f i l t e r , t h e n d i g i t a l l y sampled a t 5000 ( o r an i n t e g e r m u l t i p l e of 5000) samples per second. I n t h e French proposal the 2500 samples i n each 0 . 5 second i n t e r v a l a r e combined t o determine an impulsiveness c o e f f i c i e n t C I ( r e f . 2 ) : C I L Note t h a t t h e denominator i s t h e square of t h e mean-square v o l t a g e during t h e time i n t e r v a l , denoted i n t h e IS0 procedure as " S " . I n t h e IS0 procedure, t h e impulsiveness q u a l i t y I i s c a l c u l a t e d from t h e same sampled d a t a ( r e f . 3 ) : I = -

2500 c

2500[vi2s- l 2

i=1 Thus, C I = 1 + 1.

The values of C I and I a r e converted through t r a n s f e r f u n c t i o n s t o d e c i b e l adjustments, A , which are added t o t h e PNLT

values i n t h e same time i n t e r v a l . The form of t h e s e t r a n s f e r

f u n c t i o n s has v a r i e d a t d i f f e r e n t times during t h e i r e v o l u t i o n .

I n each case t h e a i m was t o develop a f u n c t i o n t h a t would e m p i r i c a l l y f i t t h e t h e n a v a i l a b l e s u b j e c t i v e data and, a t t h e same time, have a zero adjustment f o r "non-impulsive" n o i s e , such as that produced b y conventional j e t a i r c r a f t . The c u r r e n t forms (.refs. 2 and 3 ) of t h e t r a n s f e r f u n c t i o n s may be expressed a s :

French: A = - 6.875 + 13.75 loglo C I

( 3 )

I S O : A = - 2 . 4 + 8 loglo I

where A i s r e s t r i c t e d t o 0 < A < 5.5.

- -

Analog Analysis c l a s s i c a l way t o d e s c r i b e t h e impuls,veness o f a s,gnal i s through i t s c r e s t f a c t o r , i t s peak-to-rms r a t i o . Expressed i n d e c i b e l s , where Lpk i s t h e peak sound p r e s s u r e l e v e l , and L i s t h e mean square sound p r e s s u r e l e v e l , c r e s t f a c t o r CF = L - L.

For random n o i s e CF i s of t h e o r d e r of 1 2 d e c i b e l s and, f8r

s e v e r e h e l i c o p t e r b l a d e s l a p , may be as high as 20 o r more d e c i b e l s , depending upon what, i f any, frequency-weighting i s employed. A s a measure of h e l i c o p t e r blade s l a p , Leverton has proposed t h e c r e s t f a c t o r measured i n t h e 250 Hz octave band, w h i l e a number o f d i f f e r e n t i n v e s t i g a t o r s have used A-weighted sound l e v e l / c r e s t f a c t o r s .

I n t h i s study A-weighted sound l e v e l c r e s t f a c t o r i n d e c i b e l s is' used, as measured w i t h a B&K 2209 sound l e v e l meter.

I n making peak measurements t h e instrument uses a 1 0 microsecond RC d e t e c t o r w i t h a r e s e t a b l e peak hold c i r c u i t and provides a c c u r a t e sound l e v e l measurements f o r c r e s t f a c t o r s o f more t h a n 30 d e c i b e l s . Two d i f f e r e n t c r e s t f a c t o r s have been used i n t h e study. The simplest i s the maximum c r e s t f a c t o r measured during a f l y o v e r , i r r e s p e c t i v e of when it occurred, and i s abbreviated as CFM, A more complex measure i s t h e mean-square average o f the s e p a r a t e maximum c r e s t f a c t o r s obtained i n each 0 . 5 second i n t e r v a l of t h e f l y o v e r , abbreviated as mOa5.

SUMMARY OF SUEiJECTXVE TESTS A complete d e s c r i p t i o n of t h e s e experiments i s provided i n r e f e r e n c e 1. A b r i e f summary i s provided h e r e .

Steady-State Synthesized S i g n a l s Eight d i f f e r e n t s i g n a l s were constructed f o r judgement a g a i n s t two d i f f e r e n t non-impulsive s9gnals. All s i g n a l s had d u r a t i o n s of 1 0 seconds a t constant l e v e l , with, Q.5 second on ramps. Three d i f f e r e n t non-impulsive n o i s e s p e c t r a were and o f f used t o r e p r e s e n t d i f f e r e n t h e l i c o p t e r s p e c t r a . The firs% non-.

impulsive n o i s e was a r e p l i c a of t h e s i g n a l r e p o r t e d bp F u l l e r i n experiments at t h e National Physical Laboratory (INPLI, i n England ( r e f . 4 ) . The o t h e r two were r e p r e s e n t a t i v e of t8.e s p e c t r a o f l a r g e multi-bladed and s m a l l e r two-bladed h e l i c o p t e r s .

Impulsive n o i s e s i m u l a t i o n s were made By mixing single s i n e waye p u l s e s , repeated at a s p e c i f i e d r e p e t i t i o n rate, with t h e broad- hand non-impulsive "background" s p e c t r a . Th.e s i g n a l s may be described by t h e "b.ackground" non-impulsive spectrum, tb-e funda- mental frequency of t h e s i n e p u l s e , t h e frequency of t h e p u l s e r e p e t i t i o n r a t e , and t h e l e v e l d i f f e r e n c e i n d e c i b e l s between t h e peak sound p r e s s u r e l e v e l of t h e s i n e p u l s e s t o t h e o v e r a l l of. th.e non-impulsive spectrum. (-Note t h a t sound p r e s s u r e l e v e l t h i s i s not t h e c r e s t f a c t o r f o r th.e combined s i g n a l s . ) Time-Varying Synth.esized S i g n a l s S i x o f t h e s t e a d y - s t a t e s i g n a l s were modified t o become time varying s i m u l a t i o n s of f l y o v e r s i g n a l s by use o f a v a r i a b l e g a i n t o provide a t r a p e z o i d a l time p a t t e r n i n which t h e a m p l i f i e r o v e r a l l s i g n a l l e v e l was i n c r e a s e d a t a r a t e of 2 d e c i b e l s p e r second t o a maximum l e v e l , h e l d f o r 2 seconds, t h e n decreased i n l e v e l a t a decay r a t e of 2 d e c i b e l s p e r second (providing s i g n a l s 1 2 seconds wide a t t h e p o i n t s i n t h e t i m e p a t t e r n t h a t a r e 1 0 d e c i b e l s below t h e maximum l e v e l ) .

I I n o r d e r t o s i m u l a t e t h e e f f e c t of d i r e c t i v i t y on impulse noise during a f l y o v e r , two of t h e s i g n a l s were f u r t h e r modified t o fadeout t h e impulsive part of t h e s i g n a l during t h e 2 second maximum l e v e l p o r t i o n of t h e time p a t t e r n . Thus t h e s e s i g n a l s had impulsive content on t h e i n c r e a s i n g l e v e l p o r t i o n of t h e s i g n a l and no impulses on t h e decaying p o r t i o n of t h e s i g n a l , as i n t h e case w i t h most a c t u a l h e l i c o p t e r n o i s e s i g n a t u r e s .

Recorded H e l i c o p t e r N o i s e S i g n a l s Nine recorded h e l i c o p t e r s i g n a l s were s e l e c t e d from t h m e obtained i n a comprehensive measurement program conducted By FAA/TSC t o d e f i n e t h e n o i s e c h a r a c t e r i s t i c s produced by a v a r i e t y of h e l i c o p t e r s during l e v e l f l y o v e r and approach maneu- v e r s under c o n s i d e r a t i o n f o r n o i s e c e r t i f i c a t i o n . The s i g n a l s used i n t h e psychoacoustical tests were chosen t o r e p r e s e n t t h e range of h e l i c o p t e r designs and sizes c u r r e n t l y i n o p e r a t i o n t h a t produce s i g n i f i c a n t l y a u d i b l e b l a d e s l a p , w i t h one s i g n a l having n e g l i g i b l e b l a d e s l a p chosen as a comparison s i g n a l (S-61 i n l e v e l f l i g h t ) . The g e n e r a l c h a r a c t e r i s t i c s of the s i g n a l s s e l e c t e d and t h e o p e r a t i o n a l c o n d i t i o n s under which they were produced a r e described i n d e t a i l i n r e f e r e n c e 5. The events used i n t h i s study a r e l i s t e d i n t a b l e 1.

Experiment a1 F a c i l i t i e s A l l s t i m u l i were presented t o s u b j e c t s s e a t e d , one a t a time, i n an anechoic chamber. The i n d i v i d u a l t e s t s i g n a l s were recorded on i n d i v i d u a l magnetic tape loops t h a t could be s e l e c t e d a t w i l l through computer c o n t r o l . The playback system frequency response was equalized s o t h a t t h e s i g n a l as measured a t t h e l i s t e n e r ' s head p o s i t i o n reproduced t h e s i g n a l spectrum of t h e o r i g i n a l recording. The s i g n a l l e v e l s used i n a l l p r e s e n t a t i o n s were measured i n EPNL, as c a l c u l a t e d from real-time analyses o f t h e s i g n a l s obtained a t t h e l i s t e n e r ' s head p o s i t i o n , using t h e procedures o f FAR P a r t 36/ICAO Annex 1 6 . P a r t i c u l a r a t t e n t i o n was paid t o i n s u r e that t h e a n a l y s i s system properly measured t h e rms l e v e l s of t h e s i g n a l s w i t h h i g h c r e s t f a c t o r s .

T e s t Subjects Twenty c o l l e g e s t u d e n t s between t h e ages of 1 8 and 32 were used as s u b j e c t s . Half of t h e s e were women and half were men.

A l l s u b j e c t s were audiometrically screened t o a s s u r e t h a t t h e y were w i t h i n 20 d e c i b e l s of IS0 defined normal h e a r i n g .

T e s t Procedure was asked t o choose which of two s e q u e n t i a l l y Each l i s t e n e r reproduced s i g n a l s was more annoying. For each t e s t stimulus t h e experimental procedure, c a l l e d PEST (Parameter Estimation of S e q u e n t i a l T e s t i n g ) , i n a n i t e r a t i v e manner c o n t r o l l e d by a computer, v a r i e d t h e l e v e l of a comparison s t i m u l u s i n a succession of trials u n t i l t h e s u b j e c t ' s responses i n d i c a t e d t h a t t h e t e s t stimulus was s u b j e c t i v e l y equal t o t h e s t a n d a r d s t i m u l u s . The computer program randomizes t h e o r d e r of pre- s e n t a t i o n of t h e two s i g n a l s and v a r i e s t h e l e v e l of t h e t e s t stimulus i n both i n c r e a s i n g and decreasing f a s h i o n t o o b t a i n a convergence i n t h e judgements from t h e s u b j e c t , The convergence c r i t e r i o n used i n these tests was 1 d e c i b e l , and t h e allowed upper l i m i t i n number of trials was 3 0 . The 20 s u b j e c t s used an average of approximately 1Q t r i a l s each t o reach s t a b l e judgements f o r t h e s u b j e c t i v e e q u a l i t y between t h e t e s t and s t a n d a r d s t i m u l i . T h e d i f f e r e n c e i n EPNL between t h e t e s t and comparison s t i m u l i , averaged over a l l s u b j e c t s , was used as t h e measure of t h e s u b j e c t i v e underestimate of b l a d e s l a p b y EPNL.

Although t h e o r d e r of p r e s e n t a t i o n of t h e d i f f e r e n t t e s t s t i m u l i w a s randomized between s u b j e c t s , a l l s u b j e c t s were given a p r e t e s t t r a i n i n g s e s s i o n during which they were asked t o judge one of t h e NPL t e s t n o i s e s a g a i n s t i t s e l f . The average d i f f e r - ence i n judgements f o r t h i s t e s t w a s 0 . 4 d e c i b e l s , w i t h a s t a n d a r d e r r o r i n t h e mean of 0 . 3 d e c i b e l s .

I n a l l t e s t s t h e f i x e d l e v e l s i g n a l was reproduced a t a nominal EPNL value of 80 d e c i b e l s . The comparison s i g n a l l e v e l could b e v a r i e d as much as 30 d e c i b e l s above and below t h i s l e v e l .

SUMMARY O F RESULTS The s u d j e c t s , on average, judged impulsive s i g n a l s t o b e more annoying t h a n non-impulsive s i g n a l s by up t o 7 d e c i b e l s .

On t h e o t h e r hand, non-impulsive s i g n a l s of s u b s t a n t i a l l y d i f f e r e n t s p e c t r a l shape were equated on an EPNL basis w i t h i n 0 . 1 t o 0 . 4 d e c i b e l , on average. The standard e r r o r s i n t h e mean ( 2 0 s u b j e c t s ) ranged from 1 . 0 t o 1 . 8 , f o r a l l s i g n a l s , arzd from 1 . 0 to 1 . 4 f o r just t h e h e l i c o p t e r s . For t h e s m a l l number of s u b j e c t s , these s t a n d a r d e r r o r s are very a c c e p t a b l e .

A p h y s i c a l a n a l y s i s of each s i g n a l was made t o c a l c u l a t e t h e v a r i o u s impulsiveness measures. Adjustment f a c t o r s f o r PNLT were computed according t o t h e proposed t r a n s f e r f u n c t i o n s , added t o PNLT v a l u e s , and EPNL re-computed f o r each s i g n a l . The d i f f e r e n c e i n EPNL w i t h and without the impulse adjustment was t h e n compared t o t h e judged d i f f e r e n c e s . The d i f f e r e n c e s between EPNL computed w i t h c r e s t f a c t o r s added t o PNLT and without were compared d i r e c t l y w i t h t h e judgements.

The r e s u l t s o f t h e comparisons between c a l c u l a t e d and judged values were discouraging when a l l s i g n a l s were compared as a s e t . I n essence, the comparisons were u n c o r r e l a t e d , w i t h t h e best measure accounting f o r less than 20 percent of t h e variance i n a l i n e a r r e g r e s s i o n a n a l y s i s (r2 = 0 . 1 8 ) . When only t h e eight h e l i c o p t e r s were considered as a s u b s e t t h e p i c t u r e improved, f o r t h e French C I proposal, w i t h r2 = 0.69.

The crest f a c t o r measures d i d n o t improve, w i t h r2 = 0.17.

S c a t t e r

diagrams and t h e r e g r e s s i o n l i n e s o f A s on A c are shown i n

f i g u r e 1 f o r t h e French procedure and i n f i g u r e 2 f o r mOe5.

I n some o t h e r tests on the judged annoyance of impulsive sounds we have found preliminary evidence t h a t annoyance i n c r e a s e s w i t h crest f a c t o r when p u l s e r e p e t i t i o n rate is held c o n s t a n t , while w i t h c r e s t f a c t o r held c o n s t a n t annoyance v a r i e s with r e p e t i t i o n rate. The shape o f t h e s e n s i t i v i t y curve i s very much l i k e a v i s u a l f l i c k e r s e n s i t i v i t y curve, l i t t l e s e n s i t i v i t y a t low ( z 5 Hz) and h i g h (= 80 Hz) r e p e t i t i o n frequencies ( f l i c k e r f u s i o n i n t h e c a s e of v i s i o n ) w i t h a maximum i n s e n s i t i v i t y of r e p e t i t i o n frequencies o f the o r d e r of 30 t o 40 Hz. Using t h e zero airspeed blade passage frequency

as a measure, t h e product moment c o r r e l a t i o n between A s and

frequency accounts f o r 65% o f t h e v a r i a n c e i n t h e e i g h t h e l i - c o p t e r s i g n a l s ( r 2 = 0.65); however, f o r t h e e n t i r e s i g n a l s e t l i t t l e c o r r e l a t i o n r e s u l t e d ( r Z = 0 . 1 0 ) .

I n an attempt t o improve t h e p i c t u r e , m u l t i p l e r e g r e s s i o n s

of A s on a l i n e a r combination of c a l c u l a t e d adjustments, A C

( o r CF) and r e p e t i t i o n frequency were com u t e d . Typical r e s u l t s f o r t h e h e l i c o p t e r s were improvement i n r' f o r t h e French ad.lust- ment from 0.69 t o 0.87 and improvement i n r2 f o r CF0.5 from 0.17 t o 0.77. Standard e r r o r s f o r t h e r e g r e s s i o n improved .from 0.8 d e c i b e l s t o 0.5 decibels f o r t h e French adjustment, and from 1 . 4 t o 0 . 7 d e c i b e l s f o r moe5.

C O N F L I C T I N G VIEWS The p o s s i b l e use o f analog measures of c r e s t f a c t o r t o assess impulsiveness has not met w i t h much enthusiasm i n I S O , p a r t i c u l a r l y on t h e basis o f analyses r e p o r t e d from Franch.

Wright and Damongeot ( r e f . 6 ) argue t h a t c r e s t f a c t o r is a poor measure s i n c e , i n t h e i r tests, it provided poor r e s o l u t i o n f o r low impulsiveness s i g n a l s . Our contention i s t h a t t h e y d i d n o t follow t h e s p e c i f i e d measurement procedure, s i n c e i n t h e i r paper t h e y determined crest f a c t o r from v i s u a l a n a l y s i s on an o s c i l l o s c o p e .

I n a n o t h e r a n a l y s i s Berry and Robinson ( r e f . 7 ) found good c o r r e l a t i o n w i t h t h e i r data using a c r e s t f a c t o r determined from t h e l a r g e s t value of t h e i r d i g i t a l l y sampled v o l t a g e s used t o compute C I o r I. I n t h e one case where w e can compare t h e i r a n a l y s i s d i r e c t l y w i t h one o f o u r s , t h e i r d i g i t a l method c o r r e l a t e s w e l l (r2 of O.gl), f o r seven samples of s y n t h e s i z e d blade slap n o i s e , w i t h our analog a n a l y s i s . F u r t h e r , t h e s l o p e of t h e r e g r e s s i o n l i n e i s 1 . 0 1 , although t h e r e i s a 1 . 7 d e c i b e l o f f s e t a t CF = 0 .

A more b a s i c disagreement e x i s t s over t h e use o f r e p e t i t i o n rate i n an adjustment process. The p r i m a r y i s s u e i s t h e repe- t i t i o n rate t o be a t t r i b u t e d t o a h e l i c o p t e r w i t h dual main r o t o r s . If one takes the r e p e t i t i o n r a t e as that due t o t h e blade passage rate of one r o t o r only, t h e c o r r e l a t i o n w i t h r e p e t i t i o n rate is low. I f one takes t h e r e p e t i t i o n r a t e as twice t h i s number, t h e c o r r e l a t i o n i s r e t a i n e d . The f a c t i s , a dual set of p u l s e s e x i s t s , not q u i t e uniformly spaced ( c o n s t a n t r e p e t i t i o n time o v e r l a p of two s e t s ) , w i t h one s e t somewhat weaker t h a n t h e o t h e r , t h e i r r e l a t i v e crest f a c t o r s varying w i t h o p e r a t i n g c o n d i t i o n s .

The second f a c t i s t h a t t h i s k i n d of h e l i c o p t e r i s judged t o b e twice, i n d e c i b e l s , as annoying as o t h e r h e l i c o p t e r s having e s s e n t i a l l y the same degree of impulsiveness ,as c a l c u l a t e d by any of t h e impulsiveness measures. See f i g u r e s 1 and 2 f o r examples.

IS0 ADJUSTMENTS AND COMPARISONS OF DATA O f major i n t e r e s t at t h i s t i m e i s how w e l l t h e proposed IS0 adjustment procedure works on judged data. Analyses o f two sets o f French data and of a dubbed recording of our e i g h t h e l i c o p t e r - s i g n a l s , provided by u s , have been r e p o r t e d by A e r o s p a t i a l l e i n an I C A O working paper ( r e f . 8 ) . The following d i s c u s s i o n i s based on t h e data r e p o r t e d i n r e f e r e n c e 8.

Consider f i r s t t h e IS0 procedure a p p l i e d t o t h e e i g h t h e l i c o p t e r s . The s c a t t e r diagram showing t h e r e l a t i o n s h i p

between A s &nd A C i s p l o t t e d i n f i g u r e 3, along with the regres-

s i o n l i n e . A p o s i t i v e c o r r e l a t i o n e x i s t s , w i t h r2 of 0 . 6 0 , i n t e r c e p t of 0 . 2 , and s l o p e of 1.23.

The French judgement data a r e derived f r o m one set of t i m e - varying s i g n a l s and from one s e t of s t e a d y - s t a t e s i g n a l s . Con- s i d e r first t h e data f o r t h e time-varying s i g n a l s , as p l o t t e d

i n f i g u r e 4 . I n t h i s case, r2 = 0.38, but the values of A s a r e

n e g a t i v e l y c o r r e l a t e d , t h e r e g r e s s i o n l i n e s l o p e being - 0.49.

Combining these data w i t h t h e e i g h t h e l i c o p t e r s t o o b t a i n a l l t h e time-varying s i g n a l s r e s u l t s i n t h e p l o t of f i g u r e 5. Here t h e c o r r e l a t i o n i s meaningless, w i t h r2 = 0 . 0 4 .

The r e l a t i o n s h i p between A s and A c f o r t h e French steady-

s t a t e s i g n a l s provides a b e t t e r p i c t u r e , as s e e n i n f i g u r e 6 .

T h i s should b e b e t t e r , s i n c e these data are b a s i c a l l y t h o s e used t o d e r i v e t h e C I and IS0 t r a n s f e r f u n c t i o n s i n t h e first p l a c e . Here r2 = 0.85 and t h e s l o p e i s 1 . 0 4 .

F i n a l l y , combining t h e t h r e e sets o f data, as i n r e f e r e n c e 8, t h e p l o t i n f i g u r e 7 r e s u l t s . I n t h i s f i g u r e the 6 s i g n a l s used as comparison s t a n d a r d s have been omitted s i n c e t h e s u b j e c t i v e d i f f e r e n c e s f o r t h e impulsive s i g n a l s are judged r e l a t i v e t o these s t a n d a r d s . I n t h i s combined c a s e r2 = 0 . 3 4 and t h e s l o p e i s 0 . 6 9 .

C O N C L U D I N G REMARKS It seems c l e a r t h a t a t t h i s p o i n t o u r knowledge o f a good g’eneral p r e d i c t o r o f s u b j e c t i v e response t o impulsive n o i s e i s poor. It i s a l s o c l e a r that EPNL does underestimate annoyance due t o b l a d e s l a p . For these data t h e average underestimate i s about 3 d e c i b e l s f o r s i n g l e main rotor a i r c r a f t and about 6 f o r t h e d u a l main rotor a i r c r a f t . Considering t h e v a r i a b i l i t y i n t h e data, one might a r b i t r a r i l y use t h e s e c o n s t a n t values and i g n o r e any more e l a b o r a t e approach. One could simply apply these a d d i t i v e v a l u e s t o any h e l i c o p t e r t h a t had a maximum A-weighted crest f a c t o r o f more t h a n 1 4 d e c i b e l s .

REFERENCES 1. Galloway, W. J., I'Subjective Response t o Simulated and Actual H e l i c o p t e r Blade S l a p Noise," BBN Report No. 3573, December 1977.

2 .

Anon., "Impulsivity Indicators--Effect o f I n t e g r a t i o n T i m e on NPL I n d i c a t o r , " ICAO/CAN/WG B W P 1 6 ( F r a n c e ) , June 1977.

" F i r s t Draft Proposal f o r Amendment t o IS0 3891 'Acoustics- 3.

Procedure f o r Describing A i r c r a f t Noise Heard on t h e

Ground' Measurement o f Noise From H e l i c o p t e r s ," ISO/TC43/

SC1 ( S e c r e t a r i a t - 2 5 4 ) 356, January 1 9 7 8 .

4 . F u l l e r , H . C . , "Rating H e l i c o p t e r Noise: A Study of Sub- j e c t i v e Reaction t o Impulsive Sound," ISO/TC43/SCl/WG2 N 77, November 1976 ( B & R 3 ) .

T r u e , H . C . and Rickley, E . J . , "Noise C h a r a c t e r i s t i c s o f 5.

Eight Helicopters,11 Report No. FAA-RD-77-94, J u l y 1 9 7 7 .

6 . Wright, S. E . and Damongeot, A . , llPsychoacoustic S t u d i e s o f Impulsive Noise," Paper No. 55, T h i r d European Rotor- c r a f t and Powered L i f t A i r c r a f t Forum," September 1 9 7 7 .

Berry, B. F. and Chew, A . J . , " H e l i c o p t e r Noise: Analysis 7 .

o f Various D e s c r i p t o r s o f Impulsiveness, ISO/TC43/SC1/WG2e N 81, February 1977 (B&R 4 ) .

8. o f "Pulse R e p e t i t i o n Rate' Anon., "Analysis o f t h e E f f e c t on t h e Annoyance o f H e l i c o p t e r Impulsive Noise," ICAO/CAN/WGB WP 8 ( F r a n c e ) , A p r i l 1 9 7 8 .

TABLE 1 HELICOPTER SIGNALS USED I N SUBJECTIV€Z RESPONSE STUDY S t at i c BBN A i r c r a f t TSC Operating Rep.

No. Event Condition Rate, Hz Type 214 S-61 ( r e f . ) 115 knot l e v e l 1 7 S-64 60 knot l e v e l 1 8 . 6 215 50 2 1 6 C H - 4 7 C 2 8 150 knot l e v e l C H - 4 7 C 1 8 60 knot l e v e l 217 25 2 1 8 2 1 2 1 0 5 knot l e v e l 11 2 1 2 6 1 knot l e v e l 11 220 6 O Approach 1 2

4 7 G 19

2 2 1 s-61 20 6 O Approach 1 7 2 2 2 206 L 4 6 6 O Approach 1 3

A s

dB 0 2 4 6

AC - dB

Figure 1 . - Comparison of judged difference in EPNL between impulsive and non-impulsive signals and calculated impulse adjustment using French measure CIA.

A s 4

A s 4

dB dB 0 0 12 14 16 18 12 14 16 18

Ac - d B Ac - dB

Figure 2.- Comparison of judged difference in EPNL between impulsiE and non-impulsive signals and calculated impulse adjustment using CFbe5.

A S dB A C - dB Figure 3 . - Correlation between judged and calculated adjustment to EPNL for BBN helicopters.

s

I so

- A s 4 -0 dB , 0 2 4 6

A C - dB

Figure 4 . - Correlation between judged and calculated adjustment to EPNL for French time-varying simulations.

0 BBN e 0 FRENCH

- IS0 -

e * s 4 dB n I 0 2 4 6

AC - dB

Figure 5.- C o r r e l a t i o n between judged and c a l c u l a t e d adjustment t o EPNL f o r BBN and French time-varying s i g n a l s .

I I

I 1

- I so 0 -

Figure 6 . - adjustment t o 0 BBN o FRENCH VARYING 0 FRENCH STEADY A S dB 0 2 4 6 A C - dB Figure 7.- Correlation between judged and c a l c u l a t e d adjustment t o EPNL f o r combined s i g n a l s .

2 1

RATING HELICOPTER NOISE John W. Leverton, B. J. Southwood, and A. C. Pike Westland Helicopters Limited SUMMARY I n t h e case of h e l i c o p t e r s t h e main problem from t h e community point of v i e w is t h e n o i s e heard on t h e approach. This i s p a r t i c u l a r l y t r u e i n t h e case of h e l i c o p t e r s with high levels of b l a d e s l a p and/or t a i l r o t o r noise. The EPNL concept does n o t appear p a r t i c u l a r l y w e l l s u i t e d on quantifying h e l i c o p t e r n o i s e s i n c e - i t is i n s e n s i t i v e t o t h e n o i s e Heard on approach some d i s t a n c e from t h e flyover position. Blade s l a p and t a i l r o t o r n o i s e b o t h need an a d d i t i o n a l correction. The former i s now r e a d i l y agreed and preliminary evidence i s pre- sented t o show t h a t a similar c o r r e c t i o n is required i n t h e case of t a i l r o t o r noise. The impact of t h e use of such c o r r e c t i o n s is examined and although they improve t h e c o r r e l a t i o n , w i t h t h e p r a c t i c a l s i t u a t i o n t h e r e is s t i l l considerable d i f f i c u l t y due t o t h e i n h e r e n t c h a r a c t e r i s t i c s of t h e EPNL procedure.

INTRODUCTION The use of h e l i c o p t e r s f o r general commercial purposes has continued t o i n c r e a s e and although i n many instances they operate from remote sites on spec- i a l i s t operations, t h e i r use w i t h i n b u i l t up areas has coctinued t o grow. A s a n a t u r a l consequence o v e r f l i g h t s over many areas f a r from h e l i p o r t sites have well-defined f l i g h t paths have showed a s i g n i f i c a n t increase. I n urban areas, been e s t a b l i s h e d f o r s a f e t y and ATC reasons and, as a r e s u l t , t h e r e has been a tendency t o concentrate f l i g h t s over p a r t i c u l a r l o c a t i o n s . This wider use of h e l i c o p t e r s has given rise t o concern over n o i s e and although t h e number of complaints are s t i l l s m a l l , they have tended t o i n c r e a s e over t h e last few years. The p o i s e levels of h e l i c o p t e r s have a l s o shown a tendency t o increase as modern technology has allowed high speed r o t o r s t o be employed and helicop- ters with h i g h e r forward f l i g h t speeds. Noise c e r t i f i c a t i o n f o r h e l i c o p t e r s has now been under consideration by t h e ICAO Working Group B f o r a number of There appears t o be a years and c e r t i f i c a t i o n 'proposals are expected s h o r t l y .

need, t h e r e f o r e , t o re-examine t h e whole t o p i c of e x t e r n a l h e l i c o p t e r n o i s e and (WHL) are t h e r e s u l t s of such a survey conducted by Westland Helicopters Ltd.

presented i n this paper. I n t h i s context i t is worth noting t h a t i n a d d i t i o n t o manufacturing h e l i c o p t e r s , WHL operates t h e London (Battersea) H e l i p o r t and h a s been involved i n helicopter/community i n t e r f a c e problems encountered by t h e B r i t i s h Army of t h e m i n e i n West Germany. The WHL a i r f i e l d is a l s o s i t u a t e d within t h e boundaries of a s m a l l town; thus, considerable experience h a s been gained from t h e operators, manufacturers, and community p o i n t of view. The author has a l s o been c l o s e l y involved, as an a d v i s e r t o t h e United Kingdom (UK) delegation, with t h e ICAO Working Group B.

HELICOPTER NOISE - THE REAL PROBLEM

I n t h e case of h e l i c o p t e r s t h e main n o i s e problem from t h e community point of view is t h e n o i s e heard on approach. This is p a r t i c u l a r l y t r u e i n t h e case of h e l i c o p t e r s with hegh levels of blade s l a p and/or t a i l r o t o r noise. It is t h e n o i s e heard at d i s t a n c e which is of main concern. I f t h i s n o i s e i s im- pulsive (blade s l a p ) o r contains a d i s t i n c t i v e whine ( t a i l r o t o r n o i s e ) , then it r e a d i l y attracts a t t e n t i o n , becomes d i s t u r b i n g , and, because it can b e heard f o r a r e l a t i v e l y long period as t h e h e l i c o p t e r approaches, gives rise t o complaints.

I n many cases, b u t n o t always, blade s l a p decreases r a p i d l y in l e v e l as t h e o v e r f l i g h t p o s i t i o n i s approached. S i m i l a r l y , on p r a c t i c a l l y a l l h e l i c o p t e r s with a high degree of t a i l r o t o r n o i s e , t h e n o i s e d i e s away w e l l b e f o r e t h e o v e r f l i g h t p o i n t is reached. This is due t o t h e d i r e c t i o n a l c h a r a c t e r i s t i c s of t h e s e two n o i s e sources.

The e f f e c t s discussed above are i l l u s t r a t e d i n f i g u r e s 1, 2 , and 3 which show t i m e h i s t o r i e s (amplitude - time traces). Figure 1 shows r e s u l t s f o r a two-bladed s i n g l e r o t o r h e l i c o p t e r with and without blade s l a p where t h e blade s l a p decreases j u s t p r i o r t o t h e overhead p o s i t i o n ; as can be seen, t h e r e is a marked d i f f e r e n c e i n t h e duration of t h e noise. A l s o i n d i c a t e d on t h i s f i g u r e is t h e blade s l a p region and t h e "peak levels" determined using a peak d e t e c t o r developed w i t h i n WHL. This, i n e f f e c t , gives a measure of t h e blade s l a p and a t r u e r e p r e s e n t a t i o n of t h e d u r a t i o n a s s o c i a t e d w i t h t h i s h e l i c o p t e r . The d a t a on t h i s f i g u r e , l i k e t h a t on f i g u r e s 2 and 3 , are unfortunately l i m i t e d i n t h e sense t h a t t h e recordings w e r e n o t taken over a s u f f i c i e n t t i m e period, ( t h e d a t a w e r e Collected as a p a r t of o t h e r s t u d i e s ) and, as a r e s u l t , t h e t r u e dura- The g e n e r a l implications are, t i o n above t h e background n o i s e cannot be shown.

however, clear. Figure 2 shows r e s u l t s again f o r a h e l i c o p t e r with blade s l a p b u t i n t h i s case t h e blade s l a p occurs during t h e complete f l i g h t . These d a t a w e r e obtained from a tandem r o t o r h e l i c o p t e r and are t y p i c a l of t h e n o i s e l e v e l s generated by t h i s type of h e l i c o p t e r . Indicated on t h e f i g u r e is t h e 'peak' l e v e l f o r t h e blade s l a p case and t h e dB(A) level f o r t h e same h e l i c o p t e r when flown (during s p e c i a l t e s t s ) with no blade s l a p . Figure 3 shows t h e corres- ponding r e s u l t s f o r a h e l i c o p t e r w i t h a high degree of t a i l r o t o r n o i s e compared with a h e l i c o p t e r which has a low l e v e l of t a i l r o t o r noise. A s can be seen, t h e duration is increased when t a i l r o t o r n o i s e is present.

From a review of test d a t a and a b r i e f review of complaints and observa- t i o n of h e l i c o p t e r n o i s e , it h a s been concluded t h a t although i n many cases t h e absolute l e v e l Qf h e l i c o p t e r n o i s e is h i g h e s t a t t h e o v e r f l i g h t p o s i t i o n , t h e annoying c h a r a c t e r i s t i c s have decreased and t h e r e i s l i t t l e n o t i c e taken of t h e maximum o v e r f l i g h t noise. This is n o t t o imply t h a t t h e complaints a g a i n s t h e l i c o p t e r n o i s e are completely independent of t h e l e v e l s i n c e obviously a ldw a l t i t u d e f l y o v e r which generates a very high level w i l l prompt an adverse re- action. I n t h e real environment h e l i c o p t e r s are t y p i c a l l y 500 f t (150 m) o r mor€ from t h e n e a r e s t residence and then it would appear t h a t t h e c h a r a c t e r of t h e n o i s e is e q u a l l y , o r more, important than t h e absolute l e v e l . There h a s , how- ever, been l i t t l e o r no t e c h n i c a l s t u d i e s i n t o t h e s e a s p e c t s , although experiencc gained from h e l i c o p t e r f l i g h t s over London and generally w i t h i n t h e UK supports these g e n e r a l observations.

According t o Greater London Council t h e r e is i n London "a small b u t steady flow of complaints about noise" ( r e f , 1). Y e t , when 230 occupiers of p r o p e r t i e s within t h e v f c i n i t y of the London Battersea Heliport w e r e contacted, only 3 ob- j e c t i o n s on n o i s e were received. I n f a c t , t h e r e appears t o b e more o b j e c t i o n s from l o c a t i o n s w e l l away from t h e h e l i p o r t ; t h i s r e s u l t agrees with t h e conclusion from t h e W H L review t h a t t h e main problems arise from the n o i s e generated on approach. I n t h i s context it is a l s o worth notlng t h a t i n a d d i t i o n t o t h e ab- s o l u t e level of h e l f c o p t e r n o i s e decaying r a p i d l y a f t e r t h e o v e r f l i g h t p o i n t is reached, n e i t h e r b l a d e s l a p n o r t a i l r o t o r n o i s e is t h e r e a f t e r s u b j e c t i v e l y d e t e c t ab le.

I n t h e p r a c t i c a l s i t u a t i o n , high n o i s e levels are generated during "bank turns", manoeuvres etc. p r l o r t o landing and take-off at h e l i p o r t s . These are obviously a f u n c t i o n of t h e s p e c i f i c f l i g h t procedures used o r t h e ATC con- s t r a i n t s and t h u s should n o t b e b c l u d e d i n any c e r t i f i c a t i o n scheme. They can, and do, however, have a major influence on t h e s u b j e c t i v e r e a c t i o n t o helicop- ter n o i s e and it would appear from t h e a v a i l a b l e evidence t h a t it is such aspects wlilich d e f i n e tlie a c c e p t a b i l i t y t o t h e general p u b l i c of a p a r t i c u l a r h e l i c o p t e r near a h e l i p o r t . The d e t a i l s of t h e f l i g h t path are important i n t h i s context and if they are chosen such t h a t a h e l i c o p t e r has t o t u r n sharply t o avoid over- f l y i n g a p a r t i c u l a r l o c a t i o n , t h i s can o f t e n generate higher n o i s e l e v e l s than w a s allowed t o f l y overhead.

would occur i f t h e h e l i c o p t e r RATING HELICOPTER NOISE It follows from t h e p o i n t s o u t l i n e d previously t h a t s i n c e t h e annoyance of a h e l i c o p t e r is l a r g e l y dependent on t h e n o i s e heard on approach t h a t any r a t i n g o r c e r t i f i c a t i o n scheme f o r h e l i c o p t e r s should b e completely d i f f e r e n t from t h a t derived f o r f i x e d wing (CTOL) a i r c r a f t . This should t a k e i n t o account t h e s u b j e c t i v e c h a r a c t e r of t h e h e l i c o p t e r n o i s e on approach some d i s t a n c e from t h e f l y o v e r p o i n t a d t h e t i m e a s s o c i a t e d with t h e n o i s e , as w e l l as t h e maxi- m u m l e v e l measured during t h e o v e r f l i g h t . It is o f t e n argued t h a t it is not t h e r o l e of c e r t i f i c a t i o n t o c o n t r o l t h e "operational n o i s e s i t u a t i o n " , b u t blade s l a p n o i s e and t a i l r o t o r n o i s e are of fundamental importance i n t h e case of t h e h e l i c o p t e r , p a r t i c u l a r l y s i n c e t h e s e n o i s e sources can occur during is a very d i f f e r e n t s i t u a t i o n from " s t r a i g h t and l e v e l " c r u i s e f l i g h t . This t h e tra- t h a t a s s o c i a t e d with f i x e d wing a i r c r a f t and t h e r e f o r e , by i m p l i c a t i o n , It is d i t i o n a l m e t h o d of r a t i n g a i r c r a f t n o i s e is b a s i c a l l y inappropriate.

d i f f i c u l t , however, t o kmagine how a "new scheme" f o r h e l i c o p t e r s could be formulated p a r t i c u l a r l y s i n c e t h e v a r i o u s c e r t i f i c e t i o n a u t h o r i t i e s p l a c e signi- f i c a n t emphasis on developing a scheme which is, as f a r as p o s s i b l e , compatible with ' f i x e d wing' procedures. A s a consequence, t h e h e l i c o p t e r n o i s e certifica- t i o n concept c u r r e n t l y being considered w i t h t h e I n t e r n a t i o n a l C i v i l Aviation Organization (ICAO) and by t h e Federal Aviation Administration (FAA) are based on t h e EPNL method which is dependent mainly on t h e a b s o l u t e l e v e l of t h e n o i s e and t h e d u r a t i o n between t h e '10 dB down' p o i n t s .

The EPNL method a l s o t a k e s i n t o consideration t h e t o n a l content of t h e t h e s u b j e c t i v e noise and thus it would b e expected t h a t it would t a k e account of impact of t a i l r o t o r n o i s e as w e l l as t h e high frequency engine "whine.' Pre- liminary s t u d i e s witf-,in W H L suggested t h a t t h i s is not t h e case and t h a t t h e 'tone c o r r e c t i o n s ' are o f t e n , t o a f i r s t o r d e r , independent of t h e l e v e l of t a i l r o t o r n o i s e c ThAs is a complex s u b j e c t and t h e r e has been very l i t t l e work on t h i s t o p i c t o d a t e , b u t ft would appear from a n a l y s i s made by t h e 'author that problems arise from the f act t h a t h e l i c o p t e r n o i s e , p a r t f c u l a r l y below 500 Hz, c m t a f n s many d i s c r e t e frequencies from t h e main and t a i l r o t o r and t h e one-thlrd octave band s p e c t r a is n o t a t r u e r e f l e c t i o n of t h e annoying charac- teristics of t h e t a i l r o t o r noise. To remove t h e u n c e r t a i n t i e s of applying t h e tone c o r r e c t i o n procedure, ICAO Working Group B suggested at one s t a g e t h a t tone c o r r e c t i o n s should be applied only above 500 H z , b u t r e c e n t l y t h i s approach has been dropped and t h e latest view appears t o b e t h a t tone c o r r e c t i o n s should b e applied over t h e complete frequency range as i n t h e standard EPNL procedure (Annex 16).

A s mentioned previously;, b l a d e s l a p i s t h e most annoying source associated with t h e h e l i c o p t e r . There is ample evidence t o show t h e inadequacies of t h e standard EPNL procedure and t h a t an a d d i t i o n a l c o r r e c t i o n factor is required t o account f o r t h i s source. There is, however, some opposition t o t h i s procedure and, although t h e I n t e r n a t i o n a l Organization f o r Standardization (ISO) has pro- posed a procedure f o r accounting f o r , b l a d e s l a p , t h i s has n o t received wide acceptance i n t h e U . S . A . I n f a c t , it would appear t h a t even some groups, which have accepted i n p r i n c i p l e t h a t a b l a d e s l a p c o r r e c t i o n i s necessary, are more concerned w i t h developing a l t e r n a t i v e schemes r a t h e r than assessing t h e r e l a t i v e merits of those already proposed.

t h e use of t h e EPNL concept f o r r a t i n g The concern over t h e s u i t a b i l i t y of h e l i c o p t e r n o i s e w i l l still apply even i f a blade s l a p c o r r e c t i o n i s f i n a l l y accepted. The adoption of such a c o r r e c t i o n w i l l , however, s i g n i f i c a n t l y improve t h e r a t i n g of h e l i c o p t e r n o i s e r e l a t i v e t o t h e use of an unmodified EPNL .

RATING UNITS It is n o t proposed i n t h i s paper t o review t h e various u n i t s f o r r a t i n g h e l i c o p t e r n o i s e s i n c e it is clear t h a t f o r a number of reasons t h e perceived n o i s e l e v e l (PNL) w i l l be used as t h e b a s i c u n i t with possibly t h e dB(A) i n some s i t u a t i o n s . I n t u i t i v e l y , it would be expected t h a t t h e PNL method should adequately account f o r h e l i c o p t e r n o i s e providing t h e s j g n a l is r e l a t i v e l y broadband i n n a t u r e and without any pronounced d i s c r e t e n o i s e sources. Also, s i n c e it is used f o r r a t i n g CTOL a i r c r a f t , it should a l s o be equally a p p l i c a b l e f o r r a t i n g h e l i c o p t e r engine noise. It follows, t h e r e f o r e , t h a t with t h e ex- ception of t h e cases when t h e s i g n a l is dominated by blade s l a p and/or t a i l r o t o r n o i s e , t h e PNL u n i t , and by implication t h e dB(A), should b e s u i t a b l e f o r r a t i n g h e l i c o p t e r noise. For b l a d e s l a p and t a i l r o t o r n o i s e , some a d d i t i o n a l c o r r e c t i o n terms are required and t h i s is discussed i n t h e following s e c t i o n s .

BLADE SLAP Correcting f o r Blade Slap It is clear from t h e s t u d i e s conducted w i t h i n t h e UK by t h e National Phy- sical Laboratory (NPL) and W H L ( r e f s . 2 and 3 ) , by Aerospatiale i n France ( r e f . 4 ) and from some of t h e work conducted i n t h e States ( r e f s . 5 and 6 ) t h a t a s u b j e c t i v e c o r r e c t f o n is required t o account f o r b l a d e s l a p . There i s very l i t t l e disagreement between t h e r e s u l t s and t h e r e is a g e n e r a l consensus t h a t t h e penalty a s s o c i a t e d with severe blade s l a p is 6 dB. This is i l l u s t r a t e d i n f i g u r e 4 which shows t h e r e s u l t s of psychoacoustic tests conducted w i t h i n W H L ( r e f . 7 ) . The f i g u r e shows t h e Subjective c o r r e c t i o n i n terms of dB(A) a g a i n s t a measure of t h e impulsive n a t u r e of t h e s i g n a l based on t h e crest f a c t o r developed by WHL some y e a r s ago ( r e f s . 3 and 7). The s u b j e c t i v e r a t i n g of blade s l a p , from none t o severe, is i n d i c a t e d on t h e f i g u r e , t o g e t h e r w i t h a proposed correcti'on c u w e . For a l l p r a c t i c a l purposes, t h e r e s u l t s would b e i d e n t i c a l t h e c o r r e c t i o n s had been determined i n terms of PNL values.

if, i n s t e a d of dB(A), Since t h e analogue crest f a c t o r method developed by W H L was proposed, NPL and Aerospatiale have developed impulsive n o i s e d e s c r i p t o r s which are based on Recently, t h e r a t i n g of blade s l a p has been d i g i t a l a n a l y s i s of t h e s i g n a l s .

reviewed by t h e I n t e r n a t i o n a l Standards Organization (ISO) and they have recommended t h e u s e of a method based on t h a t o r i g i n a l l y devised by NPL (ref. 8 ) .

The NPL and Aerospatiale C I d e s c r i p t o r s are compared i n general terms on f i g u r e 5 which shows t h e r e s u l t s f o r a 250-Hz s i n e wave pulse as a f u n c t i o n of r e p e t i t i o n rate. Also i n d i c a t e d on t h i s f i g u r e is t h e t r u e crest f a c t o r and, as can b e seen, providing t h e i n t e g r a t i o n t i m e employed i n t h e NPL method is small, then f o r a l l p r a c t i c a l purposes t h e r e s u l t s of t h e NPL and t h e C I desc- r i p t o r follows c l o s e l y those given by t h e crest f a c t o r . Thus, from a fundamental point of view, t h e r e is l i t t l e t o choose between t h e methods (assuming t h e i n t e - t h e s i g n a l as orig- g r a t i o n i s 0.2 ms) and determination of t h e crest f a c t o r of chose an i n a l l y proposed by wHLi The ISO, when it adopted t h e NPL d e s c r i p t o r , i n t e g r a t i o n t i m e of 0.2 m s which is i d e n t i c a l t o t h e value used i n t h e W H L ana- logue peak d e t e c t o r ( r e f . 9 ) .

merits of t h e s e It is n o t proposed i n t h i s paper t o discuss t h e r e l a t i v e methods b u t , s i n c e i n p r a c t i c a l terms they r e s u l t i n t h e same o r d e r of correc- t i o n , t o review implications which r e s u l t from t h e i r use. There is, however, one exception t o t h i s g e n e r a l t r e n d i n t h a t t h e method proposed by Galloway contains a crest f a c t o r and r e p e t i t i o n rate term and can give very ( r e f . 6) d i f f e r e n t r e s u l t s . F i r s t l y , t h e crest f a c t o r , by d e f i n i t i o n , already contains a r e p e t i t i o n rate term and seconjily, i f t h i s procedure is adopted, then h e l i - w i l l have a high copters w i t h a high b l a d e passing frequency ( r e p e t i t i o n r a t e ) s u b j e c t i v e penalty o r c o r r e c t i o n even i f t h e impulsive content of t h e s i g n a l is s m a l l ( r e f . 10). This r e s u l t i s opposite t o what occurs i n practice, where t h e r e is a marked tendency f o r t h e s e v e r i t y of blade s l a p t o decrease with an increased.

i n c r e a s e i n t h e number of blades; hence, t h e blade passing frequency is There is a number of o t h e r d i f f i c u l t i e s a s s o c i a t e d w i t h t h e u s e of t h i s method and thus it would n o t seem t o b e a s u i t a b l e d e s c r i p t o r for blade s l a p .

The blade s l a p c o r r e c t i o n w a s o r i g i n a l l y based on steady state (hover) recordings, although r e c e n t l y Aerospatiale ( r e f . 11) and Galloway ( r e f . 6 ) have conducted physcoacoustic tests u s h g f l y o v e r recordings. These have shown t o a first order t h a t t h e steady state and flyover tests g i v e f o r a l l p r a c t i c a l pur- poses i d e n t i c a l r e s u l t s . The f l y o v e r s used appear, however, i n general t o be of s h o r t e r d u r a t i o n than s i g n a l s commonly encountered i n t h e real environment.

Even so, P t h a s been argued t h a t t h e r e s u l t s are equally applicable t o hover and flyover signals.

The b l a d e s l a p c o r r e c t i o n f a c t o r , as c u r r e n t l y proposed by I S O , i s added t o t h e PNLT t i m e h i s t o r y , i n 8 s i m i l a r manner t o t h e tone c o r r e c t i o n , every h a l f second t o give t h e PNLT(1) t i m e h i s t o r y from which t h e EPNL(1) is c a l c u l a t e d The d i f f e r e n c e between t h e EPNL(1) and t h e standard using t h e normal approach.

t h e h e l i c o p t e r n o i s e o r i n EPNL is a measure of t h e o v e r a l l impulsiveness of o t h e r words t h e magnitude of t h e annoyance associated w i t h t h e blade s l a p .

'Use of t h e Blade Slap Correction A number of s t u d i e s have been conducted w i t h i n W H L on t h e effect of using t h e various proposed blade s l a p c o r r e c t i o n procedures. These have been based on t h e NPL, t h e Aerospatiale C I and/or t h e WHL analogue crest f a c t o r d e t e c t o r methods s i n c e , as mentioned previously, t h e r e s u l t s are e s s e n t i a l l y independent of t h e method used. R e a l h e l i c o p t e r and simulated h e l i c o p t e r s i g n a l s have been t h e time a t which t h e blade s l a p d i e s away, r e l a t i v e t o used and t h e impact of t h e t i m e of t h e m a x i m u m n o i s e level, e s t a b l i s h e d . The g e n e r a l trends are ill- u s t r a t e d i n f i g u r e s 6 and 7. Two cases are shown on f i g u r e 6 which gives re- s u l t s of a t h e o r e t i c a l a n a l y s i s . I n t h e first t h e maximum c o r r e c t i o n of 6 dB, which corresponds t o severe b l a d e s l a p , is assumed t o apply over t h e b l a d e s l a p range w h i l s t i n t h e o t h e r a lower i n t e n s i t y b l a d e s l a p with a c o r r e c t i o n of 4 dB w a s considered. A s w i l l be observed, t h e impact, as expected, of applying t h e t h e t i m e from overhead p o s i t i o n is increased. Figure c o r r e c t i o n is decreased as 7 shows sfmilar r e s u l t s f o r a simulated f l y o v e r w i t h blade s l a p and r e s u l t s of real h e l i c o p t e r a n a l y s i s .

It is d i f f i c u l t from t h i s a n a l y s i s t o draw s p e c i f i c conclusions, although, if blade s l a p occurs during t h e complete f l i g h t , t h e c o r r e c t i o n is more l i k e l y t o be i n t h e o r d e r of 5.5 PNdB r a t h e r than t h e t h e o r e t i c a l 6 dB. From a review of a wide range of h e l i c o p t e r s , it would appear t h a t f o r t h e h e l i c o p t e r with severe blade s l a p on approach which "dies away" 2 o r 3 seconds p r i o r t o t h e overhead p o s i t i o n t h a t t h e c o r r e c t i o n w i l l b e i n t h e o r d e r of 4.5 EPNdB. Another d i f f i c u l t y a s s o c i a t e d with t h e b l a d e s l a p d e s c r i p t o r s c u r r e n t l y being proposed by IS0 is t h a t nonimpulsive h e l i c o p t e r s g i v e rise t o a c o r r e c t i o n i n terms of EPNL of 1 t o 2 EPNdB. This r e s u l t s from t h e f a c t t h a t although s u b j e c t i v e l y they are more impulsive than t h e broadband white n o i s e they are n o t impulsive, a reference i n determination of t h e impulsive d e s c r i p t o r . The s i g n a l used as p r a c t i c a l e f f e c t of t h i s is t h a t t h e d i f f e r e n c e between t h e impulsive and non- impulsive h e l i c o p t e r is f u r t h e r decreased.

Based on experience obtained within W H L from evaluating p u b l i c r e a c t i o n t o h e l i c o p t e r n o i s e , it seems f a i r t o conclude, t h e r e f o r e , t h a t t h e EPNL(1) 4 2 4 concept as c u r r e n t l y being proposed g t g l l underest-tes, i n r e l a t i o n t o a non- slapping h e l i c o p t e r , the impact o f h e l i c o p t e r b l a d e slap. This i s not meant t o imply t h a t the c o r r e c t i o n procedure is h a p p r o p r i a t e b u t r a t h e r t h a t b a s i c EPNL concept is inadequate, TAIL ROTOR NOISE C h a r a c t e r i s t i c s T a i l r o t o r n o i s e , l2ke blade s l a p , shows up on a n a l y s i s as a serees of p u l s e s spaced a t t h e b l a d e passing i n t e r v a l . Thus, except f o r t h e d i f f e r e n c e s i n p u l s e frequency and b l a d e passing frequency, t h e s i g n a l s , which s u b j e c t i v e l y are c l a s s e d "whine", are very s i m i l a r t o those associated with b l a d e s l a p . This i s i l l u s t r a t e d diagrammatically i n f i g u r e 8 which shows r e p r e s e n t a t i v e blade s l a p and t a i l r o t o r n o i s e p u l s e chains. Thus, from a r a t i n g p o i n t of v i e w , t a i l r o t o r n o i s e is impulsive i n c h a r a c t e r and hence "rated" by t h e v a r i o u s blade s l a p d e s c r i p t o r s discussed prevlously. This is a very important p o i n t which is over- looked by many i n v e s t i g a t o r s who simply associated t a i l r o t o r n o i s e with a dis- crete frequency spectrum.

The Tone Correction Procedure T a i l r o t o r n o i s e , as discussed previously, is very dominant on approach, but unlike blade s l a p , it d i e s away w e l l before t h e overhead p o s i t i o n , as i l l u s - When t a i l r o t o r n o i s e is pronounced, then it can be detec- t r a t e d i n f i g u r e 3.

t e d on a one-third octave band p l o t ; a t y p i c a l r e s u l t is shown on f i g u r e 9. The E1NL procedure is, however, r e l a t i v e l y i n s e n s i t i v e t o such tones as can be seen I f t h e f l y o v e r s shown i n f i g u r e 3 are exami- from r e s u l t s i n d i c a t e d on t h e f i g u r e .

ned, then it can b e shawn.that t h e EPNL f o r t h e Scout with t h e high l e v e l of t a i l r o t o r n o i s e is only 2 EPNdB higher than t h a t f o r t h e Wessex. P a r t of t h i s is due t o t h e d i f f e r e n c e i n t h e duration c o r r e c t i o n a r i s i n g from t h e s l i g h t l y d i f f e r e n t f l i g h t speeds between t h e two h e l i c o p t e r s . I f t h i s is taken i n t o consideration, then t h e c a l c u l a t e d d i f f e r e n c e is less than 1 EPNdB. I n an t h e equivalent continuous n o i s e l e v e l attempt t o h i g h l i g h t t h i s problem f u r t h e r , (Leq) has been c a l c u l a t e d f o r a Scout f l y o v e r when t h e t a i l r o t o r n o i s e is very pronounced and compared w i t h t h e p r e d i c t i o n of t h e Leq f o r an equivalent f l i g h t with no t a i l r o t o r noise. This is i l l u s t r a t e d i n f i g u r e 10 and it w i l l be noted t h a t t h e t a i l r o t o r 'hump' is w i t h i n 3 dB(A) of t h e maximum dB(A) level.

The d i f f e r e n c e between' t h e two Leq v a l u e s , based on t h e levels w i t h i n t h e

region covered by t h e 'maximum - 25 dB(A), is 1.9 dB(A) - y e t obviously t h e two

conditions sound very d i f f e r e n t , When analysing f l y o v e r s i g n a l s , it has a l s o been observed t h a t tone c o r r e c t i o n s r e s u l t i n a constant d i f f e r e n c e between t h e PNLT and PNL values, being t y p i c a l l y 1.5 dB. Owing t o t h e v a r i a b i l i t y of t h e one-third octave band s p e c t r a , t h e band which is responsible f o r t h e c o r r e c t i o n does not appear t a be represen- is considered t o b e due t o t h e complex n a t u r e tative of t h e real s i t u a t i o n . This of h e l i c o p t e r noise.

It is concluded t h e r e f o r e from the a n a l y s i s o u t l i n e d above and d e t a i l e d reviews of a wide range of f l i g h t conditions t h a t tone correctfons i n t h e EPNL procedure b e a r l i t t l e r e l a t i o n t o t h e t r u e annoyance of h e l i c o p t e r t a i l r o t o r noise.

Rating by Impulsive Noise Descriptors Since t a i l r o t o r n s i s e takes t h e same form as blade s l a p , then any impulsive d e s c r i p t o r w i l l be e q u a l l y s e n s i t i v e t o both of these sources of n o i s e and, of course, t o any o t h e r impulsive sources of noise. This can be seen on f i g u r e 5 which, i n a d d i t i o n t o v a l u e s f o r i d e a l i s e d blade s l a p pulses (low r e p e t i t i o n tail r o t o r n o i s e . A s discussed previously, r a t e s ) shows r e s u l t s f o r i d e a l i s e d values are shown f o r t h e NPL (ISO) and Aerospatiale C I d e s c r i p t o r s and t h e crest f a c t o r . The main d i f f e r e n c e between t h e two sets of r e s u l t s is t h a t whereas t h e NPL (ISO) method gives a measure of blade s l a p when an i n t e g r a t i o n This e f f e c t time of 10 ms is used, it e f f e c t i v e l y rejects t a i l r o t o r noise.

can be b e t t e r appreciated from t h e p l o t f o r i d e a l i s e d s i g n a l s shown i n f i g u r e 11.

It may appear from t h e s e p l o t s t h a t i f a 10 m s i n t e g r a t i o n t i m e w a s used, then t a i l r o t o r n o i s e would be r e j e c t e d and t h e measured value i n p r a c t f c e would depend s o l e l y on t h e l e v e l of t h e impulsive (blade s l a p ) noise. There are, however, a number of major objections t o t h i s . F i r s t l y , t h e r e l a t i o n s h i p between t h e NPL d e s c r i p t o r (10 log I) and t h e crest f a c t o r v a r i e s with r e p e t i - t i o n rate .- i n o t h e r words, on f i g u r e 5, t h e values are not p a r a l l e l t o those is t h e f a c t t h a t use of such of t h e crest f a c t o r . More importantly, however, a method would give a r e s u l t independent of t h e pulse frequency, and hence c r e s t f a c t o r , as i n d i c a t e d i n f i g u r e 12. Thus, such a s o l u t i o n is n o t p r a c t i - c a l .

I n t h e method o r i g i n a l l y proposed by WHL f o r r a t i n g blade s l a p , which w a s based on t h e use of t h e crest f a c t o r , t h i s problem w a s overcome by passing t h e s i g n a l through a band p a s s f i l t e r centered on 250 Hz i n order t h a t a l l impul- s i v e s i g n a l s except those a s s o c i a t e d with blade s l a p were r e j e c t e d ( r e f . 3).

Such a method is used w i t h i n W H L f o r assessing t h e magnitude of blade s l a p , b u t o b j e c t i o n s w e r e r a i s e d a g a i n s t t h i s method on t h e grounds t h a t s i n c e the widest standard f i l t e r which could be used w a s an octave band (177 t o 354 Hz), some h e l i c o p t e r s could generate blade s l a p with t h e main energy above t h e upper A n example o f t e n quoted is t h e Bolkow BO 105 which appears t o frequency l i m i t .

have t h e blade s l a p energy m a x i m u m centered on 600 Hz ( r e f . 12), while on most it is around 250 t o 300 Hz ( r e f . 13). It is considered t h a t o t h e r h e l i c o p t e r s t h e W H L proposal could possibly be f u r t h e r developed t o take account of such cases and it i s 'questionable whether a s i g n a l w i t h a r e p e t i t i o n rate, as is t h e case of t h e BO 105, of 28 Hz and "pulse frequency" of 600 Hz w i l l subjective11 sound t h e same as t h e blade s l a p generated by o t h e r h e l i c o p t e r s . It h a s a l s o n o t y e t been shown whether t h e standard PNL/dB(A) method f a i l s t o penalize such blade s l a p i n a similar manner t o t h a t found f o r Glade s l a p of t h e type used i n W H L h a s , however, not pursued t h i s method t h e s t u d i e s summarezed f n f i g u r e 4.

r e c e n t l y s i n c e c u r r e n t proposals by IS0 provide an adequate d e s c r i p t o r f o r b l a d e s l a p and t h e r e is an h t u i t i v e f e e l i n g - r e c e n t l y confirmed by preliminary

tests - t h a t t h e IS0 method could be used t o account for both t a i l

s u b j e c t i v e r o t o r n o i s e and b l a d e s l a p .

I f f i g u r e s 5 and I1 are examined i n d e t a i l , f t w k l l be observed t h a t t h e crest f a c t o r assocfated with t a i l r o t o r n o i s e is less than t h a t f o r blade s l a p .

This is a genuine e f f e c t and agrees w e l l with t h e real p r a c t i c a l s i t u a t i o n i n t h a t severe blade s l a p is always more pronounced and a n n o y h g than t h e c o k e s - ponding high l e v e l of t a i l r o t o r noise.

Subjective Evaluatfm Preliminary p sckoacoustlc tests have been conducted w i t h i n WHL using simulated and real h e l i c o p t e r steady state (hover) recordfngs ( r e f . 14). The This r e s u l t s obtained t o d a t e are shown i n terms of dB(A) values i n f f g u r e 13.

f i g u r e is d i r e c t l y comparable t o t h e b l a d e s l a p r e s u l t s shown i n f i g u r e 4 an?, as w i l l be noted, it suggests t h a t pronounced tail r o t o r n o i s e r e q u i r e s an addi- t i o n a l c o r r e c t i o n of 4 dB(A). It follows t h a t , t o a f i r s t o r d e r , t h i s is s i m - i l a r t o blade s l a p and t h e r e f o r e , by taking i n t o account t h a t t a i l r o t o r n o i s e g i v e s sli-ghtly lower crest f a c t o r than a s s o c i a t e d with blade s l a p , it can be argued t h a t an kmpulsive d e s c r l p t o r of t h e type proposed by IS0 can adequately I f t h i s approach w a s adopted, then account f o r both sources of impulsive noise.

obviously t h e conventfonal tone c o r r e c t i o n s would n o t b e required. The impul- s i v e r a t i n g procedures, as c u r r e n t l y envisaged, have a ' c u t o f f ' a t around 2 kHz and t h e l e v e l of t a i l r o t o r n o i s e is very low above 1 kHz. It would seemsappro- p r i a t e , t h e r e f o r e , t o l i m i t t h e tone c o r r e c t i o n procedure i n t h e present EPNL procedure t o , say, 1 kHz and above. This would provide, assuming an. impulsive impulsive n o i s e , n o i s e d e s c r i p t o r w a s used, a good measure of both sources of while ensuring t h a t high frequency d i s e r e t e s tones from t h e engine etc. were adequately covered and t h a t t a i l r o t o r n o i s e was n o t penalized twice.

THE REQUIREMENTS FOR NOISE STANDARDS t h e UK it would seem reasonable t o assume t h a t From t h e experience within t h e a i m of ' n o i s e c e r t i f i c a t i o n ' should b e , i n general, t o contain t h e c u r r e n t s i t u a t i o n s i n c e , u n l i k e CTOL a i r c r a f t , h e l i c o p t e r s do not cause any major n o i s e disturbance. There are, of course, a number of noisy h e l i c o p t e r s which are ' c e r t i f i c a t i o n ' lirrtits should be such exceptions t o t h i s g e n e r a l r u l e and hence as t o p r o h i b i t t h e i r development and ensure t h a t they are phased out of c i v i l use. Even so, t h e r e does n o t appear t o be any case f o r s e t t i n g standards which would r e q u i r e a dramatic reduction i n t h e n o i s e generated by t h e majority of h e l i c o p t e r s .

The h e l i c o p t e r s i n t h e s u b j e c t i v e l y 'noisy' category are u s u a l l y those which generate high levels of 'blade s l a p ' (impulsive main r o t o r n o i s e ) o r t o of t a i l r o t o r noise. It has been estab- a lesser e x t e n t t h o s e w i t h a high l e v e l l i s h e d t h a t t h e standard dBA and PNL (PNdB) u n i t s do n o t adequately account The p o s i t i o n r e l a t i n g t o t h e f o r blade s l a p and ' c o r r e c t i o n s ' are required.

s u b j e c t i v e impact of t a i l r o t o r n o i s e has n o t been s t u d i e d i n such depth, b u t it appears from t h e a v a i l a b l e evidence t h a t t h e standard procedures do n o t f u l l y quantify t h i s source even i f t h e 'tone c o r r e c t i o n procedure' i n t h e EPNL procedure is applied. It follows, t h e r e f o r e , t h a t it is completely f a l s e t o use a u n i t which does n o t t a k e t h e s e a s p e c t s f n t o account when s e t t i n g t h e a p p r o p r i a t e l i m i t .

Noise c e r t i f i c a t i o n is an important i s s u e and it is v i t a l t h a t care should be taken i n b o t h t h e s e l e c t i o n of t h e f l i g h t test conditions and t h e r a t h g u n i t s i n c e c e r t i f i c a t f o n w i l l Fiave a long term e f f e c t on t h e h e l i c o p t e r indus- It seems e s s e n t i a l t h a t a method and a r a t i n g u n i t t r y and t h e community.

which takes i n t o account t h e s u b j e c t i v e impact are derfved o r otherwise p u b l i c inadequate and then r e a c t i o n w i l l be based on t h e f e e l i n g t h a t c e r t i f i c a t i o n is numerous l o c a l rules w i l l be applfed. mese could be more severe than t h e cer- t i f i c a t i o n requirements and t h e i n d u s t r y would b e burdened by t h e need t o m e e t c o n f l i c t i n g requirements, It is, t h e r e f o r e , considered t h a t t h e c e r t i f i c a t i o n scheme should be based on a f u l l t e c h n i c a l evaluation of a l l t h e i s s u e s invol- ved including those o u t l f n e d i n t h i s paper.

CONCLUDING REMARIG Helicopter n o i s e is n o t t h e major problem o f t e n suggested, although ob- viously t h e r e are a number of nofsy h e l i c o p t e r s which give rise t o complaints.

and possibly lower Thus, it would appear t h a t although t h e r e is a need t o l i m i t , s l i g h t l y , t h e n o i s e l e v e l s a s s o c i a t e d with c u r r e n t h e l i c o p t e r s , t h e r e is l i t t l e j u s t i f i c a t i o n i n attempting t o o b t a i n a dramatic reduction. This is p a r t i c u - l a r l y t r u e i f t h e economic p e n a l t i e s involved i n o b t a i n n o i s e reductions on r e a l environment are h e l i c o p t e r s and t h e l i k e l y gains i n r e a l n o i s e terms i n taken i n t o account. Furthermore, it would a p p e a r from t h e experience of W H L t h a t many of t h e problems t h a t occur i n p r a c t i c e are a s s o c i a t e d almost e n t i r e l y with blade s l a p and/or high l e v e l s of t a i l r o t o r n o i s e which occur on approach during c r u i s e f l i g h t . I n simple terns t h e r e is ample evidence t o suggest t h a t a blade s l a p c o r r e c t i o n is required and t h a t when t h e blade s l a p is severe, t h i s should be 6 dB. A s i m i l a r s i t u a t i o n occurs i n t h e case of t h e t a i l r o t o r n o i s e and although preliminary r e s u l t s i n d i c a t e t h i s can be tackled i n a similar manner t o blade s l a p , it i s recognized t h a t it w i l l b e some t i m e b e f o r e such a correc- t i o n i s accepted. The main problem appears t o be, however, t h a t i f t h e s e cor- then s i n c e t h e r a t i n g r e c t i o n s a r e taken i n t o account on a "half second" b a s i s , methods are being based on t h e EPNL concept, they w i l l n o t account f o r t h e There is a l s o t h e p o s s i b i l i t y t h a t annoyance caused by h e l i c o p t e r s i n p r a c t i c e .

even i f c e r t i f i c a t i o n r e s u l t s i n a reduction of t h e m a x i m u m n o i s e emitted by a h e l i c o p t e r during f l y o v e r , t h e c h a r a c t e r ( c r e s t f a c t o r ) of t h e n o i s e generated on approach due t o blade s l a p and/or a high l e v e l of t a i l r o t o r n o i s e w i l l re- main t h e same. Thus, it is p o s s i b l e t h a t t h e annoyance t o t h e p u b l i c w i l l , f o r a l l p r a c t i c a l purposes, b e t h e same even though t h e a b s o l u t e l e v e l is reduced.

Added t o t h i s is t h e f a c t t h a t n e a r a h e l i p o r t t r a n s i e n t manoeuvre n o i s e (bank t u r n s , e t c . ) may still occur and s i n c e t o some e x t e n t t h e s e are independent of t h e maximum n o i s e generated by t h e h e l i c o p t e r , t h e impact of n o i s e c e r t i f i c a t i o n i n . t h i s case may again be s m a l l . This p a r t i c u l a r aspect can, however, b e con- t r o l l e d i n p r a c t i c e by l o c a l h e l i p o r t o r ATC r u l e s .

A review has been made of p o s s i b l e a l t e r n a t i v e methods of r a t i n g h e l i - copter nokse and t o d a t e W H L f r e e l y admits t h a t it has n o t y e t been a b l e t o devise a completely acceptable scheme. Aspects considered have included, i n t h e EPNL procedure, changing t h e c a l c u l a t i n g t h e t o t h e !maximum l e v e l - 20 dB' a correc- b u t this does n o t give any marked improvement and a method i n t i o n based on t h e crest f a c t o r of t h e signal 5n the f a r f i e l d d i t i o n ) is added t o t h e computed PNLM o r EPNL value. U s e of a l s o been evaluated, t o g e t h e r w i t h a number of 'ad hoc' allowances have been taken i n t o account f o r blade s l Since t h e r e is obviously d i f f i c u l t y at t h e p r e s e n t t l m e w5th t h e o v e r a l l r a t i n g procedure t o be used, t h i s again gives support t o t h e view t h a t c e r t i f i - c a t i o n should e s s e n t i a l l y p r w i d e a scheme which l i m i t s t h e use and develop- ment of very n o j s y "slapping" type h e l i c o p t e r s o r those with high l e v e l s of t a i l r o t o r n o i s e , r a t h e r than attempt an o v e r a l l reduction of h e l i c o p t e r noise.

I n t h i s context it is a l s o worth n o t i n g t h a t t h e p r e d i c t i o n of ' t o t a l h e l i - c o p t e r ' n o i s e is r e l a t 3 v e l y i n a c c u r a t e and less p r e c i s e than commonly a s s o c i a t e d w i t h f i x e d wing a i r c r a f t . This is understandable since( t h e research e f f o r t both i n terms of manpower and f i n a n c i a l resources has been s l g n i f i c a n t l y less than i n t h e case of CTOL a i r c r a f t . The main sources of r o t o r n o i s e are, how- e v e r , r e l a t f v e l y w e l l understood and it is t h e i n t e r a c t i o n e f f e c t s which cause problems during p r e d i c t i o n s . It a l s o implies t h a t t h e configuration (layout) of t h e h e l i c o p t e r has a s i g n i f i c a n t impact i n t h e r e s u l t i n g o v e r a l l noise. It follows, t h e r e f o r e , t h a t it is n o t p o s s i b l e t o design, w i t h i n t h e required accuracy, t o a s p e c i f i c l e v e l .

1, Greater London Council: Report (11.6.75) by t h e C o n t r o l l e r of Planning and Transportation. I t e m 11 P847.

F u l l e r H.C.: Rating Helicopter Noise - A Study of Subjective Reaction t o

2.

National P h y s i c a l Laboratory Report, November 1976.

Impulsive Sound.

3. Southwood B . J . , P i k e , A.C.: The Rating and Subjective Assessment of H e l i - copter Blade Slap. Westland Helicopters Ltd. (WHL) Applied Acoustics Department Note 1147, May 1976.

4. Damongest A: Gene Produit p a r les B r u i t s de Hature Impulsive. Aerospatiale November 1976.

Report H/DE - ER 351-67, Noise C e r t i f i c a t i o n Cor.siderations f o r Helicopters Based on Laboratory 5.

J u l y 1976.

I n v e s t i g a t i o n s - Man Acoustics Report.FAA-RD-76-116, Subjective Response t o Simulated and Actual Helicopter Noise.

6. Galloway W . J . : BBN Report 3573, December 1977.

7. Leverton J . W . , Southwood B . J . - A Correction f o r Helicopter Blade Slap.

W H L Applied Acoustics Department Note 1163, November 1976, 8. F i r s t D r a f t Proposal f o r Amendment t o IS0 3891 "Acoustics F Procedure f o r Describing Aircraft Noise Heard on t h e Ground!' --Measurement of Noise from Helicopters, ISO/TC 43/SC 1 ( S e c r e t a r i a t - 254) 356.

An Analogue Method f o r Quantifying Impulsive Pike A , C . , Leverton J . W . : 9 , Conference i n 'Zurich, Swit- Noise. Paper preeented at I n t e r n e i s e - 77.

. zerland, Marcti. 1977.

C-ents on BBN Report 3573 - Subjective Response t o

10. Southw,ood B. J. : Skmulated and Actual Helfcopter Noise, W€€L Applied Acoustics Department Note 1216, March 1978.

Psychoacoustic Studies of Impulsive Noise.

11. Wright D r . S.E., Damongeot A: ' Paper No.55 presented at t h e Third European Rotorcraft and Powered L i f t September 1977.

A i r c r a f t 'Forum, A i x e n Provence, France, 12. Laudien E, Huljer H.: Impulsive Helicopter Rotor Noise. Paper No. 2.4 presented a t t h e GARTEur - 5 S p e c i a l i s t Meeting on P r o p e l l e r and H e l i - copter Noise, Parfs, 1-2nd June 1977,

13. Leverton J . W . : Helicopter Noise - Blade Slap P a r t 2: Experimental Results

March 1972.

NASA Report CR-1983 " (Prepared by ISVR,' Southampton), 14, Southwood B . J . , Leverton J . W . , Pike A , C , : Rating T a i l Rotor Noise, Paper presented at the I O A Spring Conference, Cambridge, A p r i l 1978.

Figure 1.- Flyover t i m e h i s t o r y - 2 bladed main r o t o r h e l i c o p t e r

(UH-1B).

I m .

-10 -5 0 5

T I M € - 1 . (run ov:Rn#no W % W I ~ N )

(V107).

Figure 2.- Flyover time h i s t o r y - tandem r o t o r h e l i c o p t e r

-10 0 5 -2 5 -20 -IS OVCR CI En 0 T t M E AS. vomrront Figure 3.- Flyover t i m e history - Scout and Wessex.

Figure 4 . - Blade slap: subjective correction.

Figure 5.- C h a r a c t e r i s t i c s of impulse d e s c r i p t o r s - v a r i a t i o n with r e p e t i t i o n rate.

Figure 6.- Impact of blade s l a p d e s c r i p t o r s on EPNL ( i d e a l i s e d s i g n a l s ) .

s

E

..

\

P

r y 4 I n Y L

Q s 3

u

I

P

I

Figure 7 . - Impact of blade slap descriptors on EPNL - simulated

and real helicopter results.

BLADE S I R P I U L S E 2 S O M t ~ 4 n r &no€ ?Asshue INr€RVaI KAz 66 ~a TbiL ROTOR NOISE -4- 0.2- /UTEGRA nor4 PERtOO

I H - I n = 0.2ns

Figure 8.- Diagrammatic representation of blade slap and tail rotor noise.

HLC IcoQr&R w i r n TAIL Roroa NOISE (?AIL 90.3 P N d a PtvLr 9 2 P W B ) F E P 6 U E I N C Y-Uz Figure 9 . - octave band of helicopter noise.

Figure 10.- Flyover t i m e history. Scout helicopter wXth high level of t a i l rotor noise.

Figure 11.- V a r i a t i o n of main and t a i l r o t o r n o i s e w i t h i n t e g r a t i o n t i m e and d e s c r i p t o r .

. - -e I I Figure 12.- C h a r a c t e r i s t i c s of impulse d e s c r i p t o r s - v a r i a t i o n with p u l s e frequency.

Figure 1 3 . - Tail rotor noise. Subjective correction - preliminary

results.

ANNOYANCE OF HELICOPTER IMPULSIVE NOISE F . d'Ambra and A. Damongeot S.N.I. Aerospatiale SUMMARY Psychoacoustic studies of helicopter impulsive noise have been conducted in ordkr to qualify additional annoyance due to this feature and to develop physical impuisiveness descriptors to develop impulsivity correction methods.

The paper reviews the explored impulsiveness parameters and the subjective evaluation data. The currently proposed descriptors and methods of impulsive- ness correction are compared using a multilinear regression analysis technique.

It is shown that the presently IS0 recommended descriptor and correction method provides the best correlation with the subjective evaluations of real helicopter impulsive noises. The equipment necessary for data processing in order to apply the correction method is discussed.

INTRODUCTION During the last few years, environmental agencies of different countries have expressed a need to establish and enforce a certification rule for all types of flying vehicles and in particular for helicopters. Among the different problems to be solved in order to promote such a certification rule, the question of representative noise unit is of utmost importance.

Indeed it should obviously:

-

Reflect the true annoyance felt by the public

-

Allow comparisons with the annoyance due to operations of other types of flying machihes

-

Reflect truly the efforts that the aircraft manufacturer and operator put in the design and operations of their vehicle to fly more quietly

- Not affect the present units used for aircraft

- Be as simple as possiHe for data processing

In view of this forthcoming c e r t i f i c a t i o n scheme of h e l i c o p t e r n o i s e , several c o u n t r i e s have undertaken some work on t h e p a r t i c u l a r f e a t u r e s of h e l i c o p t e r n o i s e i n o r d e r t o assess r e p r e s e n t a t i v e n o i s e u n i t s based on c o r r e c t i o n s t o t h e p r e s e n t l y accepted a e r o n a u t i c a l n o i s e u n i t s .

These u n i t s already t a k e i n t o account t h e e f f e c t of p a r t i c u l a r d i s t r i b u t i o n of a c o u s t t c (Noy) and t h e e f f e c t of tone and duration energy i n t h e audio frequency range of t h e n o i s e (EPNL). A new f e a t u r e which has n o t y e t been taken i n t o account i n t h e n o i s e s i g n a t u r e of f l y i n g machines is t h e inipulsive type of p r e s s u r e s i g n a l s which t h e majority of h e l i c o p t e r shows i n some f l i g h t configurations.

This impulsive f e a t u r e is a l s o found i n o t h e r n o i s e sources of day t o day l i f e , l i k e r e p e t i t i v e hammer blows, pneumatic d r i l l s , and motorcycles.

The work reported i n t h i s paper w a s supported by the "Ministzre de l a

Culture et de l'environnement ," t h e "Service technique Agronautique - Section

Moteurs" and S.N.I. g r o s p a t i a l e .

For t h e motivations previously s t a t e d , t h e work has been conducted i n such a way t o promote p o s s i b l e c o r r e c t i o n methods t o already e x i s t i n g aeronauti- cal, and t o a lesser degree, c i v i l engineering, n o i s e u n i t s . A l a r g e p a r t of t h e s u b j e c t i v e d a t a which are analyzed o r i g i n a t e s from psychoacoustic tests performed i n France by a j o i n t team of t h e Helicopter and Aircraft Division of Aerospatiale. Other s u b j e c t i v e d a t a and magnetic t a p e recordings used f o r psychoacoustic tests performed i n o t h e r c o u n t r i e s have been kindly made a v a i l a b l e i n t h e f r h e w o r k of IS0 and ICAO-WG.B working sessions. These d a t a have a l s o been incorporated t o t h i s study.

The paper is divided i n t o f o u r main s e c t i o n s : ( I ) Physical impulsiveness parameters: Subjective e v a l u a t i o n methods and r e s u l t s (11) Data I n t e r p r e t a t i o n : Impulsiveness d e s c r i p t o r s and p o s s i b l e methods of c o r r e c t i o n s ~ (111) M u l t i l i n e a r Regression Analysis: Quality criteria of t h e proposed methods (IV) Instrumentation and d a t a processing Some a s p e c t s of t h i s r e p o r t have already been presented at t h e t h i r d European R o t o r c r a f t and Powered L i f t A i r c r a f t Forum by D r . S. E. Wright and A. Damongeot ( r e f . 1). They pertained mainly t o t h e above mentioned s e c t i o n s I and 11.

SYMBOLS I Impulsive n o i s e s i g n a l Nonimpulsive n o i s e s i g n a l ' n Impulsive n o i s e l e v e l ; PNdB, TPNdB, EPNdB u n i t s (1) Nonimpulsive n o i s e level; PNdB, TPNdB, EPNdB u n i t s (HI) CFMax,CFN Maximum Crest Factor (peak) (r.m.s) Crest Factor during a 0.5 sec t i m e i n t e r v a l CFO. 5

-

Mean 0.5 sec Crest Factor during a t r a n s i e n t s i g n a l CFO. 5 Pressure and "A" weighted pressure t i m e h i s t o r y , ' A

-

X,X 0.5 sec and mean value of t h e IS0 d e s c r i p t o r A , B , C C o e f f i c i e n t s of t h e regression l a w i n dB, dB p e r u n i t value of the impulsive d e s c r i p t o r , dB p e r u n i t value of t h e r e p e t i t i o n rate SA,SB,SC Standard d e v i a t i o n of A,B,C e r ,re Multiple c o r r e l a t i o n c o e f f i c i e n t s Se Overall standard deviation AS Subjective c o r r e c t i o n (dB) hC Computed c o r r e c t i o n (dB) f Pulse r e p e t i t i o n rate (Hz) IMPULSIVENESS PARAMETERS AND SUBJECTIVE EVALUATION mTHODS It has been shown elsewhere ( r e f . 2) by c a r e f u l recordings of h e l i c o p t e r n o i s e s i g n a l s performed e i t h e r with microphones on t h e ground and a i r c r a f t i n hover o r i n f l i g h t (descent'or f l y o v e r a t high speed) o r w i t h microphones set i n t h e same reference frame as t h e h e l i c o p t e r i n motion t h a t t h e impulsiveness content of h e l i c o p t e r n o i s e s i g n a l s is mainly linked t o aerodynamic phenomena on t h e main r o t o r and t o a lesser e x t e n t t o t h e t a i l r o t o r . This impulsive c h a r a c t e r arises when t h e r e is e i t h e r a s t r o n g i n t e r a c t i o n between t h e main r o t o r blades and t h e w a k e v o r t i c e s ( f l i g h t of descent o r hover) or when a high aerodynamic speed relative tp t h e t i p of t h e advancing blade is reached (compressibility and/or thickness e f f e c t ) .

Physical Impulsiveness Parameters The main physical parameters which d e s c r i b e t h e pressure trace of sta- t i o n a r y n o i s e s i g n a l i n t h e s e circumstances are shown i n f i g u r e 1.

The impulse shape may b e d i f f e r e n t , as i l l u s t r a t e d f o r a l i g h t o r a (a) heavy h e l i c o p t e r .

The degree of i m p u l s i v i t y can vary widely from a pure random n o i s e (b) t o a weak, a medium, and a s t r o n g impulsive noise.

The p u l s e r e p e t i t i o n rate - which is e q u a l t o t h e r o t o r angular

(c)

frequency t i m e s t h e number of blades - can a l s o vary i n a wide range depending

on t h e h e l i c o p t e r weight ( r o t o r diameter and t i p speed) and t h e number of b l a d e s used (two t o s i x in present design).

(d) F i n a l l y t h e n o i s e l e v e l s , expressed i n t h e p r e s e n t l y agreed n o i s e u n i t s of PNdB, can vary t o a lesser e x t e n t at the d i s t a n c e s which are a c t u a l l y sought f o r p o s s i b l e c e r t i f i c a t i o n scheme. Figure 2, which reproduces n o i s e traces taken a t d i f f e r e n t t i m e s during t h e f l y o v e r of a heavy h e l i c o p t e r , shows t h a t , f o r t r a n s i e n t n o i s e s i g n a l s , t h e r e is i n a d d i t i o n t o t h e previously s t a t e d parameters an evolution of t h e degree of impulsivity of t h e n o i s e s i g n a l , t h e t r e n d being t h a t t h i s degree i n c r e a s e s b e f o r e t h e maximum n o i s e l e v e l is obtained, then decreases sharply afterward.

Subjective Evaluation Methods t o submit t o a The p r i n c i p l e of a s u b j e c t i v e evaluation experiment is c e r t a i n j u r y t h e n o i s e s i g n a l t o be q u a l i f i e d and a reference n o i s e of known annoyance. Broad band n o i s e has been t h e siibject of many s u b j e c t i v e e v a l u a t i o n s from which t h e u n i t of PNdB (Perceived Noise decibel) has been derived. There- f o r e , it can r e p r e s e n t a very good reference, e s p e c i a l l y i f it is taken a s t h e broad band n o i s e of a h e l i c o p t e r .

Elaboration of Impulsive Noise Recordings I n o r d e r t o be a b l e t o change at w i l l t h e d i f f e r e n t parameters which w e r e pointed o u t i n t h e previous paragraph, it is necessary t o e l a b o r a t e t h e a way t h a t one can s e p a r a t e t h e s e impulsive n o i s e s i g n a l t o q u a l i f y i n such parameters, while using as much as p o s s i b l e t h e a c t u a l h e l i c o p t e r n o i s e traces. I n the'French psychoacoustic tests performed, t h i s has been achieved by e l e c t r i c a l l y mixing h e l i c o p t e r broad band n o i s e signals w i t h real h e l i c o p t e r impulse s i g n a l s , as shown i n f i g u r e 3. This allows t h e pulse amplitude and r e p e t i t i o n rate t o b e v a r i e d a t w i l l so t h a t t h e f o u r p h y s i c a l parameters, shape, degree, r e p e t i t i o n rate, and o v e r a l l n o i s e l e v e l s , could be t e s t e d s e p a r a t e l y .

The same technique has been applied t o e l a b o r a t e t h e t r a n s i e n t impulsive t o be t e s t e d . same p u l s e shape is maintained during t h e ioise s i g n a l s The :ime h i s t o r y of t h e p r e s s u r e s i g n a l b u t with a v s r i a t i o n of t h e degree of .mpulsivity according t o t h e previously mentioned trend: i n c r e a s i n g degree iefore m a x i m u m PNL, sharp decrease t h e r e a f t e r .

Method of Comparisons by P a i r s I n t h e French study, t h e method used t o s u b j e c t i v e l y e v a l u a t e impulsive To ioise is t h e method of comparison by p a i r s , i l l u s t r a t e d i n f i g u r e 4.

!valuate given impulsive n o i s e I, f i v e levels of nonimpulsive n o i s e Rn are For >layed twice. The t e n p a i r s (I,%) and (Rn,I) are compared a t random.

zach comparison, t h e J u r y is asked- simply Which' n o i s e is t h e most annoying?"

Che percentage of t h e J u r y who f i n d s t h e nonimpulsive n o i s e more annoying is :hen p l o t t e d a g a i n s t t h e d i f f e r e n c e i n l e v e l between t h e nonimpulsive (NI) and :he impulsive n o i s e (I) measured i n p r e s e n t s u b j e c t i v e n o i s e u n i t s : APNdB and lTPNdB f o r s t a t i o n a r y s i g n a l s , AEPNdB f o r t r a n s i e n t s i g n a l s , Two " s e n s i t i v i t y x r v e s " are obtained as shown i n f i g u r e 5: one r e l a t i v e t o (NI) being played > e f o r e (I), t h e o t h e r r e l a t i v e t o (NI) being played a f t e r (I). The mean curve chosen t o be t h e c h a r a c t e r i s t i c response curve. The annoyance c o r r e c t i o n Is 1s i s then considered t o be such that 50% of the J u r y f i n d the ( I ) and (NI) Levels equally annoying.

One can n o t i c e from f i g u r e 5 t h a t t h e impulsive n o i s e (I) is found more This t r e n d , mnoying when it is played a f t e r t h e nonimpulsive n o i s e (NI).

constantly 'noticed throughout t h e complete study, shows t h a t t h e r e is a memory s f f e c t which tends t o emphasize the last event f e l t by t h e J u r y as compared t o t h e previous event.

This method of comparison by p a i r s needs an anechoic chamber and a l a r g e But it does n o t r e q u i r e s u b j e c t s acquainted t o t h e s p e c i f i c problem t o jury.

b e studied.

It q u a l i f i e s i n some way t h e annoyance f e l t by a g e n e r a l p u b l i c of a n o i s e s i g - not s p e c i f i c a l l y motivated t o exaggerate some s p e c i a l f e a t u r e s n a l which they would r e s e n t due t o previous exposures as it could b e t h e case f o r i n h a b i t a n t s l o c a t e d i n t h e immediate v i c i n i t y of a h e l i p o r t .

Details on t h e number and s e l e c t i o n procedure of t h e French Jury are s p e c i f i e d i n r e f e r e n c e 1 (about 60 persons r e t a i n e d after audiometer t e s t s ) .

Method of P a i r s Adlustment It is based on t h e same p r i n c i p l e of comparison between a reference n o i s e and a n o i s e t o b e q u a l i f i e d , . b u t t h e s u b j e c t is allowed t o change t h e reference n o i s e l e v e l and play back and f o r t h t h e two n o i s e s t o be compared u n t i l t h e equal annoyance of t h e two n o i s e s is reached. The i t e r a t i o n procedure followed by t h e s u b j e c t can be recorded and allows a b e t t e r statisti- cal i n t e r p r e t a t i o n of t h e r e s u l t s obtained.

The s u b j e c t i v e evaluation of t h e impulsive content of n o i s e s i g n a l s has been evaluated by this method of adjustment i n several research centers: Westland Helicopters Ltd., National Physical Laboratory (U.K.), and Bolt Beranek and Newman (U.S.A.).

The a p p l i c a t i o n of t h e method does n o t r e q u i r e an anechoic chamber and i t can b e conducted w i t h a j u r y of smaller s i z e .

During t h e course of t h e experiments, t h e s u b j e c t s a c q u i r e more experience i n the p a r t i c u l a r f e a t u r e s of t h e n o i s e s i g n a l s t o b e t e s t e d . I n some way, t h e c o r r e c t i o n s found f o r impulsive s i g n a l s could b e c l o s e r t o t h e opinion of i n h a b i t a n t s l o c a t e d i n t h e immediate v i c i n i t y of an h e l i p o r t .

Subjective Evaluation Results Tables I and I1 provide t h e s u b j e c t i v e r e s u l t s obtained i n t h e French study. The impulse shape used is i d e n t i f i e d , t o g e t h e r w i t h t h e - value of t h e

d i f f e r e n t impulse parameters previously discussed. The CF , CF0.5, and 2

columns, which s p e c i f y t h e degree of impulsivity, w i l l be Yiscussed and i d e n t i f i e d i n t h e s e c t i o n e n t i t l e d "Impulsiveness Descriptors and P o s s i b l e Methods f o r Corrections.

Table 111 provides t h e s u b j e c t i v e r e s u l t s of real h e l i c o p t e r t r a n s i e n t n o i s e which have been k i n d l y provided by Bolt Beranek and Newman Inc., together with a d u p l i c a t e of t h e recordings of t h e t e s t e d A o i s e s i g n a l s .

Parameters p e r t a i n i n g t o t h e degree of impulsivity (CFN, CF0.5 , ; ) have been computed from t h i s tape.

Both types of experimental r e s u l t s show t h a t c o r r e c t i o n s of 0 t o 7 dB have t o be added t o conventional u n i t s of PNdB o r EPNdB t o reflect t h e annoyance e f f e c t of impulsive noises. A s shown i n reference 1, t h e j u r y re- sponses are s t a t i s t i c a l l y meaningful, giving a 90% confidence l e v e l of 21.3 dB.

IMPULSIVENESS DESCRIPTORS AND POSSIBLE METHODS FOR CORRECTIONS A n "impulsive descriptor'' is a mathematical expression which is as simple as p o s s i b l e f o r ease of d a t a processing and which could, as much as p o s s i b l e , provide a good c o r r e l a t i o n between t h e value of t h e d e s c r i p t o r and t h e s u b j e c t i v e c o r r e c t i o n AS.

S t a t i o n a r y Noise Signals : Impulsiveness Descriptors For s t a t i o n a r y s u b j e c t i v e d a t a , examination of t h e j u r y c o r r e c t i o n s AS as a f u n c t i o n of n o i s e levels (90 and 100 PNdB) shows p r a c t i c a l l y no i n f l u e n c e of t h i s parameter w i t h i n t h i s s h o r t range of v a r i a t i o n . The t h r e e o t h e r

parameters , namely shape , degree, and p u l s e r e p e t i t i o n rate, have been

combined i n t o one s i n g l e d e s c r i p t o r through use of:

-

An "A" f i l t e r e d s i g n a l which tends t o decrease t h e e f f e c t of t h e low frequency content of t h e impulse

-

P4/(?$ where PA i s t h e A f i l t e r e d - An impulsivity c o e f f i c i e n t

A sound pressure t i m e h i s t o r y A s shown i n r e f e r e n c e . 1 , f o r a pure p e r i o d i c p u l s e t r a i n of pulse width a

and period T , t h e unweighted c o e f f i c i e n t p/( 3)' turns' out t o be equal

rn t o be equal t o k where k depends on t h e pulse shape ( r e c t a n g l e , k = 1; t r i a n g l e , k = 1.8). I f T i n c r e a s e s , i.e., i f t h e pulse r e p e t i t i o n rate decreases, o r i f a decreases ( i . e o 5 i f t h e "spikyness" i n c r e a s e s ) , then t h i s impulsivity c o e f f i c i e n t increases.

Thus shape, degree, and pulse r e p e t i t i o n rate are indeed taken i n t o

-

account i n t h i s P4 A / (%)2 d e s c r i p t o r .

I n t h e framework of t h e I n t e r n a t i o n a l Standard Organization (ISO) Working Group 2 , t h i s t o p i c of impulsiveness d e s c r i p t o r has been brought f o r t h and several impulsiveness d e s c r i p t o r s have been submitted f o r examination. Among t h e d i f f e r e n t proposals (Westland Helicopters Ltd., South Africa National Research I n s t i t u t e , France SNI Aerospatiale, U.K. National Physical Laboratory) submitted before t h e i r last meeting d a t e (Dec. 5, 1977), t h e NPL proposal* has been r e t a i n e d and recommended f o r a p p l i c a t i o n t o ICAO - Working Group B.

The NPL d e s c r i p t o r is based on t h e variance of t h e square of t h e "A" "A" weighted sound pressure s i g n a l divided by t h e square value of t h e d.s.

weighted sound pressure s i g n a l : It can b e shown t h a t t h e I d e s c r i p t o r is i d e n t i c a l t o t h e French d e s c r i p t o r minus one. So t h i s d e s c r i p t o r does t a k e a l s o i n t o account t h e shape, degree, and pulse r e p e t i t i o n rate parameters of impulsive noise. The o t h e r d e s c r i p t o r s peak ). A s shown i n proposed w e r e mainly based on t h e C r e s t , F a c t o r concept ( r . m . s Zigure 6 , where comparisons are made on t h e same n o i s e s i g n a l s between x = 10

Log I and CF = 20 Log (e), t h e latter d e s c r i p t o r p r e s e n t s a l a c k of

r.m. s s e n s i t i v i t y .

*With a "short i n t e g r a t i o n time" 5 - 200 ps.

S t a t i o n a r y Noise Signals: Correction Methods Once an impulsiveness d e s c r i p t o r has been chosen, a c o r r e c t i o n method can be e a s i l y b u i l t using a b e s t f i t technique between t h e subjective- c o r r e c t i o n s A S and t h e computed c o r r e c t i o n s AC.

I n t h e IS0 N 356 proposal, t h e recommended c o r r e c t i o n law is

0.8 (X - 3)

(dB) where x = 10 Log I This c o r r e c t i o n is l i m i t e d t o t h e range of 0. K A C 45.5 dB and held constant at 5.5 dB f o r l a r g e r values of x. It is t o be noted t h a t f o r x = 3 and I z 2 , t h e n o i s e s i g n a l is purely broad band.

This c o r r e c t i o n method applied t o t h e French s u b j e c t i v e d a t a provides a standard d e v i a t i o n of - +l. 3 dB.

Transient Noise Signals: Impulsiveness Descriptors and Correction Methods I n t h e IS0 N 356 recommended procedure, t h e impulsiveness d e s c r i p t o r remains t h e same as i n t h e s t a t i o n a r y case. The I Descriptor is computed at each 0.5 sec t i m e i n t e r v a l , t h e c o r r e c t i o n AC is added t o t h e LTpN giving t i m e h i s t o r y from which t h e correcte!'gPNL is computed. This pro- a LITPBT cedure is s y n t h e t i c a l l y presented i n f i g u r e 7.

A t t h e l a s t IS0 Working Group 2 meeting, another procedure has been pro- posed t o I S 0 members ( r e f . 3 ) . A s presented s y n t h e t i c a l l y i n f i g u r e 8 , t h e impulsiveness d e s c r i p t o r is based on a "A" weighted Crest Factor CF, and t h e impulse r e p e t i t i o n rate ( f ) is taken as a complementary d e s c r i p t o r .

Two p o s s i b l e impulsiveness c o r r e c t i o n methods w e r e presented which are b r i e f l y sketched i n f i g u r e 9.

I n t h e f i r s t one, t h e "A" weighted Crest Factor i s computed every 0.5 s e c of t h e t r a n s i e n t s i g n a l and a c o r r e c t i o n l a w is applied t o LTpNy giving a L ITPN t i m e h i s t o r y from which t h e corrected EPNL is computed.

An a l t e r n a t e method presented w a s t o compute an o v e r a l l c o r r e c t i o n A,

* t o t h e EPNL which is based on t h e "A" weighted maximum Crest Factor C F

M a X measured during t h e t r a n s i e n t n o i s e s i g n a l .

The c o r r e c t i o n l a w s proposed i n t h e two cases are l i n e a r as function of Crest Factor and p u l s e r e p e t i t i o n frequency ( f ) :

AdB = A + B - (CF) + C*f

"[Max (peak)]/[Max (r.m.s)] , each f a c t o r measured independently.

and t h e A, B, and C c o e f f i c i e n t s are obtained i n each case by a m u l t i l i n e a r regression a n a l y s i s using t h e s u b j e c t i v e c o r r e c t i o n s AS of t a b l e 111 (nine experiments) as i n p u t data.

Procedures Discuss ion The u s e of a m u l t i l i n e a r regression technique t o o b t a i n a b e s t f i t correction method is indeed a very good approach, provided t h a t a l a r g e number of experiments i s taken i n t o account i n t h e computation process of t h e c o e f f i c i e n t s of the regression law. Otherwise, t h e c o r r e c t i o n l a w obtained may very w e l l f i t t h e experimental d a t a which are used as i n p u t , while p u t t i n g , on some parameter, a weight through t h e regression l a w c o e f f i c i e n t which does n o t r e f l e c t i t s t r u e importance. More p r e c i s e l y , a t a t i m e when manufacturers are t r y i n g t o increase t h e number of blades of t h e i r r o t o r s (ex., Hughes A i r c r a f t Company, Quiet Helicopter Program) i n order t o decrease t h e n o i s e , it is very important t o know i f t h e p u l s e r e p e t i t i o n rate ( f ) has t o b e an independent parameter, and i f it i s t h e case, what values should be chosen f o r i t s r e g r e s s i o n c o e f f i c i e n t C and its accuracy.

a m u l t i l i n e a r regression a n a l y s i s I n order t o answer t h e s e questions, has been performed on t h e complete set of a v a i l a b l e d a t a presented i n t a b l e s I, 11, and 111, using as p o s s i b l e d e s c r i p t o r s t h e two previous Crest Factors CF and CFMax, t h e IS0 d e s c r i p t o r x together w i t h t h e pulse r e p e t i t i o n r a ? ~ ~ f .

MULTILINEAR REGRESSION ANALYSIS The The method used i n t h i s a n a l y s i s is b r i e f l y sketched i n f i g u r e 10.

method is i d e n t i c a l t o t h e m u l t i l i n e a r regression a n a l y s i s used i n appendix A of reference 3 . Regression c o e f f i c i e n t s A , B , and C are computed using on one hand t h e s u b j e c t i v e c o r r e c t i o n s AS as s t a t e d by t h e j u r i e s and impulsiveness parameters ( I . D . = x , CF ) and pulse r e p e t i t i o n rate f on t h e o t h e r hand.

CF0.5' M That is t o say, the c o r r e c t i o n l a w assumed is and A, B, and C are computed t o minimize I n a d d i t i o n t o t h e o v e r a l l standard deviation Se which r e s u l t s from t h e and of t h e m u l t i p l e c o r r e l a t i o n c o e f f i c i e n t r, t h e present b e s t f i t technique, SB, and SC f o r t h e c o e f f i c i e n t s A , study defines a l s o t h e standard deviation SA, B, and C.

These standard deviations on t h e regression c o e f f i c i e n t s d e f i n e t h e accuracy provided on t h e s e c o e f f i c i e n t s A, B , and C by t h e method of analysis.

They show, i n a simple manner, t h e confidence level t h a t one can g r a n t t o each parameter (I.D., f ) taken i n t o account, For example, it is p o s s i b l e t h a t t h e m u l t i p l e c o r r e l a t i o n c o e f f i c i e n t r be s t a t i s t i c a l l y s i g n i f i c a n t f o r a given:confidence level, and t h a t , at t h e same time, one o b t a i n s a standard d e v i a t i o n on one of t h e c o e f f i c i e n t s as l a r g e as t h e value of t h i s c o e f f i c i e n t i t s e l f , Obviously, i n t h i s case t h e parameter r e l a t e d t o t h i s c o e f f i c i e n t has no real s i g n i f i c a n c e . These q u a l i t y criteria have been summarized i n f i g u r e 10.

Application of t h e Regression Analysis

- For t h e C F and CF d e s c r i p t o r s , t h e procedure underlined i n

Max 0.5 r e f e r e n c e 3 has been followed. It i s f i r s t necessary f o r a t r a n s i e n t s i g n a l t o For t h i s purpose, CFOe5 is com- define a mean value of t h e d e s c r i p t o r CF 0.5' puted a t each 0.5 sec t i m e i n t e k v a l from t h e "A" weighted n o i s e signal, and a "first" c o r r e c t i o n ACF i s computed 0,5

ACF0.5 = CF0.5 - 1 2 (ACFOe5 > 0)

- A c o r r e c t e d PNLT is then computed

= PNLT + ACFOe5 PNLT

corr.

is used i n t h e i n t e g r a t i o n process t o compute t h e EPNL which c o r r .

sec Crest Factor f o r t h e complete s i g n a l is obtained by t h e A mean 0.5 following expression:

-

C F ~ . 5 = (Epmcorr, - EPNL) + 1 2

- A mean value of t h e IS0 impulsiveness d e s c r i p t o r x has a l s o t o be

-

This mean d e s c r i p t o r x computed i n o r d e r t o conduct t h e regression a n a l y s i s .

i s computed following t h e same procedure as i n t h e case of t h e CF 0.5 d e s c r i p t o r .

The, EPNL c o r r e c t i o n i s f i r s t computed following t h e IS0 N 356 recommended method. Then is deduced from t h i s c o r r e c t i o n s e t t i n g .

- EPNL = 0.8 (2 - 3)

EPNL c o r r .

-

- The values of CFNax, CFOa5, and 2 are i n d i c a t e d f o r each recording used

i n t h e s u b j e c t i v e evaluation methods i n t a b l e s I, 11, and 111. It is t o b e noted t h a t throughout t h i s regression a n a l y s i s a good consistency has been maintained i n t h e hypothesis i n order t o g e t comparable r e s u l t s :

- The c o r r e c t i o n s q a t each 0.5 sec are applied from x = 3 (AC = 0)

and CFOe5 = 12 which represent t h e values of a pure broad band n o i s e s i g n a l .

- The maximum c o r r e c t i o n a t each 0.5 sec is 5.5 dB i n each case.

- The r e g r e s s i o n a n a l y s i s is conducted f o r t h e t h r e e d e s c r i p t o r s i n two cases : ( a ) Taking i n t o account t h e p u l s e r e p e t i t i o n rate ( f ) as an independent parameter (b) Discarding t h e pulse r e p e t i t i o n rate f i n t h e regression l a w .

Regression Analysis Results Table I V summarizes t h e r e s u l t s obtained i n t h i s regression analysis.

The following remarks can be drawn from t h i s t a b l e .

(a) The o v e r a l l standard d e v i a t i o n is minimum (1.4 dB) when t h e IS0 recommended method is used.

(b) The m u l t i p l e c o r r e l a t i o n c o e f f i c i e n t r is much higher (>0.75j when t h e I S 0 recommended method i s use& while it is b a r e l y s i g n i f i c a n t at 1% confidence l e v e l with t h e CF o r CFO, d e s c r i p t o r .

M a x

(c) Discarding t h e p u l s e r e p e t i t i o n rate - which has a l o w regression

c o e f f i c i e n t C with a very high (44 t o 7 5 % ) r e l a t i v e standard d e v i a t i o n

(SC/C) - improves very much t h e standard deviation SA which influences

d i r e c t l y (dB) t h e l e v e l of t h e c o r r e c t i o n AC.

the q u a l i t y criteria of t h e Table V provides a d i r e c t eomparison of From t h i s t a b l e one can conclude regression a n a l y s i s i n t h e s i x cases t r e a t e d .

t h a t :

- The pulse r e p e t i t i o n rate should b e discarded as an independent

parameter

- The I S 0 recommended procedure provides on a statistical b a s i s t h e

b e s t a v a i l a b l e method a t t h e present t i m e These conclusions are more c l e a r l y i l l u s t r a t e d i n t a b l e VI which r e p r e s e n t s t h e computed r e s u l t s obtained from a p p l i c a t i o n of t h e regression conducted l a w s without r e p e t i t i o n rate dependency on t h e psychoacoustic tests on t h e real h e l i c o p t e r n o i s e s i g n a l s of reference 3. Comparing computed c o r r e c t i o n s AC with t h e J u r y s u b j e c t i v e evaluations l e a d t o t h e following remarks : and Eo.5 g i v e high p e n a l t i e s ( 2 . 3 and 1.8 dB) f o r t h e reference

- CF

M n o i s e supposedly nonimpulsive

- Overall standard d e v i a t i o n between t h e computeLand t h e s u b j e c t i v e

method, and 1 . 7 dB r e s u l t s is 1,l dB f o r t h e IS0 method, 1 , 5 dB f o r t h e CB 0.5 f o r t h e CF method of c o r r e c t i o n M INSTRUMENTATION AND D A T A PROCESSING Procedure I n t h e I S 0 N 356 recommended method, t h e impulsiveness d e s c r i p t o r I. is defined as follows: 1. The a c o u s t i c s i g n a l is weighted through a f i l t e r "A," then sampled a t a frequency of 5000 Hz.

2. The d i g i t a l values "v 'I thus obtained are processed, every 0.5 sec, i i n two s t e p s :

- The mean square "s" of vi at each 0.5 sec (i.e., t h e square of t h e

s i g n a l r . m . s v a l u e ) , is f i r s t computed.

n

n = 2500 s = L c vi

n i= 1

- The impulsivity d e s c r i p t o r I is then

yS- s j - L L n ( v : - sl 2

-

I = - c

(1) n 2 n

s i=l

i= 1 It has been proved mathematically i n reference 4 t h a t t h i s procedure l e a d s t o t h e value of t h e previously proposed French d e s c r i p t o r minus one, t h e latter d e s c r i p t o r being defined as : 1 o r 2 can be used f o r t h e impulsivity So i n p r i n c i p l e , e i t h e r expressions d e s c r i p t o r computation process.

Instrumentation and Cost The procedure previously underlined implies t h e use of:

- t h e "A" weighting and a n t i a l i a s i n g (0 t o 2000 Hz) f i l t e r s

- an a n a l o g / d i g i t a l converter (+lo v o l t s ; 11 b i t s + s i g n ) a s s o c i a t e d with a 5000-Hz timer c o n t r o l l i n g t h e s i g n a l sampling are q u i t e inexpensive ( ~ $ 4 0 0 ) .

These instruments The t i m e r I s synchronized with the 1 / 3 octave analyzer used t o compute each 0.5 sec t h e 24 l e v e l s of t h e 1/3 octave spectrum needed f o r PNL computation.

Two cases have then t c ) b e considered (see f i g . 11) : Case 1: t o b e performed The laboratory in which t h e d a t a processing is uses a 1/3 octave analyzer coupled t o a 5000 Hz real t i m e computer f o r t h e I t a s k i n normal operations of PNL and EPNL computations. I n t h i s case, t h e computer has enough s t o r a g e capacity t o s t o r e 5000 sampled values (two t i m e i n t e r v a l s of 0.5 see, ATi-l, and AT. w i t h 2500 values v . a t each t i m e i n t e r v a l ) and enough performance t o compute tfie I value (5000 m u l h p l i c a t i o n s , 2500 s u b t r a c t i o n s , and few d i v i s i o n s ) of t h e AT t i m e i n t e r v a l during t h e t i m e

i-4

allowed (0.5 s e e ) f o r t h e "vi1' a c q u i s i t i o n o t h e next AT. t i m e i n t e r v a l .

The 1 / 3 octave analyzer is coupled t o a lower performance Case 2: computer not allowing real t i m e computation of t h e I t a s k a t t h e frequency of 5000 Be. I n t h i s case a complementary "mini computer" i s necessary t o at each 0.5 s e c t h e I value of s t o r e t h e 5000 sampled values and compute t h e previous 0.5 sec t i m e i n t e r v a l . This type of mini computer i s a t t h e present t i m e a v a i l a b l e commercially a t a low p r i c e compared t o t h e o t h e r equipment necessary f o r pure EPNL computations. F m example, i n France, t h e AMs1 type ALPHA LSI 4/90 which has t h e f l o a t i n g p o i n t computation c a p a b i l i t y is s o l d i n France a t approximately $5000 which is

(2 K - W, 8 K ROM - PROM)

t o be compared with t h e $20 t o 30 000 needed f o r a 1 / 3 octave analyzer.

GENERAL CONCLUSIONS This study has e s s e n t i a l l y shown t h a t : (1) Impulsive n o i s e needs t o be c o r r e c t e d i n order t o r e p r e s e n t t h e annoyance r e a l l y f e l t by t h e public.

(2) Corrections up t o 7 PNdB o r EPNdB have been found by r e p r e s e n t a t i v e J u r i e s i n s e v e r a l c o u n t r i e s , w i t h a s t a n d a r d d e v i a t i o n of 21.3 dB.

(3) Among t h e proposed impulsiveness d e s c r i p t o r s and c o r r e c t i o n methods, t h e recommended IS0 N 356 procedure provides t h e b e s t c o r r e l a t i o n between s u b j e c t i v e and computed c o r r e c t i o n s .

(4) There is no s p e c i a l need t o add t o t h e proposed I S 0 N 356 procedure an a d d i t i o n a l c o r r e c t j o n term based on pulse r e p e t i t i o n rate, ( 5 ) The standard deviation obtained between computed and subjective annoyance of real helicopter noises is +1.1 dB which is comparable to the standard deviation of the jury subjective responses ( 6 ) The instrumentation and computing hardware necessary for data processing are available on the market at a small price compared to the cost of the equipment needed for EPNL data processing.

REFERENCES 1 . Wright, S. E.; and Damongeot, A . : Psychoacoustic Studies of Impulsive Noise. 3rd European Rotorcraft Forum.

A Simple Theoretical 2. Schmitz, F . H.; Boxwell, A. A . ; and Yu, Y. H.: 3rd European Rotorcraft Forum.

Model of High Speed Helicopter Noiser 3. Galloway, W, J.: Subjective Response to Simulated and Actual Helicopter Blade Slap Noise, BBN Report No. 3573.

4 . Damongeot, A . : Instrumentation for Measuring the Impulsivity Indicator, ' ISO/TC-43/SC-l/WG2, proposed by N.P.L. or France.

[ PULSE SHAPE I [ DEGREE OF IMPULSIVITY 1

WEAK

SIGNATURE I REPETITION RATE 1 I NOISE LEV€(.$ I

90 and 100 PNdB

? O H 3 - 5 8 H 3

Figure 1.- Impulsiveness parameters.

SPEED 60 kts- HEAVY HELICOPTER x s 7 7 0 8 10 SEC AFTER P N L M 3CS 3 T 0 4 WEAK lNPULSlVlTY Figure 2.- Evolution of impulsiveness degree during helicopter flyover.

+

Figure 3 . - Generation of h e l i c o p t e r n o i s e s i g n a l .

I : IMPULSIVE NOISE TO EVALUATE

1 FIXED CONDITIONS : NOISE LEVEL, DEGREE OF IMPULSIVITY(x.PULS€ SHAPO, PULSE REPETlTlON FIArE

R : NONlMPUtSIV€ COMPARISON NOISE AT SDIFF€RENT NOISE LEVELS

RI TO R5 EACH PAIR( I, Rn) EVALUATED 7WICE OROER (1,Rn)cmd (Rn,I)

Figure 4 . - S u b j e c t i v e evaluation method comparison by p a i r s .

'1004 I ; IMPULflVE NOISE % OF JURY JUO61N$ 90- 80- 7Q- 60-

--

50- 40- 30- 20-

--

'10- Y N I BEFORE I Y ' " I Q l I I I I I 0 3 6

15 (Z-1)

SUWEC b C O I UNITS

4 d S A , SNdO, T P N I P , FLPIYdI) Figure 5.- J u r y response curve.

Figure 6.- Impulsiveness d e s c r i p t o r s comparison.

I S 0 N 356 RECOMMENDED METHOD

COMPUTE I DESCRIPTOR AT EACH 0.5 SEC.

a APPLY AT EACH 0.5 SEC. A CORRECTION AC

T O T H E LTPN.

WHERE x = IO Log I

PUTE T H E EPLN FROM T H E CORRECTEC

LITPN T I M E HISTORY

Figure 7.- Impulsiveness c o r r e c t i o n methods.

0 ISOREC MENDED DESCRIPTOR

I -+BASED O N T H E V A R I A N C E O F

TH E SOUAR E 0 F TH E "A 'WE I G HTE D

E HISTORY.

CRIPTOR MENTIONED TO IS0

-

CF-BASED O N T H E CREST FACTOR O F

.THE "A "WEIGHT PRESSURE

T I M E HISTORY.

- IMPULSE REPETITION RATE (f) MENTIONED

I B L E CO P L E M E NTARY D ESC R I PTO R

Figure 8.- Impulsiveness d e s c r i p t o r .

I O COMPUTE CFo.5 DESCRIPTOR AT EACH 0.5 SEC.

0 APPLY AT EACH 0.5 SEC. A CORRECTION A TO THE LTPN

O-A =5.5

A d B = A + B ~ ~ 0 . 5 + cf

A, B, C OBTAINED FROM REGRESSION ANALYSIS OF PSYCHOACOUSTIC DATA.

0 COMPUTE THE EPNL FROM THE CORRECTED L ~ T P N TIME HISTORY ALTERNATE METHOD

TlON A 0 APPLIED TO EPNL BASED ON CF MAX

WHERE A, 9, C ARE COMPUTED FROM PSYCHOACOUSTIC DATA BY REGRESSION ANALYSIS TECHNIQUES Figure 9.- Impblsiveness correction methods.

INPUT

-

0 SET OF PSYCHOACOUSTIC DATA

JUDGED'RESPONSE : ( A s)

IMPULSE DESCRIPTOR (I.D.)

IMPULSE REPETITION RATE (f) : 0 LINEAR CORRECTION LAW ASSUMED A C = A + B ( I D ) + C (f) COMPUTE A, B, C. REGRESSION LAW COEFFICIENTS TOMINIMIZE I A C - A S 1 QUALITY CRITERIA

0 MULTIPLE CORRELATION COEFFICIENT r > re FOR

SIGNIFICANCE AT l0/o LEVEL 0 OVERALL STANDARD DEVIATION Se : MINIMUM STANDARD DEVIATION ON COEFFICIENTS A, B, C : SMALL Figure 10.- Multilinear regression analysis.

45 7 TYPE ALPHA

"A" WEIGHTING 11 BITS + SIGN I

ANTI-ALI AS1 NO ( 0 - 2 O W HZ ) Figure 11.- PNL, EPNL computer system with impulsivity corrections.

TABLE 1 . - PSYCHOACOUSTIC TEST DATA - STATIONARY NOISE SIGNALS*

-

TEST f REQ.

A S TYPE OF HELICOPTER X CFM No HZ PNdB

- -

201 HEAVY - N.I. (REF.)

17.5 12.9 12.4 3.3 0

202 LIGHT - N.I. (REF.) 17.5

15.2 12.9 3.6 0

206 HEAVY - 100 PNdB 10 14.7 14.8 9.5 5.6

210 LIGHT - 100 PNdB

10 14.9 13.8 4.0 0.9

211 LIGHT - 100 PNdB 10 19.4 20.3 4.2

9.6

LIGHT - 100 PNdB 10 20.3 22. 9.6 5.4

213 LIGHT - 90 PNdB 10 15.4 14.7 5.6 2.8

HEAVY - 90 PNdB 10 14.1 12.4 3.6 0.1

215 LIGHT - 90 PNdB 4.4

17.5 16.1 14.9 8.8

216 LIGHT - 90 PNdB 25 14.7 15.9 9 .o 6.4

LIGHT - 90 PNdB 35 16.7 14.8 8.4 5.8

- -- - -- - - - ------ ----

----. ----. ---- .----.

218 MOTORCYCLE - N.I. (REF.) 30 13.9 13.7 0

4.0 220 MOTORCYCLE 1 58 14.6 13.2 7.9 7 .O 221 MOTORCYCLE 2 24 19.4 16.6 3.2 9.8

-

* FRENCH STUDIES, COMPARISON BY PAIRS

TABLE 11.- PSYCHOACOUSTIC TEST DATA - TRANSIENT NOISE SIGN&S*

-

-

TEST FREQ. AS TYPE OF HELICOPTER X CFM cF0.5 No EPNdB

I HZ

-

101 FLYOVER - 70 Kt - N.I. (REF.) 17.5 12.5 12.3 3.5 0

102 H. LIGHT - 70 Kt - 95 EPNdB 4.2

10 15.8 15.2 6.1

103 HEAVY - 70 Kt - 95 EPNdB 17.5 14.3 14.5 4.1 2.1

104 FLYOVER - 148 Kt. N.I. (REF.) 13.5 4.1 0

17.5 13.8

105 H. HEAVY - 148 Kt - 95 EPNdB 1

17.5 15.8 17.8 9 .o

106 H. HEAVY - 70 Kt - 95 EPNdB 17.5 13.7 13.1 3.8 4

107 H. HEAVY - 70 K t - 95 EPNdB 7 .O 2.3

17.5 14.8 15.6

H. HEAVY - 70 Kt - 95 EPNdB

108 17.5 15.7 15.7 6.2 2.5

109 H. HEAVY - 70 Kt - 100 EPNdB 17.5 15.9 15.8 6.5 2.1

-

FRENCH STUDIES, COMPARISON BY PAIRS

TABLE 111.- PSYCHOACOUSTIC TEST DATA - TRANSIENT NOISE SIGNALS*

B.B.N. - -

FREQ. AS TYPE OF HELICOPTER X CFM FO .5 N" HZ EPNdB

214 S.61 (REF.) - 115 K t - LEVEL 17 14.3 12.7 3.5 0

21 5 S.64 - 60 K t - LEVEL 18.6 14.4 14.1 4.6 2.7

216 CH47.C. - 150 K t - LEVEL 12.5 15.8

17.6 8.3 7 .O

217 CH47.C. - 60 K t - LEVEL

12.5 17.3 16.9 8.3 5.5

21 8 B.212 - 105 K t - LEVEL

11 14.3 15.0 6.8 3.2 21 9

8.212 - 61 K t - LEVEL 11 19.4 15.8 7 .O 3.1

220 47.G. - 6" APPR. 12 17.7 16.1 7.1 3.5

221 S.61 - 6" APPR. 17 15.6 14.6 6.4 3.8

222 8.206 L - 6" APPR. 13 21.4 15.8 7 .O 3.6

P * B.B.N. STUDIES, METHOD OF ADJUSTMENT DATA COMPUTED FROM B.B.N. MAGNETIC TAPE COPY.

TABLE 1V.- REGRESSION ANALYSIS RESULTS - 32 PSYCHOACOUSTIC TEST DATA

I.D.

A f SA 'B f SB c f sc IMPULSE DESCRIPTOR 0.08 0.04

-- I V I

I (b) I -4.01 0.36 I 0.44 0.16

(a) - 7.14 2.71 0.57 0.16 0.09 0.04

-

cF0.5 (b) -4.12 0.35 0.47 0.17 * CH47C PULSE RATE (f) AT 12.5 HZ A S = A + B (I.D.) + C.f r > 0.46 SIGNIFICANT AT 1 % LEVEL (a) PULSE REPETITION RATE TERM INCLUDED (b) PULSE REPETITION RATE TERM DROPPED TABLE V.- QUALITY CRITERIA OF DIFFERENT IMPULSIVENESS DESCRIPTORS A S = A + B (I.D.) + C.f 1 % LEVEL -r > 0.46 SIGNIFICANCE AT (a) PULSE REPETITION RATE TERM INCLUDED (b) PULSE REPETITION RATE TERM DROPPED TABLE V I . - COMPARISONS BETWEEN COMPUTED AND SUBJECTIVE IMPULSIVENESS

CORRECTION - TRANSIENT NOISE SIGNALS*

FREQ. AC EPNdB B.B.N.

TYPE OF HELICOPTER

N" Y (ISO) JURY

214 S.61 (REF.) - 115 K t - LEVEL

17 I 2.3 I 1.8 1 0.8 I 0

215 S.64 - 60 Kt - LEVEL 2.3 18.6 2.5 1.6 2.7

216 CH47.C. - 150 Kt - LEVEL 3.7 12.5 3.3 4.4 7 .O

217 CH47.C. - 60 Kt - LEVEL 3.6 12.5 3.8 4.4 5.5

218 B.212 - 105 Kt - LEVEL 2.9 3.3 3.2

2.3 11

219 8.212 - 61 Kt - LEVEL 3.4 3.1

4.5 11 3.3 3.5

220 47.0. - 6" APPR. 3.8 12 3.4 3.5

221 S.61 - 6 " APPR. 2.9 17 2.7 3.0 3.8

222 B.206 L - 6" APPR. 5.4 13 3.3 3.4 3.6

I I I * B.B.N..STUDlES, METHOD OF ADJUSTMENT.

ANNOYANCE DUE TO SIMULATED BLADE-SLAP NOISE Clemans A. Powell NASA Langley Research Center SUMMARY A study conducted a t t h e NASA Langley Research Center w a s previously reported i n which t h e e f f e c t s of several c h a r a c t e r i s t i c s of blade-slap n o i s e on annoyance response w e r e studied concurrently. These c h a r a c t e r i s t i c s o r parameters w e r e t h e sound p r e s s u r e level of t h e continuous n o i s e used t o simulate h e l i c o p t e r broadband n o i s e , t h e r a t i o of impulse peak t o broadband noise o r crest f a c t o r , t h e number of presscire excursions comprising an impulse event, t h e rise and f a l l t i m e of t h e i n d i v i d u a l impulses, and t h e r e p e t i t i o n frequency of t h e impulses. Forty s u b j e c t s made repeated judgments on a set of 36 n o i s e s t i m u l i which included 32 simulated h e l i c o p t e r blade-slap n o i s e s characterized by t h e above f i v e parameters and f o u r nonimpulsive broadband noises. Each parameter w a s found t o have a s i g n i f i c a n t e f f e c t on t h e annoyance judgments.

I n t h e p r e s e n t study, a d d i t i o n a l analyses w e r e conducted t o determine t h e c o r r e l a t i o n between s u b j e c t i v e response and v a r i o u s physical measures f o r t h e range of parameters studied. A s m a l l but s i g n i f i c a n t improvement i n t h e f a c t o r c o r r e c t i o n .

p r e d i c t i v e a b i l i t y of PNL w a s provided by an A-weighted crest No s i g n i f i c a n t improvement i n p r e d i c t i v e a b i l i t y w a s provided by a rate correction.

INTRODUCTION Human r e a c t i o n t o h e l i c o p t e r n o i s e , i n g e n e r a l , cannot be q u a n t i f i e d o r predicted as w e l l as t h e n o i s e from conventional take-off and landing a i r c r a f t .

It is generally agreed t h a t t h e discrepancy i n p r e d i c t i o n , u s u a l l y an under- estimation of annoyance response, is caused by f a c t o r s associated with t h e p u l s a t i v e n a t u r e of h e l i c o p t e r noise. Depending on t h e p a r t i c u l a r h e l i c o p t e r and f l i g h t conditions, t h e impulsiveness of h e l i c o p t e r n o i s e can range from marginally p e r c e p t i b l e modulation t o severe r e p e t i t i v e bands o r slapping sounds.

Because of t h e underestimation of annoyance of h e l i c o p t e r n o i s e by some researchers have suggested various a i r c r a f t noise-rating s c a l e s , modifying t h e noise-rating scales o r adding an impulse n o i s e c o r r e c t i o n .

Although considerable research has been conducted t o determine t h e a p p r o p r i a t e modifications o r c o r r e c t i o n s , t h e s e e f f o r t s have been generally unsuccessful o r inconclusive. F i e l d annoyance s t u d i e s s u f f e r from a l a c k of c o n t r o l over t h e physical parameters a f f e c t i n g t h e i n t e n s i t y of blade s l a p . It is g e n e r a l l y not p o s s i b l e t o s e p a r a t e t h e s u b j e c t i v e e f f e c t s of changes i n blade s l a p from s u b j e c t i v e e f f e c t s of changes i n o t h e r a c o u s t i c a l parameters which r e s u l t from using d i f f e r e n t h e l i c o p t e r types o r t h e same type under d i f f e r e n t operating conditions. Laboratory annoyance s t u d i e s using recordings of a c t u a l h e l i c o p t e r n o i s e s , while s u f f e r i n g from a similar confounding of e f f e c t s , a l s o s u f f e r from inadequacy of reproduction of t h e complex and phase s e n s i t i v e t i m e h i s t o r i e s of h e l i c o p t e r f l y o v e r noise. To s o l v e many of t h e problems associated w i t h s u b j e c t i v e tests using a c t u a l h e l i c o p t e r n o i s e s o r recordings, some r e s e a r c h e r s have r e s o r t e d t o using simulations of h e l i c o p t e r n o i s e s .

Most of t h e s e s t u d i e s , however, have been confined t o t e s t i n g only one out of many c h a r a c t e r i s t i c s of blade-slap n o i s e which could be responsible f o r t h e reported d i s c r e p a n c i e s i n p r e d i c t i o n of annoyance response t o such noises.

The study described i n t h i s paper w a s conducted t o examine t h e s u b j e c t i v e e f f e c t s of s e v e r a l c h a r a c t e r i s t i c s of r e p e t i t i v e impulse noise. Five v a r i a - b l e s were chosen t o c h a r a c t e r i z e h e l i c o p t e r b l a d e s l a p and t h e s e c h a r a c t e r i s t i c s o r parameters w e r e v a r i e d concurrently t o i n v e s t i g a t e p o s s i b l e i n t e r a c t i v e e f f e c t s . Human s u b j e c t s l i s t e n e d t o and r a t e d the annoyance of s h o r t b u r s t s of t h e simulated blade-slap noises. Some r e s u l t s of t h i s study have been previously reported i n reference 1, which i n d i c a t e d t h a t each of t h e parameters had a s i g n i f i c a n t e f f e c t on annoyance response. Additional analyses have been conducted which i n d i c a t e d t h a t v a r i o u s o b j e c t i v e measures The such as PNL and LA were a l s o s e n s i t i v e t o changes i n t h e parameters.

r e s u l t s of t h e s e analyses and t h e c o r r e l a t i o n between s u b j e c t i v e and o b j e c t i v e measures are reported herein. Comparisons are made between t h e r e s u l t s of t h i s experiment and a study conducted by Boeing Vertol ( r e f . 2 ) .

DESIGN AND PRELIMINARY RESULTS A d e t a i l e d d e s c r i p t i o n of t h e experimental design, procedures, and equipment used i n t h e experiments is given i n reference 1. The following paragraphs w i l l summarize the design and present t h e preliminary r e s u l t s as presented i n t h a t reference.

Experimental Design The following f i v e parameters w e r e chosen t o c h a r a c t e r i z e h e l i c o p t e r blade-slap n o i s e c o n s i s t l n g of a series of repeated impulses upon a continuous noise: 1 . The sound pressure of t h e continuous n o i s e used t o simulate h e l i c o p t e r broadband noise.

2. The r a t i o of impulse peak t o broadband n o i s e sound pressure levels ( i d e a l i z e d crest f a c t o r ) .

3 . The number of p r e s s u r e excursions making one complete impulse, i d e a l l y t h e number of s i n e waves i n a s i n g l e impulse.

4 .

The frequency of s i n e waves used t o synthesize t h e i n d i v i d u a l impulses.

The r e p e t i t i o n rate of t h e impulses.

5.

A set of 32 simulated blade-slap n o i s e s w a s c r e a t e d which included each of t h e s e parameters i n a high o r low condition i n t h e manner of a Z5 f a c t o r i a l design. The high and low conditions f o r each parameter are given i n t a b l e I.

I n a d d i t i o n t o t h e impulsive n o i s e s , f o u r samples of the nonimpulsive, broad- band n o i s e were included i n the set €or judgments by t h e subjects. The 36 noise s t i m u l i w e r e randomly ordered i n t o f o u r groups of n i n e s t i m u l i each.

The order of p r e s e n t a t i o n of the s t i m u l i groups w a s counterbalanced between groups of test s u b j e c t s .

Special precautions w e r e taken t o reduce t h e influence of room r e f l e c - t i o n s and t o i n s u r e t h a t t h e s u b j e c t s experienced t h e d e s i r e d waveforms.

Sound-absorbing panels which can be seen i n f i g u r e 1 w e r e used t o reduce room r e f l e c t i o n s . The impulsive and continuous portions of the s t i m u l i w e r e synthesized and recorded on s e p a r a t e channels of a s t e r e o t a p e recorder.

A s p e c i a l l y modified low-frequency loudspeaker w a s used t o reproduce t h e impulsive waveforms. During s t i m u l i preparation, t h e impulsive s i g n a l s w e r e monitored a t t h e test s u b j e c t ' s head l o c a t i o n and t h e recorded s i g n a l s modified t o reproduce t h e waveforms c a l l e d f o r i n t h e experimental design.

The s t i m u l i heard by t h e s u b j e c t s w e r e constant level 10-sec b u r s t s of n o i s e sec on ramp and o f f ramp.

with 0.5 Twenty male and twenty female s u b j e c t s made judgments on each of t h e complete sets of n o i s e s t i m u l i and a complete r e p l i c a t i o n of t h e s t i m u l i .

Each judgment w a s made on a continuous numerical scale from 0 t o 9 , from "no 'I annoyance" t o "maximum annoyance.

Data Analysis and Results The 2560 annoyance judgments made on t h e impulsive n o i s e s were analyzed is using an a n a l y s i s of variance procedure, an abbreviated version of which presented i n t a b l e 11. Each of t h e f i v e parameters w a s found t o have a s i g n i f i c a n t e f f e c t at t h e 0.01 level on t h e annoyance response of the impulsive blade-slap noises. Figure 2 i l l u s t r a t e s the magnitude and d i r e c t i o n of t h e e f f e c t of each of t h e f i v e parameters on t h e mean annoyance response.

For example, t h e mean annoyance r a t i n g f o r t h e impulsive n o i s e s with one s i n e wave p e r impulse w a s less than t h e mean annoyance r a t i n g f o r t h e impulsive waves p e r impulse. From t h i s f i g u r e , it can be seen n o i s e s with t h r e e s i n e t h a t t h e level of continuous n o i s e and t h e idealized crest f a c t o r had l a r g e , p o s i t i v e e f f e c t s on mean anndyance. The number of s i n e waves, t h e frequency waves, and t h e r e p e t i t i o n frequency had much smaller, p o s i t i v e e f f e c t s of s i n e although each w a s s t a t i s t i c a l l y s i g n i f i c a n t .

These f i n d i n g s , although of academic i n t e r e s t , do n o t resolve t h e question of whether o r not the present noise-rating scales underestimate t h e annoyance p o t e n t i a l of impulsive n o i s e s as compared with nonimpulsive noises.

46 5 To provide some information on t h i s question, t h e author of r e f e r e n c e 1 performed c o r r e l a t i o n analyses between t h e s u b j e c t i v e r a t i n g s of both t h e impulsive and nonimpulsive n o i s e s and various noise-rating scales.

These analyses i n d i c a t e d t h a t t h e perceived n o i s e level s c a l e underestimated by about 2 dB t h e annoyance p o t e n t i a l of t h e impulsive noises.

The next s e c t i o n of t h i s r e p o r t w i l l present the r e s u l t s of a d d i t i o n a l analyses which were performed on t h e d a t a from r e f e r e n c e 1 t o determine whether o r not t h i s underestimation of t h e annoyance p o t e n t i a l of t h e impulsive n o i s e s w a s r e l a t e d i n any systematic way with t h e f i v e parameters varied i n t h e experiment.

ADDITIONAL ANALYSES AND RESULTS Analyses The f i r s t s t e p of a d d i t i o n a l analyses was t o determine which of t h e noise- r a t i n g scales examined i n t h e experiment provided t h e b e s t o v e r a l l c o r r e l a t i o n w i t h t h e mean response d a t a f o r each noise condition, Linear least square regression analyses w e r e performed with t h e mean response d a t a as t h e dependent v a r i a b l e and with t h e physically measured d a t a f o r each r a t i n g scale.

as independent v a r i a b l e s . The c o r r e l a t i o n s i n terms of t h e Pearson product moment c o r r e l a t i o n c o e f f i c i e n t s f o r t h e mean response and each r a t i n g s c a l e are presented i n t a b l e 111. I n a d d i t i o n , t h e c o r r e l a t i o n s between t h e various r a t i n g s c a l e s are a l s o presented. The mean d a t a were obviously highly c o r r e l a t e d with t h e measured values of each r a t i n g scale, as were t h e measured values between r a t i n g s c a l e s .

Because of t h e high c o r r e l a t i o n between r a t i n g scales, t h e d i f f e r e n c e s i n c o r r e l a t i o n f o r t h e d i f f e r e n t r a t i n g scales with t h e mean response are not s i g n i f i c a n t . However, s i n c e PNL w a s more highly c o r r e l a t e d than t h e o t h e r r a t i n g scales and s i n c e it forms t h e b a s i c measure f o r t h e accepted standard measure (EPNL) f o r conventional a i r c r a f t n o i s e , t h e f u r t h e r analyses w e r e conducted using PNL as t h e primary physical measure.

The r e s u l t s of t h e regression a n a l y s i s of t h e mean response on t h e PNL values f o r t h e s t i m u l i are presented i n f i g u r e 3. The nonimpulsive n o i s e s t i m u l i are represented by t h e s o l i d c i r c u l a r symbols and t h e impulsive n o i s e s t i m u l i by t h e open c i r c u l a r symbols. The least squares l i n e a r regression f o r , t h e s e p o i n t s is indicated by t h e s o l i d l i n e . A s is t y p i c a l f o r scale, t h e r e appears t o be some s l i g h t curvature i n trend t h i s type judgment of t h e d a t a p o i n t s at t h e ends of t h e range. I n order t o reduce t h i s nonlinear t h e following procedure was used t o convert t h e mean s u b j e c t i v e d a t a behavior, i n t o s u b j e c t i v e l y equivalent n o i s e l e v e l s f o r f u r t h e r comparison between t h e various n o i s e conditions. A polynomial regression w a s performed i n t h e form

Xi = a + byi + cyi 2 + dyi 3

where X . is t h e PNL value f o r t h e i t h stimulus and y is t h e corresponding i mean s u 6 j e c t i v e response. The r e s u l t i n g b e s t f i t regression w a s found t o be 2 3

+ 0 . 0 4 4 ~

X = 64.76 + 6 . 8 7 4 ~ - 0 . 6 7 0 ~

The predicted o r s u b j e c t i v e l y equivalent n o i s e level f o r each stimulus w a s calculated by s u b s t i t u t i n g t h e r e s p e c t i v e mean response value i n t o t h e regression r e l a t i o n s h i p . For f u r t h e r discussion, the s u b j e c t i v e l y equivalent n o i s e l e v e l s w i l l be designated simply as equivalent levels. The equivalent level (Eq.L.) of each stimulus is p l o t t e d i n f i g u r e 4 a g a i n s t t h e r e s p e c t i v e measured PNL values. A c l o s e comparison of the d a t a i n f i g u r e s 3 and 4 i n d i c a t e s t h e improvement i n l i n e a r i t y between t h e s u b j e c t i v e response and n o i s e level i n PNL.

i n annoyance between t h e impulsive n o i s e s t i m u l i and t h e The d i f f e r e n c e simulated h e l i c o p t e r broadband n o i s e w a s continuous n o i s e which served as determined by s u b t r a c t i n g t h e equivalent levels of t h e nonimpulsive n o i s e t h e r e s p e c t i v e impulsive n o i s e s t i m u l i . The values s t i m u l i from those of (AEq.L.) represent t h e i n c r e a s e i n annoyance due t o t h e a d d i t i o n t h u s obtained Similarly, t h e of t h e impulsive n o i s e on t h e continuous background noise.

d i f f e r e n c e i n t h e PNL values of t h e impulsive noise s t i m u l i and t h e respective nonimpulsive n o i s e s t i m u l i (APNL) r e p r e s e n t s t h e i n c r e a s e i n PNL a t t r i b u t e d t o t h e a d d i t i o n of t h e impulsive noise. A comparison of t h e s e two sets of values is presented i n f i g u r e 5. The open symbols represent those d a t a w i t h a continuous n o i s e level of 65 dB (OASPL) and t h e s o l i d symbols represent those with a continuous l e v e l of 80 dB (OASPL). .From t h i s f i g u r e , it can be seen t h a t , i n general, t h e a d d i t i o n of t h e impulsive n o i s e produced i n PNL.

a g r e a t e r i n c r e a s e i n annoyance than w a s accounted f o r by t h e i n c r e a s e The excess annoyance did not appear t o be s t r o n g l y r e l a t e d t o t h e level of continuous noise.

The same d a t a are reproduced i n f i g u r e s 6 t o 9 with t h e o t h e r f a c t o r s of t h e experimental design as separable parameters. The d a t a i n f i g u r e 6 are separated by t h e d i f f e r e n t symbols i n t o t h e conditions of high and low i d e a l i z e d crest f a c t o r . Although t h e d a t a are c l e a r l y grouped by t h i s parameter, t h e change i n PNL m i r r o r s t h e change i n e f f e c t i v e n o i s e l e v e l equally as w e l l f o r t h e high i d e a l i z e d c r e s t f a c t o r as f o r t h e low i d e a l i z e d crest f a c t o r conditions. I n f i g u r e 7, t h e d a t a are separated by t h e r e p e t i t i o n rate of t h e impulses. N o clear s e p a r a t i o n of t h e d a t a i s provided by t h e r e p e t i t i o n r a t e , f a c t o r . I n f i g u r e 8, t h e d a t a are separated by t h e number of s i n e waves i n t h e impulse events. Based on the g r e a t e r number of d a t a p o i n t s below t h e l i n e of e q u a l i t y f o r t h e 3-sine wave condition as compared w i t h t h e 1-sine wave condition, t h e r e appears t o be some r e l a t i o n s h i p between t h e annoyance of t h e impulsive n o i s e s and t h e number of s i n e waves t h a t is not accounted f o r by PNL. Figure 9 p r e s e n t s t h e d a t a separated by t h e frequency of s i n e waves i n t h e impulse events. There appears t o b e no c o n s i s t e n t e f f e c t of t h e frequency of s i n e waves on the i n c r e a s e i n annoyance due t o impulsiveness which i s n o t accounted f o r by a change i n PNL.

t h e v a r i o u s f a c t o r s I n o r d e r t o more a c c u r a t e l y quantify t h e e f f e c t s of of t h e experiment, a c o r r e l a t i o n a n a l y s i s w a s performed between t h e f a c t o r s , s u b j e c t i v e measures, and o b j e c t i v e measures previously described. Two a d d i t i o n a l c o r r e l a t e s w e r e considered i n t h e a n a l y s i s and are defined as follows. The underestimation of PNL t o account f o r t h e s u b j e c t i v e d i f f e r e n c e s between t h e impulsive and nonimpulsive n o i s e annoyance w a s defined as

A S = AEq.L. - APNL

where AEq.L. w a s t h e d i f f e r e n c e between t h e equivalent level f o r t h e impulsive and nonimpulsive n o i s e s and APNL w a s the d i f f e r e n c e between the perceived n o i s e levels of t h e impulsive and nonimpulsive noises. A n A-weighted impulsive c o r r e c t i o n w a s defined as

ACFA = LA(peak) - L A ( r m s ) - 1 2

The c o r r e l a t i o n matrix f o r t h e s u b j e c t i v e measures, o b j e c t i v e measures, and experimental f a c t o r s i s presented i n t a b l e I V .

The high c o r r e l a t i o n of t h e e f f e c t i v e level with PNL is i n d i c a t i v e t h a t , i n general, PNL predicted t h e s u b j e c t i v e response very w e l l , t h e unexplained e r r o r being only 4 percent of the t o t a l v a r i a t i o n i n s u b j e c t i v e response over a wide range (28 PNdB) of n o i s e l e v e l s . The standard e r r o r of estimate using PNL as a p r e d i c t o r of e f f e c t i v e n o i s e level w a s 1.72 dB. The only experimental f a c t o r which w a s found t o be s i g n i f i c a n t l y c o r r e l a t e d with t h e equivalent level w a s t h e i d e a l i z e d crest f a c t o r . The i d e a l i z e d crest f a c t o r , however, w a s a l s o found t o be s i g n i f i c a n t l y c o r r e l a t e d with PNL t o approximately t h e same degree.

S i m i l a r l y , t h e change i n equivalent l e v e l between t h e impulsive and found t o be s i g n i f i c a n t l y c o r r e l a t e d with t h e change nonimpulsive n o i s e s w a s i n PNL and t h e i d e a l i z e d crest f a c t o r . Again, however, t h e i d e a l i z e d crest f a c t o r w a s found t o be s i g n i f i c a n t l y c o r r e l a t e d with a change i n PNL.

The d i f f e r e n c e (AS) between t h e change i n equivalent level and t h e change i n PNL w a s found t o be s i g n i f i c a n t l y c o r r e l a t e d w i t h t h e crest f a c t o r c o r r e c t i o n but not with t h e i d e a l i z e d crest f a c t o r . There w a s , however, a s i g n i f i c a n t n e g a t i v e c o r r e l a t i o n of A S and t h e number of s i n e waves comprising t h e impulse events. Q u a l i t a t i v e i n d i c a t i o n s of t h i s trend w e r e presented i n f i g u r e 8 and i n previous discussions. The number of s i n e waves w a s a l s o s u f f i c i e n t l y and negatively c o r r e l a t e d with ACFA s o t h a t it is doubtful t h a t any improvement i n p r e d i c t i o n beyond t h a t afforded by ACFA would be r e a l i z e d .

Least square r e g r e s s i o n analyses ( f i g . 10) w e r e performed w i t h t h e underestimation of PNL f o r impulsive n o i s e s A S as t h e dependent v a r i a b l e and t h e impulsive c o r r e c t i o n ACF as the independent v a r i a b l e . The regression A equation thus obtained w a s A S = -0.04 + .400 X A C F , The standard e r r o r of estimate (SEE) f o r t h e regression w a s 1.52 dB. It should be pointed o u t , however, t h a t t h i s value is only 0.2 dB improvement i n t h e p r e d i c t i v e a b i l i t y of PNL with no impulsive c o r r e c t i o n .

There has been r e c e n t evidence ( r e f . 3) t h a t t h e rate of t h e impulse events c o r r e l a t e s equally as w e l l with t h e underestimate of PNL o r EPNL as does v a r i o u s impulsive corrections. This trend has not been confirmed with t h e r e s u l t s of t h e present experiment, Comparison With Other Research I n a recent experiment conducted by Boeing Vertol and reported i n reference 2 , s u b j e c t s adjusted t h e impulsiveness of simulated blade-slap n o i s e s u n t i l they w e r e as equally annoying as continuous n o i s e s with s p e c t r a simulative of h e l i c o p t e r broadband noise. The impulsive n o i s e s w e r e presented simultaneously with broadband n o i s e with t h e same spectra as t h e reference n o i s e s but a t a lower f i x e d level. The s u b j e c t s ' t a s k w a s t o vary t h e level of t h e impulsive p o r t i o n of t h e test s t i m u l i t o match t h e annoyance of t h e r e f e r e n c e s t i m u l i . The experiment was f a c t o r i a l i n design and consisted of 108 p a i r s of s t i m u l i comprised of t h r e e d i f f e r e n t broadband spectra, t h r e e l e v e l s of reference broadband n o i s e , t h r e e impulsiveness conditions, and f o u r impulse r e p e t i t i o n rates. A t t h e completion of each adjustment f o r e q u a l i t y of annoyance, t h e l e v e l of t h e impulsive n o i s e s i n terms of v a r i o u s physical measures w a s recorded. The average of t h e s e l e v e l s over s u b j e c t s provided measures of t h e level f o r equal annoyance f o r t h e impulsive s t i m u l i . The d i f f e r e n c e i n l e v e l between t h e reference broadband s t i m u l i and t h e t e s t impulsive s t i m u l i at t h e point of e q u a l i t y thereby represented t h e underestimation of t h e physical measure, Regression analyses w e r e performed with the underestimation ( i n terms of PNL) as t h e dependent rate as independent v a r i a b l e s . S i g n i f i c a n t v a r i a b l e and with ACFA and c o r r e l a t i o n w a s found only f o r t h e c r e s t f a c t o r c o r r e c t i o n . The r e l a t i o n s h i p w a s found t o be

A S = -3.37 + 0.113ACFA

with a c o r r e l a t i o n c o e f f i c i e n t of 0.265, which f o r 106 degrees of freedom i s s i g n i f i c a n t a t t h e 0.99 l e v e l . The standard e r r o r of estimate w a s 2.65 dB.

Although a s i g n i f i c a n t dependence w a s found on t h e A-weighted crest f a c t o r c o r r e c t i o n , t h e slope f o r t h e dependence w a s considerably less than w a s found i n t h e NASA experiment.

7 469 One p o s s i b l e reason f o r t h e d i f f e r e n c e s i n r e s u l t s could be t h e d i f f e r e n c e s i n t h e manner of p r e s e n t a t i o n of t h e n o i s e s t i m u l i . The s t i m u l i f o r t h e NASA experiment were presented via loudspeaker whereas those f o r t h e Boeing Vertol experiment w e r e presented over headphones. The d i f f e r e n c e s i n r e s u l t s , thereby, could have been t h e r e s u l t of t h e d i f f e r e n c e i n whole-body response and a u d i t o r y response.

CONCLUDING REMARKS Additional analyses have been conducted on d a t a obtained from a previously reported experiment which w a s conducted t o s y s t e m a t i c a l l y i n v e s t i g a t e t h e e f f e c t s of v a r i o u s parameters of h e l i c o p t e r blade-slap n o i s e . Five, parameters w e r e chosen t o synthesize blade-slap noise.

These w e r e t h e sound pressure of t h e continuous broadband n o i s e , l e v e l t h e i d e a l i z e d crest f a c t o r of t h e impulses above t h e continuous n o i s e , t h e number of s i n e waves i n a s i n g l e impulse, t h e frequency of t h e s i n e waves, and t h e impulse r e p e t i t i o n frequency. Forty s u b j e c t s judged t h e annoyance of each of t h e noises.

Although each of t h e parameters w a s found t o have a p o s i t i v e and s i g n i f i c a n t e f f e c t on judged annoyance, each parameter w a s found i n t h e a n a l y s i s reported h e r e i n t o produce a s i m i l a r change i n measured noise l e v e l i n t e r m s of PNL.

A s l i g h t b u t s i g n i f i c a n t improvement i n t h e p r e d i c t i v e a b i l i t y of PNL N o w a s provided by t h e a d d i t i o n of an A-weighted crest f a c t o r c o r r e c t i o n .

s i g n i f i c a n t improvement w a s provided by t h e a d d i t i o n of a c o r r e c t i o n proportional t o t h e rate of impulses.

Further a n a l y s i s of a r e c e n t experiment conducted by Boeing Vertol under N A S A c o n t r a c t indicated a similar l a c k of need f o r a rate c o r r e c t i o n .

Results from t h i s experiment, however, i n d i c a t e d a s i g n i f i c a n t but smaller crest f a c t o r c o r r e c t i o n than w a s indicated i n t h e N A S A experiment.

REFERENCES 1. Lawton, Ben W i l l i a m : Subjective Assessment of Simulated Helicopter Blade-Slap Noise. N A S A TN D-8359, 1977.

2. S t e r n f e l d , Harry, Jr.; and Doyle, Linda Bukowski: Evaluation of t h e Annoyance Due t o Helicopter Rotor Noise. NASA CR-3001, 1978.

3. Galloway, W i l l i a m J.: Subjective Evaluation of Helicopter Blade-Slap Noise. Helicopter Acoustics, NASA CP-2052, P t . 11, 1978. (Paper no. 20 of t h i s compilation.)

Value of Parameter Parameter Low High 1 3 Number of s i n e waves i n impulse . . . . . . . . . .

200 400 Sine wave frequency, Hz . . . . . . . . . . . . . .

8 20 Repetition frequency of impulses, Hz. . . . . . . .

65 80 Level of continuous n o i s e , dBa. . . . . . . . . . .

I d e a l i z e d . c r e s t f a c t o r b of impulsive noise, dB. . , 15 25

Peak pressure Crest f a c t o r = rms p r e s s u r e When converted t o dB scale, crest f a c t o r is peak SPL minus r m s SPL. For purposes of d e f i n i n g n o i s e s used i n t h i s study, an i d e a l i z e d crest f a c t o r rms SPL of continuous noise.

w a s s p e c i f i e d , peak SPL of impulses minus TABLE 11.- RESULTS O F ABBREVIATED ANALYSIS O F VARIANCE Frequency of s i n e waves Impulse r e p e t it i o n frequency Level of continuous n o i s e Idealized crest f a c t o r a These F-ratio v a l u e s are s i g n i f i c a n t a t 0.01 level. For one and i n f i n i t e degrees of freedom at t h i s level, t h e critical F-value equals 6.63.

TABLE 111.- CORRELATION MATRIX OF MEAN SUBJECTIVE RESPONSE AND PHYSICAL DESCRIPTORS OASPL, OASPL, LA, LA 3 LA, PNL rms peak r m s peak impulse M e a n response 0.965 0.972 0.976 0.954 0.966 0.978 OASPL, rms .975 ,976 921 .964 .990 OASPL, peak .974 .969 * 974 .977 ms .968 .993 .994 .925 .947 LA, peak LA, impulse .984 TABLE 1V.- CORRELATION MATRIX OF SUBJECTIVE MEASURES, OBJECTIVE MEASURES, AND EXPERIMENTAL FACTORS t Num- Freq- AS AEq.L. PNL APNL dCFA b e r uency f a c t o r a a 0.268 0.005 Eq.L. 547 0.370 980 0.076 0.417 0.079 0.107 AEq.L.

.202 .112 .499 .890 .180 ,244 a PNL .073 -.lo3 .457 ,372 .115 .166 .052 a APNL -.267 -.129 .383 .119 A S

b.513 -.065 - -441 -.206 .259

A C P , .203 b-. 522 b-. 454 b . 559 a C o r r e l a t i o n c o e f f i c i e n t s i g n i f i c a n t a t 0.05 level.

b c o r r e l a t i o n c o e f f i c i e n t s i g n i f i c a n t at o 01 level.

Figure 1.- Photograph of test chamber showing o r i e n t a t i o n of subject and loudspeakers.

LEVEL OF CONT I NUOUS NOISE, FREQUENCY OF dB MEAN S I N E WAVES, u7 ANNOYANCE 5 I l L

- RATING

3 . 7 4 4D9

E412 54.07’

-

2 . 0 1 0 ~~ 1 1 3 15 25 IMPULSE IDEALIZED NO. OF REPETITION CREST FACTOR, S I N E WAVES FREQUENCY, dB H Z Figure 2.- Annoyance e f f e c t s of f i v e parameters used t o synthesize impulsive test noises.

9- NOISE TYPE 0 IMPULSIVE 8- 0 NON IMPULS IVE 7- 6- MEAN

-

SUBJECTIVE 5 RESPONSE 4- 3- 2 - 1-

L I '

65 7b ; 5 d 0 8 ' 5 d 0 ds Id0

PNL, PNd6 Figure 3.- Mean subjective response to impulsive and nonimpulsive noises.

0 IMPULSIVE 0 NON IMPULS IVE EQU [VALENT LEVEL, 85 dB

SEE = 1.71 dB

I I 1 I 1 1 i d5 70 75 80 85 90 95 100 PNL, PNdB Figure 4.- Correlation of equivalent noise level with PNL for impulsive and nonimpulsive noises.

LEVEL OF CONTINUOUS NOISE 0 65 dB AEQUIVALENT LEVEL, 0 - dB O O I I I I I I 1 0 2 4 6 8 1 0 1 2 1 4 APNL, PNdB Figure 5.- Effects of impulsiveness for two levels of continuous noise.

I DEALJZED CREST FACTOR APNL, PNdB Figure 6 . - Effects of impulsiveness for two levels of idealized crest factor.

REPETITION RATE OF IMPULSES, SEC-'

1 4

A€QUIVALENT 8

LEVEL, dB 6

2 4 6 8 10 12 1 4

APNL, PNdB Figure 7.- Effects of impulsiveness for two repetition rates v of impulsive events.

NUMBER OF SINE WAVES 0 .

AEQUIVALENT LEVEL, dB APNL, PNdB ffects of impulsiveness for two numbers of sine waves comprising an impulse event.

FREQUENCY OF SINE WAVES A EQU I VALENT 8 LEVEL, dB 0 2 4 6 8 10 12 14 APNL, PNdB Figure 9.- Effects of impulsiveness for two frequencies of sine waves comprising an impulse event.

AS, dB 0 -1 ~~ -4

'[ -5 0

1 I 1 I I I -2 0 2 4 6 8 ACFA, dB Figure 10.- Underestimation of PNL for impulsive noises.

HUMAN RESPONSE TO AIRCRAFT-NOISE-INDUCED BUILDING VIBRATION Jimmy M. Cawthorn, Thomas K. Dempsey, and Richard DeLoach N A S A Langley Research Center SUMMARY A series of p i l o t s t u d i e s has been conducted i n both t h e f i e l d and t h e Laboratory t o i n v e s t i g a t e t h e e f f e c t s of noise-induced building s t r u c t u r e J i b r a t i o n and t h e r a t t l e of objects'on human response t o a i r c r a f t flyover noise. The f i e l d s t u d i e s w e r e conducted i n a c t u a l a i r p o r t communities, md t h e s p e c i f i c o b j e c t i v e s included t h e determination of s u b j e c t i v e d e t e c t i o n thresholds f o r v i b r a t i o n and rattle, as w e l l as t h e e f f e c t of v i b r a t i o n and rattle upon a i r c r a f t n o i s e annoyance. The laboratory study, conducted a t Langley Research Center, w a s concerned primarily with t h e f a c t o r of rattle.

Ilhe s p e c i f i c o b j e c t i v e s included t h e determination of r a t t l e d e t e c t i o n thres- holds, r a t t l e annoyance t h r e s h o l d s , and t h e e f f e c t of r a t t l e on t h e o v e r a l l annoyance response t o an a i r c r a f t flyover n o i s e event.

A s a r e s u l t of t h e s e s t u d i e s , t h e v i b r a t i o n d e t e c t i o n threshold w a s determined and building s t r u c t u r a l v i b r a t i o n w a s found t o i n c r e a s e t h e annoyance The rattle of o b j e c t s w a s response produced by a n a i r c r a f t n o i s e event.

observed very i n f r e q u e n t l y i n t h e f i e l d study and i n t h e laboratory study r a t t l e w a s found t o be of no s i g n i f i c a n c e and not important.

INTRODUCTION Airport community n o i s e surveys ( r e f . 1 ) and complaint records ( r e f . 2) have o f t e n highlighted building v i b r a t i o n s and a s s o c i a t e d r a t t l e of o b j e c t s within buildings as a source of annoyance t o r e s i d e n t s l i v i n g i n a i r p o r t communities. This v i b r a t i o n / r a t t l e may be a p o t e n t i a l detriment t o h e l i c o p t e r development and operations because of t h e low frequency and impulsive n a t u r e of h e l i c o p t e r n o i s e s i g n a l s , The impact of a i r c r a f t n o i s e ( i n c l u s i v e of h e l i c o p t e r noise) on people who are exposed t o t h e n o i s e while indoors can be i l l u s t r a t e d as shown schematically i n f i g u r e 1, A i r c r a f t operations generate a house and i s then transmitted n o i s e which impinges upon t h e e x t e r i o r of through t h e house s t r u c t u r e t o t h e i n t e r i o r where it is perceived by the r e s i d e n t . I n some cases, t h e n o i s e impingement and sound transmission process w i l l produce s t r u c t u r a l v i b r a t i o n and/or t h e r a t t l e of o b j e c t s within t h e home.

I f t h e magnitude of any (or a l l t h r e e ) of t h e s e physical s t i m u l i are above an i n d i v i d u a l ' s d e t e c t i o n threshold, it is l i k e l y t h a t t h e i n d i v i d u a l w i l l not only w i l l combine them i n some way t o produce a t o t a l annoyance perceive them but response. a community r e s i d e n t t o an air- The r e s u l t a n t annoyance response of c r a f t n o i s e is, t h e r e f o r e , dependent upon some type of i n t e g r a t i o n of t h e t h r e e physical s t i m u l i .

The n o i s e generation and propagation i n t o t h e i n t e r i o r of t h e house together with t h e associated building v i b r a t i o n and rattle c o n s t i t u t e t h e major physical aspects of t h i s environment. The s u b j e c t i v e response t o t h e 4 79 physical environment, however, involves a determination of s e p a r a t e noise, v i b r a t i o n , and r a t t l e d e t e c t i o n thresholds and t h e manner i n which t h e s e s e p a r a t 9 physical f a c t o r s combine t o produce a t o t a l annoyance. This paper presents t h e r e s u l t s of a series of f i e l d and l a b o r a t o r y s t u d i e s t h a t were conducted t o (1) o b t a i n d e t a i l e d measurements of t h e n o i s e l v i b r a t i o n l r a t t l e environment, (2) o b t a i n s u b j e c t i v e annoyance responses t o t h e combined envi- ronment, and (3) attempt t o d e f i n e t h e psychophysical r e l a t i o n s h i p between t h e s u b j e c t i v e responses and t h e physical environment.

The f i e l d s t u d i e s w e r e conducted i n a c t u a l a i r p o r t communities. Spe- c i f i c o b j e c t i v e s of t h e community s t u d i e s included t h e determination of sub- jective d e t e c t i o n thresholds f o r v i b r a t i o n and r a t t l e , as w e l l as t h e e f f e c t of v i b r a t i o d r a t t l e upon a i r c r a f t n o i s e annoyance, The l a b o r a t o r y study at Langley Research Center w a s concerned primarily with i n v e s t i g a t i n g conducted t h e relative importance o r influence of t h e f a c t o r of rattle upon s u b j e c t i v e annoyance. The l a b o r a t o r y s e t t i n g w a s s e l e c t e d f o r t h e r a t t l e i n v e s t i g a t i o n s i n c e it is very d i f f i c u l t t o d e f i n e and measure r a t t l e i n t h e complex environ- ment of a f i e l d study. S p e c i f i c o b j e c t i v e s of t h e l a b o r a t o r y study w e r e t o determine, under c o n t r o l l e d conditions, t h e r a t t l e d e t e c t i o n threshold, r a t t l e annoyance threshold, and t h e e f f e c t of r a t t l e on a i r c r a f t n o i s e annoyance.

COMMUNITY STUDY Procedure The community study reported herein w a s conducted i n the communities surrounding John F. Kennedy I n t e r n a t i o n a l Airport (New York C i t y ) i n conjunction with t h e government’s assessment program of t h e Concorde supersonic t r a n s p o r t ( r e f . 2 ) . A number of homes i n t h e Kennedy Airport area w e r e u t i l i z e d f o r t h i s study, and t h e s u b j e c t s who p a r t i c i p a t e d included both r e s i d e n t s and members of t h e N A S A monitoring team. The test procedures can be b e s t described i n This photograph shows a t y p i c a l conjunction with t h e photograph of f i g u r e 2.

group of s u b j e c t s p a r t i c i p a t i n g i n t h e s u b j e c t i v e response tests as w e l l as some of t h e instrumentation used t o o b t a i n t h e physical n o i s e and v i b r a t i o n measurements. Microphones were located both indoors and outdoors (not shown) f o r recording t h e a i r c r a f t n o i s e levels. Accelerometers w e r e used t o measure t h e a c c e l e r a t i o n levels of t h e window, w a l l , and f l o o r ( v e r t i c a l and h o r i z o n t a l ) .

A l l t h e physical d a t a w e r e recorded on magnetic t a p e i n a mobile a c o u s t i c van (located o u t s i d e t h e residence) f o r later a n a l y s i s . Since t h i s w a s a community study u t i l i z i n g a i r c r a f t flyover events as they n a t u r a l l y occurred, no c o n t r o l over t h e sound sources w a s possible. Consequently, both s u b j e c t i v e r a t i n g s and physical measurements w e r e obtained f o r each flyover event t h a t occurred during a test s e s s i o n which nominally l a s t e d about 1 / 2 t o 1 hour a t each s i t e .

The technique used t o o b t a i n t h e s u b j e c t i v e response r a t i n g s f o r each flyover is i l l u s t r a t e d by t h e sample flyover event r a t i n g form of f i g u r e 3 .

Each flyover event w a s assigned a flyover number by t h e test d i r e c t o r , and t h e number w a s w r i t t e n on t h e r a t i n g form i n t h e a p p r o p r i a t e space by each s u b j e c t .

A t t h e conclusion of each event, t h e test d i r e c t o r i n s t r u c t e d t h e s u b j e c t s t o rate t h e flyover a t which t i m e t h e s u b j e c t s would i n d i c a t e on t h e i r r a t i n g form whether o r n o t they had detected v i b r a t i o n , r a t t l e , o r noise; whether o r not t h e v i b r a t i o n , rattle, o r n o i s e w a s annoying; and f i n a l l y a n o v e r a l l annoyance r a t i n g of t h e flyover on a numerical category scale which ranged from 0 t o 9 , where "0" w a s defined as "zero annoyance'! and "9" w a s defined as "maximum annoyance.'' Since many of t h e r e s i d e n t s u b j e c t s had d i f f i c u l t y i n d i f f e r e n t i a t i n g between noise, v i b r a t i o n , and rattle and i n properly using t h e r a t i n g form, only t h e d a t a from t h e N A S A "trained" s u b j e c t s of t h e assessment t e a m are u t i l i z e d i n t h i s paper.

T e s t R e s u l t s A t o t a l of 109 a i r c r a f t flyover events a t 8 houses w e r e experienced by a t o t a l of 16 test s u b j e c t s . With regard t o r a t t l e d e t e c t i o n , t h e s u b j e c t i v e r a t i n g s indicated t h a t on only t h r e e occasions d i d one half o r more of t h e s u b j e c t s d e t e c t rattle. This implies t h a t r a t t l e may e x e r t only a minor influence upon a person's annoyance response t o a i r c r a f t noise. However, due t o t h e s c a r c i t y of d a t a , it w a s determined t h a t a f i n a l conclusion with respect t o t h e importance of r a t t l e should a w a i t t h e r e s u l t s of a laboratory i n v e s t i g a t i o n i n w a s studied under c o n t r o l l e d conditions. These r e s u l t s are which t h i s f a c t o r presented later i n t h i s paper.

Vibration w a s d e t e c t e d by 50 percent o r more of t h e s u b j e c t s on 21 occasions a t 3 of t h e 8 houses. A l l t h r e e of t h e s e houses were located i n s i d e t h e 40 NEF contour, and each had conventional wooden f l o o r s above a crawl space.

The r e s u l t s of t h i s phase of t h e experiment are shown i n f i g u r e 4 i n which t h e s u b j e c t s d e t e c t i n g v i b r a t i o n i s p l o t t e d as a f u n c t i o n of t h e level of percent of v e r t i c a l f l o o r v i b r a t i o n i n dB u n i t s ( r e f . 1 pg). The threshold of v i b r a t i o n d e t e c t i o n i s defined as t h e l e v e l a t which 50 percent of t h e observers f e e l t h e v i b r a t i o n . Consequently, f o r t h e s e d a t a , t h e threshold of d e t e c t i o n is seen t o be i n t h e range of from 62 t o 68 dB, v e r t i c a l f l o o r a c c e l e r a t i o n . The range of 62 t o 68 dB corresponds approximately t o 100 t o 105 dB, o u t s i d e sound pressure l e v e l . Thus, it appears t h a t aircraft-generated o u t s i d e sound pressure levels g r e a t e r than 100 dB are capable of inducing v i b r a t i o n s of a magnitude s u f f i c i e n t t o exceed t h e threshold of v i b r a t i o n d e t e c t i o n of t h e occupants within.

Figure 5 p r e s e n t s t h e r e s u l t s of t h e category s c a l i n g experiment of t h e o v e r a l l annoyance r a t i n g of t h e flyover events. Average annoyance r a t i n g s are shown as a function of o u t s i d e A-weighted sound pressure level f o r two c a t e g o r i e s of events, namely f o r a i r c r a f t f l y o v e r s i n which t h e threshold of v i b r a t i o n d e t e c t i o n w a s achieved, and f o r those a i r c r a f t flyover events f o r which t h e threshold of v i b r a t i o n d e t e c t i o n w a s n o t achieved. The l i n e s shown i n t h e f i g u r e were drawn based on a least-square l i n e a r f i t of t h e two sets of d a t a . The f i g u r e and "paired t-tests" (based on, t h e a c t u a l d a t a of one curve v e r s u s t h e predicted d a t a of t h e o t h e r curve) show t h a t a i r c r a f t f l y o v e r s f o r which t h e r e w a s v i b r a t i o n d e t e c t i o n were evaluated as s i g n i f i c a n t l y more annoying than a i r c r a f t f l y o v e r s f o r which Vibration w a s n o t detected.

An implication of t h e s e d a t a is t h a t s t r u c t u r a l building v i b r a t i o n does have a s i g n i f i c a n t and detrimental e f f e c t on t h e annoyance response of people t o a i r c r a f t flyover noise. Since most s t u d i e s of t h e e f f e c t s of a i r c r a f t n o i s e on people have n o t considered building v i b r a t i o n , t h e s e r e s u l t s may e x p l a i n some of the s u b j e c t i v e response v a r i a t i o n t h a t occurs w i t h i n and between v a r i o u s s t u d i e s , LABORATORY STUDIES Rattle Detection and Annoyance Thresholds The l a h o r a t o r y study of r a t t l e d e t e c t i o n , r a t t l e annoyance, and r a t t l e e f f e c t s on a i r c r a f t n o i s e annoyance w a s conducted i n t h e Langley a i r c r a f t The f a c i l i t y used i n t h e study w a s t h e i n t e r i o r n o i s e reduction laboratory.

e f f e c t s room (IER) which is shown i n f i g u r e 6. This room i s configured t o resemble a t y p i c a l r e s i d e n t i a l l i v i n g room, and its c o n s t r u c t i o n i s considered r e p r e s e n t a t i v e of t h a t found i n a standard r e s i d e n t i a l house. It c o n s i s t s of painted dry w a l l over 50.8-mm by 101.6-mm (2 by 4) s t u d s on 406.4-mm (16 i n ) c e n t e r s . The dimensions of t h e room are approximately 4 by 6 by 2.5 meters. Noise s t i m u l i are presented i n t h e room by means of f o u r loud- speakers which are located o u t s i d e and above each corner of t h e room.

Experimental design.- The experimental design of t h e r a t t l e d e t e c t i o n and t h e r a t t l e annoyance experiments involved p r e s e n t a t i o n of a n a i r c r a f t n o i s e a t a constant level with varying amounts of r a t t l e . A s shown i n f i g u r e 7 , t h e peak A-weighted sound pressure level of t h e a i r c r a f t flyover n o i s e w a s held constant a t 7 1 dB while t h e rattle l e v e l w a s v a r i e d from 45 t o 6 1 dB(A) (measured a t t h e s u b j e c t s ' s e a t i n g p o s i t i o n s ) . For c o n t r o l and r e p e a t a b i l i t y , t h e r a t t l e level w a s produced by tape recording t h e sound of drinking g l a s s e s r a t t l i n g when e x c i t e d by a n electromechanical shaker mounted t o a china c a b i n e t containing t h e g l a s s e s and which w a s d r i v e n by a tape recording of t h e a i r c r a f t flyover. For playback, t h e tape recording of t h e rattle w a s synchronized with t h e a i r c r a f t flyover n o i s e and w a s introduced i n t o t h e IER through a small loudspeaker located under a china cabinet while t h e a i r c r a f t n o i s e w a s played through t h e overhead, o u t s i d e speakers. The level of t h e r a t t l e n o i s e w a s adjusted by changing t h e g a i n s e t t i n g of t h e t a p e recorder playback.

A t o t a l of 24 paid, volunteer s u b j e c t s p a r t i c i p a t e d i n t h i s experiment.

Each n o i s e and r a t t l e level combination w a s randomized and repeated once and each s u b j e c t heard every combination. A s f u r t h e r shown i n f i g u r e 7 , t h e determination of d e t e c t i o n and annoyance thresholds w a s addressed i n two tasks. For d e t e c t i o n , t h e s u b j e c t s w e r e asked t o rate whether o r n o t s e p a r a t e they detected t h e o b j e c t s i n t h e china c a b i n e t r a t t l i n g . For t h e annoyance threshold t a s k , s u b j e c t s were asked t o rate whether o r not t h e r a t t l i n g sounds they heard w e r e annoying. I n e i t h e r case, t h e expected r e s u l t s would be a n increasing number of yes responses with increasing r a t t l e level.

The thresholds f o r d e t e c t i o n and f o r annoyance are defined as t h e level f o r which 50 percent of t h e s u b j e c t s d e t e c t e d t h e rattle.

Results.- The r e s u l t s of t h e threshold determination t a s k s are shown i n f i g u r e 8 which d i s p l a y s t h e percent of "yes" responses f o r both tasks as a f u n c t i o n of rattle level i n dB(A) u n i t s , The r e s u l t s shown i n f i g u r e 8 i n d i c a t e t h a t t h e threshold of annoyance occurred a t approximately 56 dB(A) which is about 9 dB kigher than t h e threshold of d e t e c t i o n which occurred a t 47 dB(A), That is, t h e threshold of annoyance is 9 dB higher than t h e threshold of d e t e c t i o n and probably r e p r e s e n t s an important r e s u l t , Also, t h e 56 dB(A) required t o achieve r a t t l e annoyance threshold i s believed t o be abnormally high i n a household s e t t i n g . Therefore, i f r a t t l e i s important as a f a c t o r t o evaluation of a i r c r a f t noise, i t s e f f e c t s could o p e r a t e only through d e t e c t i o n s i n c e rattle levels needed t o achieve annoyance are n o t believed t o occur.

E f f e c t of Rattle on A i r c r a f t Noise Annoyance Experimental design,- I n order t o determine i f t h e perception of r a t t l e a f f e c t s s u b j e c t i v e response t o a i r c r a f t flyover noise, a n a d d i t i o n a l experiment was conducted based upon t h e experimental design shown i n f i g u r e 9. Tape recorded a i r c r a f t n o i s e w a s presented t o t h e s u b j e c t s a t four d i f f e r e n t A-weighted sound pressure levels, both w i t h and without accompanying r a t t l e .

A t o t a l of 24 s u b j e c t s p a r t i c i p a t e d i n t h e test. T w e l v e of t h e s u b j e c t s were exposed t o n o i s e l r a t t l e combinations, whereas t h e remaining 12 s u b j e c t s w e r e exposed t o n o i s e only. The s u b j e c t s used t h e magnitude estimation procedure t o provide s u b j e c t i v e evaluations of t h e a i r c r a f t n o i s e s with o r without rattle. For t h i s t a s k , a l l s u b j e c t s w e r e presented with an accompanying a i r c r a f t n o i s e a t a level of 76 dB with no rattle. This standard sound w a s assigned a n annoyance v a l u e of 100 and w a s presented p e r i o d i c a l l y throughout t e s t i n g . The evaluation t a s k f o r t h e s u b j e c t s w a s t o a s s i g n numbers t o successive comparison a i r c r a f t n o i s e s (given i n f i g . 9) t o r e f l e c t how much g r e a t e r o r less t h e annoyance of t h e comparison n o i s e w a s r e l a t i v e t o t h e standard noise. For example, i f t h e annoyance of t h e comparison n o i s e w a s f e l t t o be t w i c e , t h r e e t i m e s , one-tenth, o r one-half t h e annoyance of t h e standard noise, t h e s u b j e c t would a s s i g n 200, 300, 10 o r 50 t o t h e comparison noise, respectively.

I f rattle has a n adverse e f f e c t on a person's annoyance of a i r c r a f t noise, then a n o i s e with r a t t l e would be r a t e d as s u b j e c t i v e l y more annoying than a noise without rattle. This determination could be made s i n c e , as mentioned earlier, only half of t h e s u b j e c t s were exposed t o combined n o i s e and rattle.

Results.- The r e s u l t s of t h i s study are presented i n f i g u r e 10. This f i g u r e shows t h e magnitude estimations of s u b j e c t i v e annoyance obtained from as a function of t h e v a r i o u s s u b j e c t groups (both with and without r a t t l e ) a i r c r a f t A-weighted n o i s e level. Figure 1 0 i n d i c a t e s t h a t increases of n o i s e level produces increased magnitude estimations of annoyance r e g a r d l e s s of whether o r not r a t t l e w a s p r e s e n t , However, t h e most important implication of t h e d a t a of f i g u r e 10 is t h e f a c t t h a t t h e r e is no appreciable d i f f e r e n c e between t h e "rattle" and "no rattle" conditions, That is, t h e presence of rattle d i d n o t , i n a p r a c t i c a l sense, a f f e c t t h e s u b j e c t i v e response t o a i r c r a f t noise.

The implications from t h e s e l a b o r a t o r y s t u d i e s are t h a t any rattle produced by a i r c r a f t flyover n o i s e should n o t , of i t s e l f , produce annoyance i n a l i s t e n e r . Furthermore, t h e presence of r a t t l e does n o t increase t h e annoyance caused by a i r c r a f t flyover n o i s e , Caution should be exercised, however, i n e x t r a p o l a t i n g t h e s e l a b o r a t o r y f i n d i n g s t o t h e real-world environment of t h e a i r p o r t community where t h e noise impact i s confounded by complicating f a c t o r s n o t present i n t h e l a b o r a t o r y such as p r o p r i e t o r s h i p , i n t r u s i o n i n t o r e l a x a t i o n t i m e , etc.).

HELICOPTER FLYOVER STUDY A h e l i c o p t e r flyover s u b j e c t i v e response study has very r e c e n t l y been conducted a t Wallops F l i g h t Center, and d a t a a n a l y s i s is c u r r e n t l y underway.

This study u t i l i z e d approximately 100 test s u b j e c t s ( f i g . 11) i n f o u r groups; two indoors and two outdoors. Two types of house s t r u c t u r e s were u t i l i z e d (one b r i c k veneer and one wood s i d i n g ) and extensive physical measure- ments of s t r u c t u r a l v i b r a t i o n w e r e obtained i n t h e s a m e mammer as t h e Kennedy Airport study. I n t h e d a t a a n a l y s i s of t h i s study, a c c e l e r a t i o n levels of building s t r u c t u r a l elements w i l l be q u a n t i f i e d as a function of t h e h e l i c o p t e r n o i s e level and w i l l be compared with t h e s i m i l a r d a t a from CTOL a i r c r a f t . A t o determine i f t h e h e l i c o p t e r n o i s e d a t a primary question of t h i s study i s c o r r e l a t e s with CTOL n o i s e d a t a , o r i f some c h a r a c t e r i s t i c of t h e h e l i c o p t e r n o i s e s i g n a l ( r o t o r bang, low frequency, etc.) causes it t o be unique. I n a d d i t i o n , t h e s u b j e c t i v e response d a t a w i l l be analyzed t o determine i f it is r e l a t e d t o amount of h e l i c o p t e r caused building v i b r a t i o n i n a fashion analogous t o t h e way t h e s u b j e c t i v e d a t a w e r e r e l a t e d t o t h e CTOL n o i s e d a t a f o r t h e study conducted a t John F. Kennedy I n t e r n a t i o n a l Airport.

CONCLUDING REMARKS Based on a n extensive physical measurement program and a l i m i t e d s u b j e c t i v e response p i l o t study conducted i n t h e J. F. Kennedy I n t e r n a t i o n a l Airport communities of New York C i t y ( i n conjunction with t h e government's assess- ment program of t h e Concorde SST), t h e following concluding remarks can be made: 1. A v i b r a t i o n d e t e c t i o n threshold w a s determined f o r CTOL aircraft and w a s found t o correspond t o a n o u t s i d e o v e r a l l sound pressure level of approximately 100 t o 105 dB. This implies t h a t aircraft-generated n o i s e s of t h i s level can produce perceivable s t r u c t u r a l v i b r a t i o n s .

2.

The perception of v i b r a t i o n w a s found t o produce a n i n c r e a s e i n t h e annoyance a s s o c i a t e d with a n a i r c r a f t flyover event giving t h e implication t h a t v i b r a t i o n is a n important f a c t o r which should be considered i n t h e assessment of a i r c r a f t f lyover noise.

3. The r e s u l t s of t h e community study and a laboratory p i l o t study suggested t h a t t h e e f f e c t s of rattle upon s u b j e c t i v e a i r c r a f t n o i s e annoyance are n e g l i g i b l e .

This is based upon t h e f a c t t h a t t h e phenomenon of r a t t l e w a s observed f o r less :han 3 percent o f t h e a i r c r a f t n o i s e events during t h e f i e l d study, and i n t h e .aboratory study, t h e presence of r a t t l e d i d n o t appreciably influence u b j e c t i v e evaluations of annoyance t o flyover noise, REFERENCES

.. Second Survey of A i r c r a f t Noise Annoyance Around London (Heathrow)

Airport. H.M.S.O., London, 1971.

I. Staff-Langley Research Center: Concorde Noise-Induced Building Vibrations I n t e r n a t i o n a l Airport Dulles - F i n a l Report. N A S A TM 74083, 1977.

I 1 ’

I

NO1 SE

V I BRATION RES I DENT I I

PHYSICAL I SUBJECTIVE

Figure 1.- Schematic illustration of impact of aircraft noise on people exposed to noise while indoors.

Figure 2.- Community noise study; human response to noise, vibration and rattle.

FLYOVER DETECTION ANNOYANCE NO.

ANNOYANCE RATING OF FLYOVER YES NO YES NO e

- VIBRATION- - - -

I I . 1 . I I

I I I I I 5 6 7 8 9 0 1 2 3 4

RATTLE - - --

ANNOYANCE RATING: 0 - ZERO ANNOYANCE

9 - MAX I M U M ANNOYANCE

Figure 3 , - Sample flyover event rating form used i n communfty noise study.

1 0 0 I DETECT1ON 50 1 I I I I

40 50 60 70 80

VERT1 CAL FLOOR ACCELERATION, dB Figure 4 . - Results of experiment t o determine vibration detection threshold.

V I BRATION o DETECTION AVERAGE 5 ANNOYANCE I 1 I I f L I 0 70 80 90 100 110 120

OUTS I DE LA, d B

Figure 5.- E f f e c t of v i b r a t i o n on o v e r a l l s u b j e c t i v e annoyance t o a i r c r a f t f l y o v e r noise.

Figure 6 . - I n t e r i o r e f f e c t s room of Langley a i r c r a f t n o i s e r e d u c t i o n l a b o r a t o r y .

4 5 4 7 51 5 4 56 59 61 NO. OF SUBJECTS: 24 TASK: I I I YES/NO DETECT1 N YES/NO ANNOYANCE EXPECTED RESULTS: % THRESHOLD YES DETECT1ON ANNOYANCE RATTLE LEVEL RATTLE LEVEL Figure 7.- Experimental design of determination of r a t t l e d e t e c t i o n threshold and r a t t l e annoyance threshold.

PERCENT YES RESPONSES Figure 8.- Comparison of rattle d e t e c t i o n and rattle annoyance thresholds.

RATTLE LEVEL, dB (A) NO1 SE LEVEL, dB (A) NO. OF SUBJECTS: 24 TASK: MAGNITUDE ESTIMATION ANNOYANCE TO AIRCRAFT EXPECTED RESULTS SUBJECT ANNOYANCE AIRCRAFT NOISE LEVEL Figure 9.- Experimental design of determination of effect of rattle on aircraft noise annoyance.

0 RATTLE

- - - - - NO RATTLE

cl MAGNITUDE EST1MATI ON , ~ ~ , 1 0 72 74 76 78 80 82

' NOISE LEVEL, I dB (A) ' 1 I

Figure 10.- Effect of rattle on aircraft noise annoyance.

NUMBER OF SUBJECTS: z 2 100 SUBJECT LOCATION: INDOORS AND OUTDOORS PHYSICAL MEASUREMENTS: STRUCTURAL V I BRATION SOURCE DIFFERENCES: AMOUNT OF ROTOR BANG HIGH V I BRATI ON HELl COPTER

/ SUBJECT

ACCELERATION V I BRATION LEVEL ?

HELl COPTER = CTOL

NO1 SE LEVEL NO1SE LEVEL Figure 11.- Subjective response t o helicopter flyover noise study.

A METHOD FOR DETERMINING INTERNAL NOISE CRITERIA BASED ON PRACTICAL SPEECH COMMLTNICATION APPLIED TO HELICOPTERS* Harry Sternfeld, Jr., and Linda Bukowski Doyle Boeing Vertol Company SUMMARY The objective of this study was to provide information regarding the relation- ship between the internal noise environment of helicopters and the ability of personnel to understand commands and instructions. A test program was conducted to relate speech intelligibility to a standard measurement called Articulation Index. An acoustical simulator was used to provide noise envi- ronments typical of Army helicopters. Speech material ("Command" sentences and Phonetically Balanced Word Lists) were presented at several voice levels in each helicopter environment. Recommended helicopter internal noise crite- ria, based on speech communication, were derived and the effectiveness of hearing protection devices were evaluated. Similar limits to satisfy other types of criteria can be developed using the methods presented in this paper.

INTRODUCTION An investigation being conducted by the Army's Working Group on Aircraft Noise indicates that speech communication requirements, as well as hearing damage risk criteria, are important factors in determining interior noise level criteria for helicopters and that the speech communication requirements are especially important in troop carrying helicopters where misunderstood instructions in combat situations may lead to serious consequences.

Speech communication in the current generation of Army helicopters is known to be inadequate but little information has been quantified in order to assist in developing adequate criteria. Although several studies have been conducted in the general area of speech communication in office-type environments, no predictive method had been validated as being completely adequate for all In the helicop- types of noise environments and all types of speech content.

ter both of these factors are somewhat unique. The ambient noise is generally of higher level and more dominated by both low frequency noise (from rotors) and pure tones (from transmissions and turbines) than most environments previ- ously tested; while the speech tends to be simple sentences of limited content, also quite different in structure than the material used in more generalized speech testing.

* This study was conducted under contract with the U.S. Army Aviation

Systems Command, Contract DAAJOZ-74-C- 2054.

In order to provide the needed information the Army (AVSCOM) has sponsored the study which is described in this paper to assist in establishing the rela- tionship between the internal noise environment of helicopters and the ability of troops flying in helicopters to uaderstand commands and instructions. In- formation of this type, which was not previously available, can be used as a basis for reevaluating and, if necessary, revising Specification MIL-A-8806-A, "Acoustical Noise Level in Aircraft, General Specification For."

The program, therefore, applied standard speech communication test procedures as specified in Reference 1, but added noise environments and speech material typical of those encountered in troop-carrying helicopter operations in order to assess Army requirements. The same techniques can also be applied to develop criteria for application to other purposes such as passenger carrying civil helicopters.

TEST PROGRAM Speech Intelligibility Testing Any test involved with speech communication is called speech intelligibility testing. Two measurements which frequently come up in tests regarding speech intelligibility are Articulation Index and Speech Intelligibility Scores.

Intelligibility refers to those units of speech material which are complete and meaningful words, phrases, or sentences. Test subjects are asked to listen to noise and speech simultaneously; they then record, in writing, what has been spoken. The subjects are scored on the percentage of speech recorded correctly.

These scores are called their Speech Intelligibility Scores.

2 ) , is a Articulation Index, developed by French and Steinberg (Reference rather powerful tool for measuring speech intelligibility. The Articulation Index (AI) is a weighted number representing, for a given set of speech and noise conditions, the effective proportion of the normal speech intelligibil- ity. AI is computed from acoustical measurements (or estimates) of the speech spectrum and of the effective masking spectrum. The detailed method for cal- culating Articulation Index may be found in Reference l which also gives suggestions for refinements to AI to take into account such things as inter- ruption in the noise, the reverberation in the listening sifuation, the vocal effort used by the talker and face to face talking.

Figure 1, from Reference 1, shows the correlation between the speech intelli- gibility scores and articulation index in certain noise and speech environ- ments. By use' of this figure AI can be used to predict speech intelligibility.

In intelligibility testing there are four main categories of test materials (i.e., speech material) used most often (Reference 3 ) : ( 1 ) nonsense syllables (2) monosyllabic words ( 3 ) spondiac words ( 4 ) sentences Nonsense syllables are superior to words or sentences as test items when it is desired to determine accurately the effectivenss of a device in transmitting particular speech sounds.

Two practical dis- advantages are the speaker must pronounce the speech sounds very precisely and the test subjects must record the sound they hear in phonetic symbols.

A standard monosyllabic list contains words so chosen that all speech sounds are approximately according to their frequency of occurrence

in normal speech; hence, they are termed "Phonetically Balanced (PB) . 'I

There are 20 such lists of 50 words each, the spread of difficulty being approximately the same in each list and each list having nearly the same average difficulty. Rare and unfamiliar words have been avoided as much as possible.

Two variations on the monosyllabic word lists are the rhyme test and the modified rhyme test which are tests of phonemic differentiation.

The rhyme test vocabulary consists of 50 monosyllabic word sets of five rhyming words each; the modified rhyme test consists of 50 mono- syllabic word sets of six rhyming words each (Reference 4 and 5 ) .

Within a given set, the rhyming words differ in a consonantal spelling. The subjects are scored on whether they supplied the correct consonant. Reference 5 contains more information on the differences between the rhyme and modified rhyme tests.

Spondiac lists contain words of homogenous audibility, i.e., lists in which each individual word is as difficult as each other word.

Such words are most easily selected from those which have the sylla- bles spoken with equal stress on each syllable; e.g., railroad, horseshoe, airplane. They are especially useful in tests whose design is to establish accurately the amplification or power level at the threshold of hearing.

There are different forms of sentence lists. In one form, the test subject is required to respond to questions or commands by an appro- priate word or phrase. A sentence would then be either "right!' or F 1 wrong," depending on whether or not it was clear that the subject understood the meaning.

In another type of test, the listener is required to record in writing the sentence that is read t o him. In sentences of this type, there are "key words!' that are to be marked as right or wrong. An effort is made to avoid clichgs, proverbs, and other stereotyped constructions, as well as the too frequent use of any one word.

Noise Simulation One of the obvious problems in conducting a speech intelligibility program involving helicopter noise is that of providing enough different helicop- ter interior environments ( a n d systematic variations in environment), which are stable and repeatable enough to permit the required testing. Prior to this program, Boeing Vertol had developed a useful capability for synthesis of internal noise. This simulator (Figure 2) consists of a full-scale mockup of a 15-passenger helicopter cabin and cockpit. Loudspeakers and horns are mounted at various locations on and around the fuselage in order to provide an independent sound source for each noise component. Each speaker receives its signal from a separate amplifier which, in turn, is input from a separate track of a 14-track tape playback. There are three fundamental types of signals which are employed in the simulation, broadband noise, continuous pure tones and pulsed harmonic sets.

The broadband noise, such as engine and boundary layer, are produced by random noise generators and then shaped through a one-third octave band filter set to achieve the desired spectrum.

Pure tone components, such as transmission gears, occur in harmonic sets.

Initially an oscillator is used to generate each harmonic and that signal, at its required level, is put on a separate track of a magnetic tape. This tape is then dubbed so that the desired harmonic combinations are recorded on a single track of the final tape.

The pulsed harmonic combinations are used for main and tail rotors. In this case each harmonic is stored in the memory of a digital averager until the The averager, which had been espe- entire required combination is in memory.

cially modified, was then instructed to read the stored signal out at a rate corresponding to the blade passage period. The key to making this system work lPes in.the use of on-line real-time data analysis of signals sensed by micro- In phones at desired locations in the aircraft as the signals are generated.

this manner the input is adjusted so that the desired output is obtained, thereby automatically accounting for electronic component frequency character- istics and for "room" acoustics.

In any finite size enclosure with directive sources, the sound field, espe- cially of pure tone components, will not be uniform. Anyone who is familiar with helicopters knows that substantial differences in transmission noise level can be experienced by merely moving one's head. Therefore any meaning- ful sound pressure level data must be obtained by space averaging both during simulation development and final analysis. To achieve this, occupancy was limited to four test subjects sitting in a rather tight square. During simu- lation development a microphone on a rotating turntable was used. Each noise component was ,provided from the appropriate loudspeaker and its level adjusted so that its average level during rotation reached the desired value.

A preliminary review was conducted of available data on internal noise in current Army helicopters. This compilation resulted in the scatter shown in Figure 3 . Also shown is the level defined by specification MIL-A-8806-A indicating that most procurement, to date, has deviated from established requirements.

Since the main concern of this study is directed at current Army troop- carrying helicopters, this constraint limits the field to two models: the Bell UH-1 (Huey) and the Boeing Vertol CH-47 (Chinook) series. Accordingly, the following were selected as representative for the test program: 1. CH-47C at 22 680 kg (50 000 lb) gross weight (Current Army configuration).

2.

CH-47A at 14 969 kg (33 000 lb) gross weight (Original A m y Chinook Configuration which contained more acoustical treatment in the aft cabin than the CH-4 7 C ) .

A "Paper Design" CH-47 which is a predicted spectrum of a Chinook 3 .

designed to meet MIL-A-8806-A. This will be referred to as CH-47 'MIL Spec.'

4. UH-LH at 4309 kg (9500 Ib) gross weight.

5. UH-1H with doors open which is a configuration often used to permit rapid egress of troops when approaching landing zones.

In order to develop the required simulations, the aircraft noise must be studied in detail in order to identify individual components by means of narrow band spectra shown in Figures 4 and 5. Analyses of this type of data from several locations in each aircraft was done to develop the spectra which were the basis for preparation of the test samples. A typical simulation consistfng of two broadband and thirty harmonic components is presented in Figure 6.

Speech Material This program, after much consideration of the available speech materials, used both standard Phonetically Balanced (PB) word lists and non-standard "Military Jargon" sentence lists. The 1000 PB word list is from Reference 6. The purpose of the PB testing is to provide continuity between this program and the results of other speech communication studies.

Since this program is concerned with determining the effect of helicopter noise on communication with troops, the "Military Jargon" sentence lists were devel- oped as samples designed to reflect somewhat typical commands and questions which might occur in an assault helicopter. Some typical sentences were: Hold it - move now.

Advance to the northeast sector.

Out the front doo,r.

Sentences of this type are more pertinent to actual Army usage than the more PB word lists. There were five different military sentence lists, theoretical each one composed of twenty sentences. Each of the five lists was randomly scrambled once; hence, there was a total of ten lists (two hundred sentences) of the military type employed in the program. The two hundred sentences contained a total of 300 key words on which scoring was based. It will be noted that the military sentences on the whole are much shorter than the stand- ard sentences of Reference 3 of the type whose test results are shown in Figure 1. Also, they are limited to fewer words; whereas the standard sen- tences avoid cliches and stereotyped expressions, the military sentences are composed mainly of these.

The PB word lists and "Military Jargon" sentence lists were tape recorded for playback through a loudspeaker located so that the sound came from the front of the cabin. The individual selected as the speaker had no strong regional accent and had clear diction. As the data was recorded the peak levels were monitored on a graphic level recorder to assist in maintaining a constant voice level. A raised voice was used rather than a conversational tone to produce the type of compressed range associated with higher voice levels. A calibration tone was included on each tape for the purpose of providing a con- stant reference for setting the desired playback volume levels. The use of this procedure ensured repeatability of data between test sessions.

Test Procedures The subjects in this program were twelve males who were members of various engineering staffs of the Boeing Vertol Company. Potential candidates were screened by audiograms, conducted by the Boeing Vertol Medical Department.

Prior to the start of the test program, the twelve test subjects were divided into three groups, consisting of four subjects each (the subjects were allowed to form their own groups of four). At a pre-test session, the subjects were given their instructions, they familiarized themselves with the military sen- tences and a practice session was held. Since testing with phonetically bal- anced words is a much more rigorous procedure which requires training to a level where other sub-Jects score at least 90% in a quiet environment, these practice.sessionswere held at each test session scheduled for military jargon testing. By the time PB word testing all subjects had exceeded the minimum requirements.

Each subject was equipped with writing utensils, a lap board and answer sheets.

During a particular test ( a test is defined as one aircraft level and one speech level), the helicopter noise was continuous; one word and/or sentence at a time was played with a finite time interval for each subject to write what he heard or what he thought he heard. Minor rests of about 1 minute were given at approximately 5-minute intervals with longer rest periods at every 15 minutes.

In order to ensure that each test would produce a meaningful range of results, a pretest was performed for each aircraft noise environment. In this pretest the voice playback level was varied until, in the opinion of the Test Director, only a few of'the messages could be understood. The level was then readjusted At each of until it was judged that most of the messages could be understood.

these levels the value of the 1000 Hz reference tone was measured. These cal- ibration levels and one additional level which was the average of the above two were then used to set the voice levels prior to each test.

In addition to the full aircraft spectra discussed above, additional variations were obtained by completely eliminating individual components by disconnecting This the appropriate tape track from the simulation system during playback.

not only provided more test environments but was also used to evaluate the relative importance of the various noise sources in affecting speech communi- cation.

Additional testing was conducted in one single rotor and one tandem rotor con- figuration to evaluate the effect of hearing protective devices.

RESULTS Data Analysis The phonetically balanced word lists were simply scored as correct or incorrect for each individual word. Phonetic spelling and/or misspellings were counted as correct providing the word was recognizable. Failure to fill in any word was counted as incorrect. In the sentence lists, failure to correctly recognize any key Word was scored as not understanding the sentence, hence there was usually more than one chance to miss a sentence.

The principles of calculating A.I., which are fully described in Reference 1, require measurement of both the ambient noise and speech levels at the observ- er's location. In order to define the aircraft noise levels, one-third octave band measurements were made at the left and right ear position of each subject for each separate environment tested. The left and right ear spectra were to obtain a single spectrum for each seat location for arithmetically averaged each test.

Reference 1 recommends the use of long-term rms spectra for speech measurement.

PB tape were played through the speaker sys- Accordingly four sections of the tem and analyzed in a manner which provided one-third octave band spectra 32 seconds for each sample. Averaging these samples thus pro- integrated over duced a 128 second m s spectrum. Since different playback levels were used for each test, a curve of absolute level against calibration tone level was made in order to assign amplitude values to the long term rms spectrum shape.

Articulation Index (A.I.) was calculated as defined in Reference 1. Of the three methods (full octave, one-third octave and 20 band) described in that reference, the one-third octave band method was used due to compatibility with available analyzing equipment. Speech peaks were determined by adding 12 dB to the measured long-term rms values of the speech actually used in preference to use of the idealized voice of Reference 1.

Speech Communication The fundamental results of this program are shown in Figure 7 for all the com- Since we are binations of aircraft noise condition and speech levels tested.

concerned with evaluating the general situation in the cabin of military heli- copters and not the characteristics of specific locations, the data for each of the four test locations has been combined to give a single value represent- ative of the aircraft/speech combination tested. Each PB data point of Fig- ure 8 is based on 1200 and each Military Jargon data point on 720 separate evaluations.

Comparison of these results with the more.genera1 data of Reference 1 is illus- trated in Figure 8 and indicates that these latter criteria would be too con- servative if applied to the helicopter. Since this difference appears to be true of both the sentences and PB words it does not seem that the type of speech material used is responsible, therefore the difference probably is due to the particular characteristic of helicopter cabin noise. Although the scope of this program cannot rigorously define the reason, a possible explanation may lie in the fact that in the speech interference range the helicopter noise is often predominated by pure tones generated by dynamic components. In this case the value of a one-third octave band of noise may be set by only a very narrow portion of the bandwidth, leaving the rest of the band available for much better speech communication than would be indicated by an Articulation Index calculated on the basis of the one-third octave band (or even the twenty band method).

This implies that the use of an equivalent band level concept (by

adjusting the measured band level by 10 log b, (where bt is the bandwidth of

the tone and B the bandwidth of the one-thirdBoctave band in question) might be applicable. Such a procedure would have to be applied exercising great judgment in cases where a particular level is set by several tones, or a com- bination of tones and broadband noise. A more practical approach is to assume the data developed in this program as being more correct than the Reference 1 curves for application to helicopter internal noise.

In evaluating speech communication in a given aircraft, there are essentially three parameters which must be considered: The voice level which may be used.

( 1 ) (2) The distance over which communication must take place.

The reliability of understanding which must be achieved.

(3) In order to perform the required evaluations, it was necessary to assume some voice levels and spectra. For the greatest general applicability, the spectrum shape used was the ideal voice spectrum of Reference 1 as opposed to the spe- cific spectrum of the talker used in this program. The rms overall sound pressure levels corresponding to the descriptions were obtained using a simple test in which three males (who had helicopter experience) used voice levels which were felt to be typical of those employed in military helicopters. The levels were measured using a one-third octave band analyzer with a graphic level recorder set such that the response was an rms level corresponding to a standard sound level meter set for "slow" response. (This corresponds to a pen decay rate of about 40 dB/sec at a writing speed of 80 mm/sec or a 1-second time constant.)

For discussion purposes, three voice levels were evaluated: "Loud Shout" (90 dB rms at speech peak value at 1 meter) was a level which could be used to issue command sentences to troops and was not unlike those employed in addressing ground troops during military drill.

"Short Duration Shoutt1 (100 dB nus at 1 meter) was a level which could be sustained for up to about ten continuous words without rest.

"Maximum Effort" (110 dB rms at 1 meter) was a level which could only be sustained for a few words and resulted in some vocal strain; this level might be expected in emergency situations.

Having defined the voice spectra as just described, the Articulation Index can be calculated for each aircraft by using the appropriate internal noise data and the method of Reference 1. The "Military Jargon" curve of Figure 7 , which was developed by this study, can then be used to convert the abstract Articu- lation Index to a more applicable speech intelligibility percent.

In order to assess the adequacy of speech recognition in Army helicopters, it is necessary to establish requirements which will permit troops to perform their mission. Unfortunately, no such standard has been established. Lacking any published guidelines, two interim criteria are suggested.

A minimum requirement should be based on communication essential to safety.

it should For example, when using maximum vocal effort (110 dB rms at 1 meter), be possible to achieve 50% intelligibility which would at least attract the attention of a person anywhere in the cabin even if he cannot accurately under- stand the message content.

Although maintenance of a minimum communication in emergency communication in emergency situations is of first priority, this does not in itself ensure a comprehension which will permit troops to adequately follow instructions necessary to successful completion of their required missions. A reasonable suggestion might be based on 80% communication using a "Loud Shouting" level (90 dB rms at a distance of 1 meter).

This would ensure that a troop com- mander, using the type of voice which might be used to instruct a platoon on the drill field, could achieve good simultaneous communication with personnel located within a radius of 1 meter of the speaker.

I f it is desired to improve the speech communication reliability by increasing A.I. in an aircraft in an efficient manner, the reductions should be made in Figure 9 illus- the frequency bands which contribute most directly to the A.I.

trates the relative A.I. weighting factors normalized to their maximum value and suggest that noise reduction outside the frequency range of 100 Hz to 5 kHz will have relatively little payoff. Coupled with the above is the fact that it generally requires less weight to attenuate high frequency noise than low frequency. Pure mass attenuation, for example, provides 5-dB increased attenuation per doubling of frequency for the same surface density, while materials such as fiberglass greatly exceed that rate at the higher frequen- cies.

The procedure for establishing an internal noise level to meet a given commu- nication requirement in an efficient manner is illustrated in Figure 10. The A.I. is calculated as described in Reference 1 . The score of the aircraft is found by trial noise limit required to obtain a desired value of the A.I.

and error solution.

At frequencies below 500 Hz, the aircraft noise levels need not be dictqted by speech communication requirements but rather by hearing damage risk criteria.

Application of t h e above procedures t o t h e suggested conditions r e s u l t s i n t h e h e l i c o p t e r n o i s e criteria curve of Figures 11 and 12. Other cri.teria based on o t h e r assumed requirements can be constructed i n a similar manner.

Hearing P r o t e c t i o n The n o i s e levels i n h e l i c o p t e r s n o t only a f f e c t communication b u t may, i f high enough, cause temporary threshold s h i f t , hearing damage, o r a t least discom- f o r t .

For t h e s e reasons, t h e use of p r o t e c t i v e devices may be recommended.

It is important, however, t o know i f t h e s e devices have any adverse e f f e c t on understanding of spoken commands. To i n v e s t i g a t e t h i s , t e s t i n g w a s repeated i n t h e CH-47C and UH-1H (doors open) configurations with t h e following protec- tive devices: 1) Army SPH-4 f l i g h t c r e w helmets.

2) Disposable ear plug (a spongy material which w a s compressed, i n s e r t e d i n t h e ears and then allowed t o expand).

3) A non-disposable f i t t e d ear plug ( s i z e f o r each person based on ear canal measurement).

The r e s u l t s which are shown i n Figure 1 3 reveal t h a t no adverse e f f e c t on speech communication can be a t t r i b u t e d t o t h e use of ear plugs and, i n f a c t , they even enhance understanding. I n t h e case of t h e SPH-4 helmet when eval- uated i n t h e UH-lH, a s i g n i f i c a n t d e t e r i o r a t i o n occurred. Although no measure- ments of n o i s e i n s i d e t h e helmets w e r e made, spontaneous comments from several of t h e test s u b j e c t s i n d i c a t e d t h a t with t h e helmet on, t h e t a i l r o t o r n o i s e w a s extremely annoying. It i s t h e r e f o r e suspected t h a t t h e SPH-4 helmet may have a resonance i n t h e t a i l r o t o r frequency range and may be amplifying one o r more harmonics of UH-1 t a i l r o t o r noise.

The disposable ear plugs t e s t e d s l i g h t l y b e t t e r than t h e f i t t e d ones, probably due t o being u n c r i t i c a l with regard t o f i t . They w e r e a l s o judged t o be more comfortable.

CONCLUDING REMARKS I n o r d e r t o a s s u r e adequate speech communication i n m i l i t a r y h e l i c o p t e r s , i n t e r n a l n o i s e standards should provide f o r n o i s e levels which w i l l r e s u l t i n A r t i c u l a t i o n Index(es) i n accordance with t h e following sketch (which is r e p l o t t e d from Figure 7 ) : ARTlC U LATlON INDEX SPEECH INTELLIGIBILITY TEST SCORES - % CORRECT In order to relate the Articulation Index to a sound pressure level spectrum, it will be necessary to further define the voice level to be required (a) the distance over which communication is required ( b ) the required reliability of communication ( c ) Additional investigation of Army requirements is needed to establish these parameters.

Ear plugs can be used to protect hearing with little or no degradation in speech communication.

The procedures developed during this program can be used to develop internal noise criteria for other applications such as civil transports.

REFERENCES 1. Anonymous; METHODS FOR THE CALCULATION OF THE ARTICULBTIOI? INDEX, ANSI S3.5-1969, American National Standards Institute, New York.

2 . Kryter, Karl D . : THE EFFECTS OF NOISE ON MAN, Academic Press, New York, 1970.

3. Beranek, Leo L . : ACOUSTIC MEASUREMENTS, John Wiley and Sons, Inc., New York, 1962.

4 . Fairbanks, Grant: TEST OF PHONEMIC DIFFERENTIATION: THE RHYME TEST, Journal of the Acoustical Society of America, Vol. 30, No; 7, July 1958, pp. 596-600.

5. Kryter, Karl D., and whitman, Edward C . : SOME COMPARISONS BETWEEN RHYME AND PB-WORD INTELLIGIBILITY TESTS, Letters to the Editor, Journal of the Acoustical Society of America, Vol. 37, No. 6, June 1965, p . 1146.

6. Anonymous: METHOD FOR MEASUREMENT OF MONOSYLLABIC WORD INTELLIGIBILITY, ANSI S3.2-1960 (Revised 1971), American National Standards Institute, New York.

TEST VOCABULARY LIMITED TO 32 PB WORDS

, F/~'PB;RD: 1 1

(1000 DIFFERENT WORDS) NONSENSE SYLLABLES (1000 DIFFERENT SYLLABLES) PERCENT OF SYLLABLES, TEST VOCABUiARY LIMITED WORDS, OR SENTENCES 256 PB WORDS TO UNDERSTOOD CORRECTLY 40 THESE RELATIONS ARE' APPROXIMATE.

THEY DEPEND UPON TYPE OF MATERIAL AND SKILL OF TALKERS AND LISTENER!

0 I I 1 I 1

0 0 . 1 0.2 0.3 0 . 4 0.5 0.6 0.7 0.8 0.9 ARTICULATION INDEX Figure 1.- Relationship between articulation index and speech intelligibility (ref. 1 ) .

SOURCES GENERATION RECORWPLAYBACK SIMULATOR MAIN ROTOR NOISE L - i m i m d FREflUENCY -Hz Figure 2 . - Noise simulation.

SOUND PRESSURE LEVEL

- d 6 RE 2 x 10J N/m2

9c

I

7 0

OCTAVE BAND CENTER FREQUENCY - HZ

(DATA INCLUDES CH-47C, CH-47A, UH-1H, AND OH-58) Figure 3.- Scatter of center cabin noise data i n army helicopters.

SOUND PRESSURE 100

LEVEL - dB RE

2 x N/m2 90 I I I I I I I I I 1 7 0 ' 0 1 2 3 4 5 6 7 a 9 10 FREQUENCY, k H z Figure 4 . - Noise spectrum of CH-47C interior at station 320 (20 Hz bandwidth).

0 FREQUENCY IN H Z 500 LOWER PLANET FIRST HARMONICS FUNDAMENTAL AND SIDEBANDS UPPER PLANET FOURTH HARMONIC SPL IN dB 1.5 FREQUENCY IN kHZ 2 Figure 5.- High r e s o l u t i o n n o i s e s p e c t r a , CH-47C cabin.

SOUND PRESSURE LEVEL

- dB RE 2 x IO-' N/m'

0 ROTOR 0 SYNCHRONIZING SHAFT A UPPER STAGE PLANET GEAR V LOWER STAGE PLANET GEAR 0 SPIRAL BEVEL GEAR 0 ENGINE COMBINER GEAR -0- AERODYNAMIC BROADBAND -am- TURBINE BROADBAND

- TOTAL SPECTRUM LEVEL

- Hz

OCTAVE BAND CENTER FREQUENCY Figure 6 . - Noise components of the CH-47C.

SPEECH INTELLlGlSlLlTY

TEST SCORES -

% CORRECT

- SCATTER LIMIT l o

- - - MEAN

ARTICULATION INDEX

Figure 7 . - Test results - relation between speech intelligibility and

articulation index in helicopters.

SPEECH INTELLIGIBILITY TEST SCORES

- % CORRECT

ARTICULATION INDEX Figure 8.- Comparison of helicopter and general test results.

1 .o 0 .a ARTICULATION INDEX 0.6

WEIGHTING FACTORS -

RE MAX VALUE 0.4 0.2 Figure 9.- Articulation index weighting factors.

VOICE - REF 1

SOUND LEVEL RE 400 T GREAT ENOUGH OCTAVEBANDCENTERFREQUENCY- HZ Figure 10.- Principles involved i n construction of an aircraft noise criterion which r e s u l t s i n minimum weight penalty.

SOUND PRESSURE

LEVEL -

dB R E 2 x N/m2

OCTAVE BEND CENTER FREQUENCY - HZ

Figure 11.- Helicopter i n t e r i o r n o i s e l e v e l s f o r required emergency commands, 50% speech i n t e l l i g i b i l i t y using a maximum e f f o r t voice, 110 dB r m s at 1 meter.

SOUND PRESSURE LEVEL

- dB RE 2 x N/m2

7 0

OCTAVE BAND CENTER FREQUENCY - HZ

Figure 12.- Helicopter i n t e r i o r n o i s e l e v e l s f o r troop i n s t r u c t i o n , 80% speech i n t e l l i g i b i l i t y using a loud shouting voice, 100 dB r m s a t 1 meter.

PB WORDS n NO PROTECTION 0.6 SPH-4 0.4 HELMETS 0.2 DISPOSABLE EARPLUGS FITTED

UH-1 H m

EARPLUGS DOORS OPEN Figure 13.- E f f e c t of hearing p r o t e c t i o n on speech i n t e l l i g i b i l i t y .

THE EFFECTIVE ACOUSTIC ENVIRONMENT OF HELICOPTER CREWMEN

Robert T. Camp, Jr., and Ben T. Mozo

Bioacoustics Di vi sion

US Army Aeromedical Research Laboratory

SUMMARY

Internal and external noise levels of helicopters are usually measured

to determine the acoustic environment of the crewmen. These types of measure-

ments alone are inadequate for assessing the real acoustic hazards of

personnel. The attenuation characteristics of helmets and hearing protectors

and the variables of the physiology of the human ear must be taken into

account in determining the effective acoustic environment of Army he1 icopter

crewmen. Also, the acoustic hazards of voice communications systems noise

may influence the overall acoustic environment of the personnel.

The compo-

site acoustic environment can be determined only with complex acoustic

measurements that are necessary to quantify the effective acoustic environment

of the crewmen.

I NTRODUCT I ON

Noise characterdstics of he1 icopters should be given consideration in

the design and purchase of aircraft for several reasons. The high sound

pressure levels associated with the operat4on of military aircraft are hazard-

ous to hearing b y most damage risk criteria. Also, the high level sounds may

interfere with communications and operational efficiency of crewmen.

The traditional approach to the task of ascertaining the acoustic

environment that may affect the personnel i s to measure sound pressure levels

at various crewmen's positions under various operational conditions. In dis-

cussions of helicopter noise problems in popular and scientific articles, it is

often assumed that the sound pressure levels within the aircraft are the

actual pressure level values that impinge on the ears of the personnel who

operate the aircraft. It is assumed that the ambient sound pressure levels

are the same as the effective acoustical environment of the crewmen. The

effective acoustical environment is defined as the actual acoustic energy

that is received by the hearing system.

Measuring the effective acoustic environment of he1 icopter crewmen is a

complex process. The purpose of this paper is to discuss the differences

between the ambient acoustic level environment of the aircraft and the actual

acoustic levels that affect the aircraft crewmen's ears.

HUMAN AUDITORY SYSTEM EFFECTS

The human ear is divided into three sections: the external, middle, and

inner ear. The inner ear contains a complex system of membranes, nerves, and

hair cells that may be damaged b y noise. The amount of noise-induced hearing

loss is determined by the effective acoustic environment or the actual energy

transmitted into the inner ear.

The middle ear is a pathway to the inner ear and is equipped with certain

protective mechanisms that affect the acoustic input. It contains two muscles

that limit the input to the inner ear.

The external canal is shaped as an irregular tube. There are various

forms and sizes. The variations of sizes and shapes may produce various

amounts of protection, especially in the high frequencies. Unpredictable

limiting characteristics of the middle ear plus the variations of sizes and

shapes of the external ear canal make it difficult to measure the actual

acoustic stimuli traveling through the system to the inner ear.

HEARING PROTECTOR AND VOICE COMMUNICATIONS SYSTEMS EFFECTS

The human auditory system just discussed is not the only source of vari-

ables that may affect the actual sound transmitted to the ear. Hearing

protective devices , earplugs , headsets, and helmets contribute to the trans-

formation of the acoustic environment to sound spectrum characteristics and

sound pressure levels that may be vastly different from the values obtained

using free field measurements in the cockpit.

In general, all types of hearing

protectors, both insert and circumaural , attenuate high frequencies much more

efficiently than they attenuate low frequencies.

Let us take, for example, the case of a typical pilot flying an Artqy

CH-47 helicopter wearing the standard SPH-4 helmet. One would expect that the

change of sound characteristics beneath the helmet when fitted on the head of

the aircrewman would be the original sound spectrum on the outside of the

helmet minus the attenuation characteristics of the helmet. I f the pilot were

flying with no voice communication system or warning signals activated, the

resultant spectrum at the external ear would be approximately the values as

determined b y this method. One must remember that the helmet earcups are

sealed tightly over the ears of the crewmen and contain earphones that gener-

ate a sound source beneath the helmet. One normally does not associate very

high sound pressure levels with an earphone. Coupled to small volumes such as

the canal of the human ear, earphones are capable of generating up to 120 dB

sound pressure level Our investiqations of sound sources from communication

systems, warning signals and navigation signals, have revealed that a very signi-

ficantly high sound pressure level may be transmitted from the earphone. These

levels are much greater than the ambient noise that is transmitted through the

helmet.

Another source of noise that may enter in the total effective acoustic

environment is the distortion created b y the design of voice comnunications

electronic systems. It is well-known that military voice communication systems

are designed with distortion.

Some military specifications require only 70% intelligibility and specify certain various amounts o f peak clipping which yield harmful distortion harmonics. The effects of peak clipping have been

thoroughly discussed in some of my previous presentations to this group. Peak

clipping may cause a decrement of intelligibility and also creates unnecessary

harmonics that add to the total excessive energy that is transmitted to the

ears and thereby the contributor to hearing loss. Added to these harmonics, often we have found inverter power-line noise that yields very high acoustic signals through the earphones and therefore is a significant source of acoustic hazard.

TRANSDUCER CHARACTERISTICS EFFECTS Another aspect of the problem that one should note is the characteristics of transducers.

Ideally, microphones and earphones should have flat response within the audio range. If either the microphone or the earphones have peaks, the crewmen may set gains of voice communications and other signals emitted through the earphones that are unnecessarily high level. By some evaluations the M-87 microphone has been considered one of the best military type noise cancel1 ing microphones available. The response characteristics

contain a very large peak in the 4000 Hz range. So when one receives messages

transmitted through this microphone, there is an unnecessarily high emphasis of the 4000 Hz range. With our present-day knowledge of hearing loss causes, it is well established that this is the most vulnerable portion of the audio spectrum for hearing damage. If this microphone were used in low ambient

noise conditions and if the reception were in quiet environments, there would

probably be no significant amounts of hearing loss. However, in its applica-

tion in high noise environments where gain settings are very high, the net

result is that the ear is subjected to large quantities of energy which, for long periods of time, may cause significant hearing loss for some personnel.

, Earphones should also have flat response. If they produce peak response

one would expect the same potential hearing damaging spectra in high 1 eve1

noise for much the same reason that peak microphones cause problems. The recent trend to change to lighter earphones should be watched carefu ly to avoid regression of headphone response. Some of the small earphones may produce high distortion'when driven at the levels required in the mi i tary noize environment.

With the excess energy caused b y variables of transducer response plus the many other signals such as voice communication messages, warning signals, navigation signals, and special instrumentation, there are significantly high acoustic levels generated beneath the helmet earcups.

5 15 SPECIAL PROBLEMS OF LOW FREQUENCY NOISE One other aspect of the helicopter noise problem that may affect the effective acoustic environment of the crewmen is extremely low rotary blade passing frequency which is usually below the octave bands traditionally given i r the analyses of survey data. In this region of the audio spectrum, measurement is seldom made of the aircraft noise spectra. The attenuation characteristics of the hearing protectors that are also worn in helicopters are not usually reported. The questions arise: Are these low frequencies passing through the helmets? Do they cause high frequency harmonic distribution that contributes to the total noise in the effective acoustic environment?

Our laboratory is presently engaged in the investigation of the effects of low frequencies on animal ears. We have found significant temporary in the high frequencies when the animals were exposed to long threshold shift periods of low frequency band centered at 63 Hz.

CONCLUDING REMARKS In summary, the purpose of our presentation is to call attention to the fact that the ambient sound pressure level measurements often made in helicop- ter cockpits with sound level meters will not yield essential information about the noise characteristics that truly affect the he1 icopter crewmen. We should be aware of the difference between these levels and the actual effective acoustic environment that must be determined by other means. We have shown how the ambient acoustics level transmitted through the helmets is transformed by the attenuation characteristics of the helmet or other hearing protective devices that the crewmen may be wearing. In addition to these transformations, there are added signals caused bv the desian of microphones, earphones, hearing protective devices, and electronic systems. In a word, the actual acoustic energy at the eardrums of the crewmen is usually totally different from the acoustic environment measured around the various positions inside the aircraft.

We have also discussed the variables of the middle ear that limit and modify

All of these variables

the acoustic spectrum that is produced at the eardrum.

and unknowns make it very difficult to assess the real energy that reaches the inner ear, the'locale of hearing damage. We have methods by which we measure the output o f earphones and the magnitudes of various warnings and navigation signals necessary i n the operation of he1 icopters.

Precise narrow band measurements are necessary, and in addition, standard real -ear attenuation characteristics of helmets and hearing protective devices must be accomplished. A better estimate of the total acoustic input to the crewmen is made by the insertion of a tiny microphone in the ear with small 5 16 vires that will not affect the attenuation characteristics of the helmets.

\lso, we have developed a portable cement chamber that is very useful in istimating the output o f earphones as used in the he1 icopter operations.

It is recommended that future he1 icopters voice communication systems, iuditory warning systems, and other special instrumentation with auditory so as to signals be designed with consideration for the total systems output ninimize the acoustic hazards that presently exist. This approach is necessary For realizing the most efficient operation of the crewmen. The evaluation of this low frequency problem can be done only after sufficient research has been xcomplished. The recent findings about the significance o f low frequency 9oise spectra cast doubt about the present universal application of dBA for reporting helicopter noise measurements. It is also recommended that as more sophisticated instrumentation is obtained for the measurement of helicopter noise, mo-re attention be given to the extremely low frequency of the rotary bl ade.

THE EFFECT OF OPERATIONS ON THE GROUND NOISE FOOTPRINTS ASSOCIATED W I T H A LARGE MULTIBLADED, NONBANGING HELICOPTER David A. Hilton, Herbert R . Henderson and Domenic J. Maglieri NASA Langley Research Center William B. Bigler I1 University of V i r g i n i a INTRODUCTION Pending n o i s e c e r t i f i c a t i o n of helicopters has focused a t t e n t i o n on t h e effects of their o p e r a t i o n s on ground noise exposure. Knowledge of t h e effects of helicopter c o n f i g u r a t i o n s (blade numbers, a i r f o i l c o n f i g u r a t i o n , etc.) and o p e r a t i o n s p l a y s an important role i n t h e evolution of f i n a l procedures to be u t i l i z e d for t h e noise c e r t i f i c a t i o n of h e l i c o p t e r s . I n a d d i t i o n , as noted i n reference 1 , t h e p r e d i c t i o n of realistic and r e p r e s e n t a t i v e ground noise con- tours, or f o o t p r i n t s , caused by various a i r c r a f t operations, including h e l i - copters, can aid s i g n i f i c a n t l y i n minimizing the noise i n t r u s i o n .

Considerable effort has been expended i n an attempt to b e t t e r understand helicopter o p e r a t i o n a l effects so that improved noise p r e d i c t i o n techniques can be developed (ref. 2 ) . Measurements made to d a t e are providing a b e t t e r under- standing of the p h y s i c a l phenomenon involved. However, t h e results from s i n g l e p o i n t measurement programs suggest t h a t the determination of t r u e ground expo- sure n e c e s s i t a t e s t h e use of numerous ground noise measurements a t m u l t i p l e l o c a t i o n s along, and perpendicular to, t h e aircraft ground track. The r e c e n t l y developed remotely operated m u l t i p l e a r r a y acoustics range (ROMAAR) a t N A S A Wallops F l i g h t Center, reference 3 , was designed to provide an arrangement for obtaining noise measurements along t h e a i r c r a f t ground track and a t v a r i o u s lateral p o s i t i o n s simultaneously along w i t h information on a i r c r a f t p o s i t i o n , operating parameters, and local meteorological information.

I n order to expand the data base of helicopter e x t e r n a l noise character- istics, NASA conducted a flyover noise measurement program u t i l i z i n g t h e NASA C i v i l Helicopter Research Aircraft. I n these s t u d i e s , both the ROMAAR and a 2560-m l i n e a r microphone a r r a y , l a i d o u t along a runway a t NASA Wallops F l i g h t Center, were u t i l i z e d for t h e purpose of documenting t h e noise characteristics of the test h e l i c o p t e r during flyby and landing operations. By u t i l i z i n g both t h e RopllAAR concept and the l i n e a r a r r a y , t h e data necessary to plot t h e ground noise f o o t p r i n t s and noise r a d i a t i o n p a t t e r n s were obtained.

5 19 The purpose of t h i s paper is to p r e s e n t examples of t h e measured n o i s e s i g n a t u r e of t h e test h e l i c o p t e r , the ground noise f o o t p r i n t or contours, and t h e d i r e c t i v i t y p a t t e r n s measured during l e v e l flyby and landing o p e r a t i o n s of a large, multibladed, nonbanging h e l i c o p t e r , the CH-53.

APPARATUS AND METHODS T e s t Site of V i r g i n i a was chosen The NASA Wallops F l i g h t Center on t h e Eastern Shore as t h e test s i t e of these a c o u s t i c s tests. The sketch of f i g u r e 1 i n d i c a t e s t h e geographical r e l a t i o n s h i p between t h e Wallops F l i g h t Center and t h e Langley Research Center. The i n s e r t i n t h e f i g u r e , an aerial photograph of t h e airfielc a t Wallops F l i g h t Center, g i v e s an i n d i c a t i o n of t h e general runway l a y o u t and the type of t e r r a i n surrounding t h e test area.

The Wallops F l i g h t Center offers a number of d e s i r a b l e c h a r a c t e r i s t i c s thai are necessary for flyover noise t e s t i n g . The RClMAAR (ref. 3) is located a t t h e Wallops F l i g h t Center and was made a v a i l a b l e for t h i s series of a c o u s t i c s tests.

The RcplIAAR is located i n an area south of t h e a i r f i e l d where ambient n o i s e is l o w and where a i r c r a f t o t h e r than t h e one under test do not operate routinely.

The microphone a r r a y is located i n r e l a t i v e l y f l a t , open areas which tend to minimize s h i e l d i n g , r e f l e c t i o n , and shadow-zones, etc. The range is supported by aircraft tracking and weather observation f a c i l i t i e s .

The general ROMAAR and linear-array concept combines t h e basic elements of flyover noise t e s t i n g , which include acoustic measurements, a i r c r a f t p o s i t i o n measurements, and weather measurements. The noise measurement elements of t h e range c o n s i s t of analog s t a t i o n s (manned) and/or d i g i t a l s t a t i o n s (unmanned) For these tests, aircraft p o s i t i o n s over t h e range and over t h e l i n e a r micro- phone a r r a y were determined by radar, and t h e a i r c r a f t operating parameters werc read from t h e standard instrument panel aboard t h e test a i r c r a f t . Data from a l l of t h e above sources were t i m e c o r r e l a t e d using a t i m e s i g n a l generated by WWVB.

Test Helicopter The test aircraft for t h i s series of noise experiments was a modified CH-53A turbine-powered t r a n s p o r t h e l i c o p t e r t h a t is being u t i l i z e d as a test bed i n t h e NASA C i v i l Helicopter Technology Program. The test h e l i c o p t e r has uprated engines and d r i v e systems and has a normal gross weight of 16 330 kg (the uprated engines and d r i v e system make t h i s machine comparable to t h e l a t e r D model). The test a i r c r a f t has also been o u t f i t t e d with a 16 seat passenger A photograph of t h e h e l i c o p t e r u t i l i z e d compartment f o r r i d e q u a l i t y research.

for these tests is shown a t t h e top of f i g u r e 2. The g e n e r a l dimensions of t h e CH-53 are shown i n t h e drawing a t t h e b o t t o m of t h e f i g u r e . Briefly, t h e main rotor has s i x blades and has a 22-m diameter, and t h e t a i l rotor c o n t a i n s four blades and is 4.9 m i n diameter.

A i r c r a f t Opera t i o n s RCMAAR.- A s i n d i c a t e d by t h e schematic drawings of f i g u r e s 3(a) and (b), .anding and level f l y b y o p e r a t i o n s were performed o v e r t h e ROMAAR. The ROMAAR : o n s i s t e d of 38 measurement p o s i t i o n s which were l o c a t e d to t h e s o u t h of run- ray 04 and covered a n area approximately 500 m to e i t h e r s i d e of t h e extended : e n t e r l i n e of runway 04 and approximately 10 km downrange. The circles i n t h e ;ketches i n d i c a t e t h e approximate deployment of t h e microphones.

For t h e l a n d i n g approach measurements, t h e a i r c r a f t approached t h e range i t a n a l t i t u d e of approximately 457 m u n t i l t h e six-degree g l i d e s l o p e was intercepted; a t t h a t p o i n t a d e s c e n t was made to a f u l l stop l a n d i n g a t a p o i n t t p p r o x h a t e l y 40 m i n s i d e t h e approach end o f runway 04/22. For t h e s e opera- :ions t h e approach was always made from t h e south, d i r e c t l y over t h e ROMAAR.

For t h e l e v e l f l y b y n o i s e measurements, c o n s t a n t a l t i t u d e f l y o v e r s were nade along t h e extended c e n t e r l i n e of runway 04/22, d i r e c t l y over t h e RQMAAR.

!!he helicopter was flown a t a nominal a l t i t u d e of approximately 152 m a t an air- speed of 49 m/s. For t h i s particular series of tests, t h e h e l i c o p t e r was flown )n reciprocal headings over the ROMAAR. The h e l i c o p t e r f l i g h t p a t h and power :onditions were s t a b l e approximately 1 km prior to range e n t r y and 1 km a f t e r Leaving t h e range.

During t h e s e o p e r a t i o n s , t h e s p a t i a l p o s i t i o n of t h e test a i r c r a f t was getermined by u t i l i z i n g a p r e c i s i o n r a d a r ; t h e g e n e r a l s p e c i f i c a t i o n s are l i s t e d in r e f e r e n c e 3. For both t h e l a n d i n g approach o p e r a t i o n s and t h e l e v e l f l y b y > p e r a t i o n s over the ROMAAR range, the onboard f l i g h t parameters were read from ?ilot d i s p l a y instruments.

Linear microphone array.- As i n d i c a t e d i n f i g u r e 3(c), a l i n e a r microphone was e s t a b l i s h e d along t h e c e n t e r l i n e of runway 04/22. U t i l i z i n g t h e s r r a y standard analog measurement systems, microphone p o s i t i o n s were e s t a b l i s h e d a t 3istances i n d i c a t e d i n t h e sketch. For t h i s series of f l i g h t s , t h e f l i g h t track Has e s t a b l i s h e d p e r p e n d i c u l a r to t h e a r r a y and overhead of t h e c e n t e r microphone.

F l i g h t s were made on reciprocal headings a t airspeeds o f 47 and 82 m / s and a t a l t i t u d e s of 76 and 152 m. For t h i s series of tests, t h e aircraft was i n s t a b i l i z e d c o n d i t i o n approximately 1 km b e f o r e p a s s i n g over t h e microphone a r r a y and t h e s e c o n d i t i o n s were maintained approximately 1 km a f t e r p a s s i n g overhead .

Shown- i n f i g u r e s 4(a) to (c) are r a d a r data concerning t h e a l t i t u d e , lateral displacement, and v e l o c i t y t i m e h i s t o r i e s a s s o c i a t e d w i t h t h e f l i g h t o p e r a t i o n s over ROMAAR f o r t h e six-degree approach and l e v e l f l i g h t c o n d i t i o n s and for l e v e l f l i g h t o p e r a t i o n s over t h e l i n e a r a r r a y . It c a n be noted from t h e s e f i g u r e s t h a t s l i g h t v a r i a t i o n s i n a l t i t u d e , l a t e r a l displacement, and v e l o c i t y e x i s t e d from f l i g h t to f l i g h t . These v a r i a t i o n s i n t h e f l i g h t path and v e l o c i t y c o u l d have r e s u l t e d i n some minor v a r i a t i o n s i n the n o i s e l e v e l v a l u e s from f l i g h t to f l i g h t . C o r r e c t i o n s can be made for t h e s e v a r i a t i o n s ; however, i n t h i s case t h e y were f e l t to be so small t h a t no attempt was made to i n c l u d e such c o r r e c t i o n s for t h e data of t h i s paper.

METEROLOGICAL CONDITIONS Shown i n f i g u r e s S ( a ) and (b) are t h e measured v a r i a t i o n s i n selected meterological q u a n t i t i e s for t h e two time p e r i o d s d u r i n g which t h e s u b j e c t test, were conducted. Temperature, r e l a t i v e humidity, wind d i r e c t i o n , and wind speed are plotted as f u n c t i o n s of a l t i t u d e . The hatched r e g i o n r e p r e s e n t s t h e r a n g e s and measured v a l u e s a t a l t i t u d e s up to and beyond the test a l t i t u d e s . These d a t a were o b t a i n e d by means of s t a n d a r d rawinsondes. Data concerning a l t i t u d e , temperature, and r e l a t i v e humidity are telemetered to a r e c e i v e r on the ground, and d u a l theodolities are used to track t h e l i f t i n g b a l l o o n i n order to o b t a i n the wind information. A l t e r n a t i n g temperature and humidity sondes were release4 a t approximately 1/2-hour i n t e r v a l s throughout t h e test period. These data are p r e s e n t e d for information o n l y and no c o r r e c t i o n s have been made to t h e measurec n o i s e data of this paper to r e f e r e n c e day c o n d i t i o n s .

TEST RESULTS Noise Characteristics Typical n o i s e characteristics of t h e test h e l i c o p t e r d u r i n g a l e v e l f l y - over operation a t an a i r s p e e d of 49 m / s and an a l t i t u d e of 1 5 2 m are i l l u s t r a t e 4 i n figure 6. Presented i n t h e left-hand p o r t i o n of t h e f i g u r e is a sound pres- sure l e v e l time h i s t o r y measured on the ground track d i r e c t l y under t h e test aircraft. The time of overhead passage of t h e helicopter is noted on t h e f i g - u r e , and it can be s e e n t h a t for t h i s p a r t i c u l a r h e l i c o p t e r t h e maximum sound pressure l e v e l o c c u r s a t approximately overhead passage. A frequency spectrum of the n o i s e a t t h e time of maximum sound pressure l e v e l is shown i n t h e r i g h t - hand p o r t i o n of t h e f i g u r e s . It can be seen t h a t t h e spectrum is dominated by t h e r e l a t i v e l y l o w frequency components associated w i t h t h e main- and t a i l - rotor blade r o t a t i o n a l noise. For a l l of t h e measurements made a t the analog s t a t i o n , data of the type i l l u s t r a t e d i n f i g u r e 6 are a v a i l a b l e . For t h e dig- i t a l s t a t i o n s , the dB(A) descriptor was selected for t h e purpose of r e p o r t i n g and d B ( A ) t i m e h i s t o r i e s are a v a i l a b l e from each of t h e s e measurement s t a t i o n s .

Noise Level V a r i a b i l i t y The data of f i g u r e s 7, 8, and 9 are included to i n d i c a t e t h e data spread and v a r i a b i l i t y experienced for t h i s series of tests. An i n d i c a t i o n of t h e n o i s e v a r i a b i l i t y c a n be o b t a i n e d by examining t h e histograms of f i g u r e s 7 and 8. Shown . i n these f i g u r e s are histograms which i n d i c a t e the v a r i a t i o n s i n maximum dB(A) for s i x f l i g h t s of the test aircraft over t h e 2560-m l i n e a r a r r a y a t an a l t i t u d e of 152 m and an airspeed of 49 m / s and during an approach opera- t i o n o v e r the ROMAAR. Data are grouped i n i n t e r v a l s of 5 dB(A) . I n g e n e r a l , t h e s e figures i n d i c a t e that less scatter is associated w i t h s t a t i o n s located under the aircraft, along the f l i g h t track t h a n a t t h e l a t e r a l l o c a t i o n s . A s d i s t a n c e i n c r e a s e s and/or as l o o k a n g l e decreases, the v a r i a b i l i t y i n t h e n o i s e measurements i n c r e a s e s .

Presented i n f i g u r e 9 are average data from four passes of the test h e l i - copter Over t h e ROMAAR a t a n a l t i t u d e of 152 m and a n a i r s p e e d of approximately 49 m/s. The data, presented a t t h e top of t h e figure, are for t h e on-track microphones only. The average l e v e l s frcan these microphones are shown as f u n c t i o n s of p o s i t i o n on t h e track: as an aid to i n t e r p r e t a t i o n , a l h e has been drawn a t a nominal l e v e l of 80 d B ( A ) . It is i n t e r e s t i n g to n o t e t h a t , over the e n t i r e 10-km length of the ROMAN?, the on-track v a r i a t i o n s are on t h e order of approximately 3 dB (A).

U t i l i z i n g t h e average dB(A) levels from a l l measurement s t a t i o n s i n t h e RCMAAR a r r a y for t h e same four l e v e l f l y b y s , d a t a are presented i n t h e lower p o r t i o n of f i g u r e 9 which i n d i c a t e the e x t e n t of t h e ground n o i s e f o o t p r i n t of the h e l i c o p t e r . For each measurement s t a t i o n , the e r a g e dB(A) l e v e l for t h e four f l i g h t s is shown a t t h e measurement s t a t i o n l o c a t i o n . Based on these average l e v e l s , the 75-dB(A) and 70-dB(A) noise contours, or f o o t p r i n t s , were developed. These contours were obtained by appropriate c r o s s - p l o t t i n g and e x t r a p o l a t i o n of the average values of the measured n o i s e l e v e l parallel to and perpendicular t o t h e aircraft ground track. Inspection of the contour i n d i c a t e s t h a t the p a t t e r n is q u i t e symmetrical i n n a t u r e around t h e aircraft ground track.

Ground Noise Contour and D i r e c t i v i t y P a t t e r n s Six-degree landing approach operations.- The ground noise contours for a six-degree landing approach of t h e test helicopter are shown i n figure 10.

These contours were constructed u t i l i z i n g t h e average of t h e maximum d B ( A ) l e v e l s measured a t each of t h e 38 ground noise meaurement s t a t i o n s during f i v e 6O landing o p e r a t i o n s over t h e ROMAAR. Again, these contours were constructed by c r o s s - p l o t t i n g and e x t r a p o l a t i o n of t h e average of the values of the measured l e v e l s parallel and perpendicular t o t h e a i r c r a f t ground t r a c k . Again, it should be noted t h a t the o r d i n a t e and a b s c i s s a are p l o t t e d to t h e same scale.

One can see t h a t during t h e approach t h e contours are parallel to t h e f l i g h t t r a c k and are symmetrical i n nature. A s t h e helicopter begins its descent o t h e r contours appear and, due to t h e descending f l i g h t , these contours close.

These closed contours are also symmetrical about t h e f l i g h t track which suggests t h a t t h i s particular h e l i c o p t e r does n o t e x h i b i t any s h a r p l y d i r e c t i o n a l n o i s e pat tern.

Further i n s i g h t i n t o t h e noise r a d i a t i o n p a t t e r n s for t h e test helicopter can be obtained by examining the results from the l e v e l f l y b y s over t h e l i n e a r microphone array.

Level flybys.- The data of f i g u r e 11 represent a i r c r a f t n o i s e d i r e c t i v i t y p a t t e r n s obtained as a r e s u l t of analyzing the n o i s e data for s i x passes of the test aircraft over t h e l i n e a r microphone array. The aircraft was frozen over- head of the c e n t e r microphone of t h e l i n e a r a r r a y and a t s e v e r a l l o c a t i o n s during t h e approach to and the departure from t h e a r r a y .

The aircraft was frozen a t five-second i n t e r v a l s along the f l i g h t track, and t h e dB(A) levels were read from the time histories a t each measurement l o c a t i o n for t h a t partic- ular t i m e .

I n t h i s manner a grid of dB(A) v a l u e s was e s t a b l i s h e d for each pass.

These data were then e x t r a p o l a t e d and cross-plotted: one could then choose a particular dB(A) value and a c o n s t a n t dB(A) contour or l i n e drawn thkough t h e i n t e r c e p t points. The d a t a of f i g u r e 11 show that t h e r a d i a t i o n p a t t e r n is symmetrical about the f l i g h t path a x i s ; however, t h e l e v e l s seem to be somewhat higher to either side of the a i r c r a f t than those to t h e f r o n t and rear of the aircraft. These d a t a g r a p h i c a l l y i l l u s t r a t e t h e symmetrical p a t t e r n s t h a t were suggested by the previous f i g u r e .

The v a r i a t i o n i n t h e d i r e c t i v i t y p a t t e r n s for t h e test helicopter as a f u n c t i o n of a l t i t u d e and airspeed are i l l u s t r a t e d i n f i g u r e 12. It can be seen t h a t increasing a l t i t u d e r e s u l t s i n very small changes i n t h e shape or s i z e of the r a d i a t i o n p a t t e r n s . This suggests t h a t a uniform r a d i a t i o n p a t t e r n e x i s t s On t h e o t h e r hand r a d i a t i o n p a t t e r n s developed from d a t a below t h e h e l i c o p t e r .

measured a t the same altitude,,but a t two airspeeds show an i n c r e a s e i n the radi- a t i o n p a t t e r n s i z e i n a d d i t i o n to a change i n shape. The i n c r e a s e i n l e v e l is thought to be due to the i n c r e a s e i n disc loading a t t h e higher speeds, while t h e shape is thought to be modified by a change i n t h e tilt to t h e blade t i p path.

CONCLUDING RElvIARKS A f i e l d measurement program was conducted u t i l i z i n g the NASA C i v i l H e l i - copter Research Aircraft. I n these s t u d i e s both t h e remotely operated m u l t i p l e a r r a y acoustic range (RCMAAR) and 2560-m l i n e a r microphone a r r a y l a i d o u t along a runway a t t h e NASA Wallops F l i g h t Center were u t i l i z e d for t h e purpose of documenting the n o i s e characteristics of t h e test h e l i c o p t e r during f l y b y and landing operations. By u t i l i z i n g both t h e ROMAAR concept and t h e l i n e a r a r r a y , the data necessary to plot ground noise f o o t p r i n t s and n o i s e r a d i a t i o n p a t t e r n s were obtained.

The results of these tests i n d i c a t e that both t h e ground n o i s e contours and t h e r a d i a t i o n p a t t e r n s for t h i s p a r t i c u l a r helicopter are symmetrical about the f l i g h t a x i s of the h e l i c o p t e r . It was also shown t h a t i n c r e a s i n g a l t i t u d e d i d n o t s i g n i f i c a n t l y modify t h e r a d i a t i o n p a t t e r n while i n c r e a s i n g speed caused higher l e v e l s and a d i s t i n c t change i n t h e r a d i a t i o n p a t t e r n .

REFERENCES 1. Stepniewski, W. Z . ; and Schmitz, F, H . : P o s s i b i l i t i e s and Problems of Achieving Community Noise Acceptance of VTOL. Presented a t Eight Congress of I n t e r n a t i o n a l Council of Aeronautical Sciences, ICAS Paper No. 72-34, Amsterdam, The Netherlands, Aug. 1972.

2. Maglieri, Domenic J.; Henderson, Herbert R.; and H i l t o n , David A.: Observed V a r i a b i l i t y of Aircraft Noise F o o t p r i n t Measurements. ‘NOISE- CON 77, Proceedings, George C. Maling, Jr., ed,, Noise Control Found.

c.1977, pp. 443-458.

3. H i l t o n , David A.. and Henderson, Herbert R.: A Remotely Operated Multiple Array A c o u s t i c Range (ROMAAR) and Its Application for t h e Measurement of Airplane Flyover Noise Footprints. NASA TM X-73986, 1976.

5 2 4 Figure 1.- Schematic illustration of geographical relationship between Wallops and Langley.

Figure 2.- Turbine-powered helicopter used in tests.

(b) Level flyby over R O U .

Figure 3.- Schematfc drawings of landing and level flyby operations performed by test helicopter.

(c) Level flyby over linear array.

Figure 3.- Concluded.

ALTITUDE, 300 m 200 1 0 0

4 FLIGHT DIRECTION

LATERAL D I S PLACEMENT, 0 m 100

AIRCRAFT *Oar

VELOC ITY, 100 mlsec 0 2 4 6 8 10 DISTANCE TO TOUCHDOWN, km (a) Six-degree landing approach over ROMAAR.

Figure 4.- Radar tracking and aircraft velocity information.

200 r

ALTITUDE, m 150

t~

0 FLIGHT DIRECTION

50 r

LATERAL D I S PLACEMENT, 0 m

t

50 I.- A I RC RAFT mlsec 0 2 4 6 8 10 DISTANCE ALONG GROUND TRACK, km (b) Level flyby over ROMAAR.

ALTITUDE, 150 m

' 0 FLIGHT DIRECTION

LATERAL D I S PLACEMENT, 0 m

L

'\ RUNWAY

100 r 1 I A I RC R A m VELOCITY, 50

c

I ml sec I I L L 1.5 0 1.5

DISTANCE FROM $ ARRAY, k m

(c) Level f l y b y over l i n e a r a r r a y .

Figure 4.- Concluded.

ALT I TU DE, m I , & , 50 60 70 80 10 20 30 40 RELATIVE HUM1 D ITY, PERCENT TEMPERATURE, OC ALTITUDE, rn 0 5 10 15 deg.

W I N D DIRECTION, W I N D VELOC ITY, mlsec (a) R O W operations.

ALTITUDE, 300 rn I I I 10 20 30 40 40 50 60 70 RELATIVE H U M I D I T Y , PERCENT TEMPERATURE, OC ALTITUDE, 3oo m 0 5 10 15 0 90 180 270 360 W I N D VELOC ITY, mlsec W I N D DIRECTION, deg (b) Linear a r r a y o p e r a t i o n s .

Figure 5.- V a r i a t i o n s i n meteorological c o n d i t i o n s f o r t w o test periods.

ROTOR rpm = 100 PERCENT TORQUE= 40 PERCENT STATION 8

s PL

- I 1 1 1 1 1 1 1 I I I I 1 1 1 1 I I I I

40 20 0 20 40 .01 .I 1 1@ T IME, sec FREQUENCY, kHz Figure 6.- Typical noise characteristics of test helicopter during level flyby. Altitude =i 152 m, airspeed = 49 m/sec.

FLIGHT DIRECTION STATION 6 STATION 1 = 90'

p = 70

f3 = 23

SR = 152 m

SR = 1290 m

SR = 397 m

NUMBER OF EVENTS ~

60 70 80 90 1 0 0

60 7 0 80 90 1 0 0

MAX I M U M LEVEL, dB(A) Figure 7 . - Variation in maximum dB(A) levels over linear microphone array for six flights at 152-m altitude at speed of 49 m/sec.

STATION 6 STATION 9

Q = 84O #3 = 90' #3 = 19'

OF

EVENTS 2

60 70 80 90 100 60 70 80 90 1 0 0 60 70 80 90 1 0 0

MAX [MUM LEVEL, d B(A)

Figure 8.- V a r i a t i o n i n maximum dB(A) levels f o r f i v e six-degree landing approach f l i g h t s over ROMAAR.

AVERAGE OF ON-TRACK V A R I A T I O N MAX. LEVELS, dB(A) I 1 I J LATERAL LATERAL V A R I A T I O N D I STANCE, k m 70 dB(A) ESTIMATED CONTOUR I 0 2 4 6 8 DISTANCE ALONG GROUND TRACK, k m Figure 9.- Noise v a r i a b i l i t y along ground t r a c k using ROMAAR f o r 152-m a l t i t u d e and 49-m/sec airspeed.

four l e v e l flybys a t ESTIMATED CONTOURS

80 d B ( N

AVERAGE OF MEASURED

----

75 d B ( N

M A X I M U M VALUES FOR --c-

70 d B ( N

F I V E FLIGHTS LATERAL D 1 S TANC E, km

I I

I 1

4 6 0 2

D ISTANCE TO TOUCHDOWN, km

Figure 10.- Ground n o i s e contours from f i v e f l i g h t s u t i l i z i n g six-degree approach over R O W .

FLIGHT DIRECTION Q ESTIMATED NOISE 60 60 61 62 63 63 65 67 65 64 64 62 60 60 60 PATTERNS I 62 62 63 64 65 66 67 66 67 66 66 64 62 61 61 80dB(A) 75 dBtA)

63 63 65 66 69 71 'h.. 68 65 64 63 63

-.+.- 70 dB(A) 65 65 67 ' I FROZEN'' POSITION OF HELICOPTER -65 -66 -67- - 6 5 ~ - RUNWAY$ 65 65 67 65 2_ 70 dB(A1 GROUND 64 64 65 64 INTERSECTION POINTS 62 64 66 66 65 64 66 68 66 65 65 66 64 64 62 I 61 62 62 64 62 61 64 67 64 62 62 64 63 62 61 LEVELS FOR ALL 6 LEVEL FLY-BYS Figure 11.- Ground n o i s e d i r e c t i v i t y patterns f o r six level flybys over l i n e a r array at 152-m a l t i t u d e and 49-m/sec airspeed.

152-m ALTITUDE, 49 m/s 76-m ALTITUDE, 49 m/s $.

k--1 km+ 152-m ALTITUDE, 82 m/s 76-m ALTITUDE, 82 mls INTERSECTION

1 PO I NTS

Figure 12.- Variations in 75-dB(A) noise directivity patterns over linear array for altitudes of 76 and 152 m and airspeeds of 49 and ,82 m/sec.

A STATIC ACOUSTIC SIGNATURE SYSTEM FOR THE ANALYSIS OF DYNAMIC FLIGHT INFORMATION DANIEL J. M E R U.F. ARMY ARMAMEElT RESEARCH AND DEVELOPNENT COMMAND SUMMARY The Army family of helicopters was analyzed to measure the polar A static octave band acoustic signature i n various mcdes of flight.

array of calibrated microphones was used to simultaneously acquire the signature and differential times required to mathematically position the aircraft i n space. The signature was then reconstructed, mathematically normalized to a f i x e d radius around the aircraft.

INTRODUCTION A number of years ago, this organization was asked to measure t h e polar octave band output of each of the Army family of helicopters i n various modes of flight. The i n t e n t was to, f i r s t , model the perception, based on t h i s signature study, of an unsophisticated enemy force who would be expected to f i r e on friendly aircraft w i t h small arms and, second, pzedict Since each aircraft was to and minimize attrition based on f l i g h t tactics.

be traveling a t some considerable velocity during data acqusition, some means of measuring the output through 360 degrees had to be devised.

of microphones was rejected as was towing a Ehplacing a circular array microphone behind a lead aircraft. It was decided to combine t h i s organi- w i t h signature analysis to mathematically zation's sound ranging techniques t h e desired family of polar signatures.

construct This paper, then, w i l l describe t h e application of a measurement technique normally restricted to transient phenmena to that signature task.

SYMBOLS C velocity of sound L distance between any two adjacent microphones differential propagation distance between f i r s t and second "1 excited microphones differential propagation distance between f i r s t and third excited "2 microphones R radial distance between origin microphone and sound source velocity of propagation VS vector average of wind velocity during sound propagation V W X one Cartesian distance between origin microphone and sound source remaining Cartesian distance between origin microphone and sound Y source

8 propagation m e d i u m s h i f t

differential time between f i r s t and second excited microphones A t 1 2 differential time between f i r s t and third excited microphones A t13 THE THEORY AND APPLICATION OF SOUND RANGING Sound ranging is a technique which can be used to accurately determine t h e location of an impulsive sound source i n three-dimensional space.

This technique was developed to measure munition performance i n jungle canopy where radar and optical techniques are impractical.

To implement sound ranging data acquisition, a geometric array of microphones is emplaced i n a surveyed position a t a target site, The shape of t h e array was determined by error analysis of various mathematical models, A typical array consists of from 3 to 6 microphones, although any number greater than three could be used. Sound emanating from the source propagates through t h e air a t a relatively constant velocity, and the wave- front arrives a t each microphone a t a specific time after t h e sound has occurred. The differences i n arrival time among the various microphones are accurately determined. Equations, derived from the Euclidean geometry, the velocity of sound, and the differential times between the microphones, w i l l yield the position coordinates of t h e sound source.

A typical two-dimensional sound ranging transducer array is shown i n the mathematical solution can be figure 1. For t h i s transducer system, obtained f r m the geometry as follows: expanding (2) and (3) and substituting (1) for x, ~ Z r 2 ~ n + n 2 = ~ 2 - y 2 + ~ 2 - 2 ~ y + y 2 (4) 1 1

~ 2 1 - 2m + n 2 2 = R 2- y 2 + 4 ~ 2 + y 2 - 4 ~ y (5)

which reduces to

2 ~ n + n 2 = L 2- 2 ~ y

1 1

2 ~ n + n 2= 4 ~ 2 - 4 ~ y

2 2 solving for R from (6)

( ~ 2 - 2Ly - n12)

R = 2nl substituting in (7)

2 ( ~ 2 - 2 ~ y - n ~ ) 2 n + n 2= 4x12- 4 ~ y

2n solving for y (8) yields R and substituting R and Substituting t h i s value of y i n y i n (1) yields x.

A second array, either p a r a l l e l or perpendicular to the f i r s t , yields a second set of coordinates which a r e cclmbined with the f i r s t set to derive the third dimensional distance, 2.

There are significant sources of error w i t h t h i s approach, however.

Signal-to-noise and resolution errors can be a problem but, since these are related to the nature of t h e specific acoustic signature produced by t h e test item, for t h e purposes of t h i s paper w i l l not be discussed.

That leaves the assumption that the velocity of sound is constant, which of coulse, it is not. Wind and wind’gradients and, to a lesser extent, temperature gradients w i l l a l l distort the velocity of propagation. Wind, more properly, represents the migration of the total m e d i u m during propagation, but for t h e purposes of error analysis can be considered a propagation velocity s h i f t (figure 2 ) .

In figures 3 and 4, the errors i n x and y induced by a 1% error i n t h e velocity of propagation are shown. After a f i r s t glance, there is a tendency t o structure the test to place the microphones a t several times the microphone array dimension from the test item t o minimize this type of error; however, i n reality, the larger t h i s distance, the more w i l l become. Thus the trade-off must be appreciable velocity gradients made between improved accuracy i n constant wind velocity, against reduced accuracy caused by topographically induced wind gradients.

An error correcting scheme for wind has &en devised; computer modeling indicates accuracy improvements of up to an order of magnitude result.

Briefly, t h i s technique involves the storage of total vector anemometer history, the computation of position coordinates based on a constant propagation velocity, the reconstruction of propagation velocity based on wind data over the computed distance, the recomputation of coordinates based on the new velocity of sound, and the reconstruction and recamputation repeated u n t i l t h e result is converged upon. The tolerance between successive recomputations determines the e x i t from the algorithm.

The hardware necessary to execute this technique include t o t a l vector anemometers, microphones, l i n e drivers, transient recotders, t h e code generator and digital processor (figure 5). T h i s system is designed to store data and compute the coordinates of a transient forcing function.

The techniques, however, are applicable t o repetitive transient producing vehicles, such as helicopters.

APPLYIJS SOUND RANGING TO HELICOPTERS Having been given t h e task of measuring the polar signature around an aircraft i n flight, it was decided to f l y t h e aircraft by a sound ranging array of calibrated microphones t o simultaneously acquire the signature and position information. A magnetic tape m e d i u m was selected since t h e Real-Thie Sound Ranging System was incapable of keeping up w i t h the data rates. A data processing scheme was devised to compute the position of the aircraft i n polar coordinates, measure the r m s amplitude of each octave band, correct those amplitudes for spreading loss and absorption, and finally, list t h e polar distributions of octave band energy.

Clearly, this methodology depended on the ability to distinguish a long vavelength repetitive phenmenon i n t h e acoustic signature of each aircraft to utilize sound ranging; and, conveniently, the helicopter slade slap provided just such a waveform. A Huey Cobra, for example, w i l l typically produce a blade slap rate of 14Hz, a Huey approximately 12Hz.

rhus, w i t h fundmental wavelengths i n excess of 23 meters, a substantially This permitted sized microphone array could be surveyed i n place.

scceptable accuracies over the distances required to measure the sircraft signatures a t low angles of incidence. Further, those long davelengths avoided the necessity of keeping track of multiple cycles vhen the wavelength is exceeded by the inter-microphorie distances.

This data acquisition and reduction task was greatly simplified by the funding organization's request to operate i n still air, for i n addition to the instrumentation, acccanpanYing personnel took part in simultaneous threshold perception tests.

To accurately range Figure 6 illustrates the reduction hardware.

the aircraft during data reduction, it was necessary to insure differential t i m e measurements between the same point on each succeeding cycle of acoustic signature. Thus, phase lock loops were u t i l i z e d to lock onto the fundamental blade slap frequency transduced by each microphone. Each channel from the magnetic tape recorder drove a phase l o c k loop; the lock range was adjusted to a reasonable range around t h e fundamental blade slap frequency for each aircraft and the loop time constant made long, approximately 1 sec, to "flywheel" over dropouts caused by topographical multipath, the tape m e d i u m o r air turbulence. The square wave output of each phase lock loop was differentiated to yield a pulse of 100 psec i n width which, i n turn, drove both t h e stop and s t a r t enable ports of each counter and the f irst-in start logic.

The first-in s t a r t logic generated a s t a r t pulse for each counter only when it received a pulse from the f i r s t arrival microphone It then counted t h e remaining microphone outputs and circuitry.

started over again; since t h i s circuit produced a 50 psec wide pulse, the counter associated w i t h the f i r s t excited microphone always read zero. The s t a r t logic also contained sequencing logic that would permit the selection of the i d e n t i t y of the f i r s t excited microphone (this was predictable since t h e aircraft flew a prescribed path); this avoided getting locked into an improper data acquisition sequence.

Thus, a t the end of one multimicrophone cycle, each counter would be The stopped a t some time with respect to t h e f i r s t counter stimulated.

The transition busy b i t was generated when any counter was counting.

to busy cued the Analog-to-Digital Converter to converge and the transition to not busy cued the computer to f i r s t , accept t h e data from the counters and, second, test end-of-conversion u n t i l the digitization was ready for transfer.

The computer sampled the data asynchronously, buffering sufficient data to reform the polar signatures. The computer terminal was used to periodically enter the aircraft type, flight mode and temperature (to calculate t h e velocity of sound) . The program did the rest, listing the results on the l i n e printer as per the funder's preference.

CONCLUSION Figure 7 illustrates the format of the data as reported and figure 8, the graphical representation. It was found that t h i s technique worked extremely w e l l on a variety of aircraft whether the blade slap was particularly audible or not. To d a t e , > i t i s n ' t clear whether a wind correcting algorithm w i l l be possible for this technique b u t one w i l l be persued with the interest of a funding organization. I t was found, however, that i n relatively still air, this signature technique is both cost effective and timely.

Figure 1.- 3 microphone vertical array.

TPROPAGATION = RMEASURED/”S

6 = (VW) (TPROPAGATION)

Figure 2 . - Plan view of 3 microphone array in wind.

MIC3 X MIC2 MICl , x .

1L 2L 3L 4L 5L 6L 7L 8L F i g u r e 3 . - E r r o r i n x caused by 1% e r r o r i n V .

S y 40% 20% 12% 6% 3% 2% 1 % 4L 3L MIC3 2L MIC2 1L MlCl X 1L 2L 3L 4L 5L 6L 7L 8L Figure 4 . - E r r o r i n y caused by 1% e r r o r in V,.

1 .

TOTAL TOTAL

VECTOR I I VECTOR

REAL-TIME PARAMETERS: ANEMOMETER ANEMOMETER

I

0 TEMPERATURE (. . Vs) 0 TOTAL VECTOR WIND VELOCITY 0 TIME 0 PRESENCE OF BURST 0 TIME OF BURST CALCULATED PARAMETERS: e vs AT BURST TIME 0 INTEGRATION OF WINO TIME HISTORY BUFFER/MULT I PLEXERiANALOG DETECTORS 0 CORRECTIONS FOR TEMPERATURE GRADIENTS AT ARRAY CALCULATED RESULTS: 0 THREE-DIMENSIONAL POSITION COORDINATES OF BURST GENERATOR 0 TIME OF BURST e DURATION OF FUZE TIME

I *"'""*'I [ I

PROCESSOR - PRINToUT

Figure 5.- R e a l - t i m e sound ranging block diagram.

CONVERTER Figure 6.- Data processing system block diagram.

53.80 dB R U N NUMBER 1 AMBIENT NOISE LEVEL SPLATT SPLU THETA SPLC SP1.R 0 . 0 E 0 . 5 1 14.84 81.37 20.86 2 0 . 0 8 78.28 18.32 0 . 0 59.97 25.32 79.01 16.41 0 . 0 62.60 0.0 61.01 30.38 75.97 14.95 0 . 0 61.77 39.60 74.72 12.94 45.73 77.68 11.93 0.0 65.75 50.12 78.95 11.93 0 . 0 67.62 59.74 81.54 10.30 0.0 71.24 0 . 0 69.55 63.95 79.51 9.96 0.0 74.47 68.37 84.13 9.66 75.93 54.10 9.29 0 . 0 74.81 85.07 9.18 0 . 0 75.89 79.38 0.0 78.38 84.94 87.45 9.06 0.0 80.37 90.00 89.40 9.03 95.06 88.02 9.06 0 . 0 78.96 0 . 0 79.45 100.62 88.63 9.18 104.07 88.04 9.29 0 . 0 78.74 87.49 9.66 0 . 0 77.83 111.63 0.0 77.53 116.05 87.49 9.96 0 . 0 74.98 120.26 85.28 10.30 129.88 83.48 11.33 0.0 72.51 134.27 83.13 11.93 0.0 71.20 0.0 68.75 140.40 81.69 12.94 0.0 63.06 149.62 78.02 14.95 154.68 75.28 1 6 . 4 1 0.0 58.87 159.92 71.41 18.32 0 . 0 53.09 0 . 0 51.77 165.16 72.63 20.86 Figure 7.- Typical computer printout.

Figure 8.- Typical polar plot.

AN ACTIVE NOISE REDUCTION SYSTEM FOR AIRCREW HELMETS Peter D. Wheeler and David Rawlinson Wolfson Unit for Noise and Vibration Control, I.S.V.R.

Sou t hamp ton University Stephen F. Pelc and Tony P. Dorey Department of Electronics, Southampton University INTRODUCTION In a high noise environment conventional ear-defenders may be incapable of ,roviding sufficient noise attenuation.

An active noise reduction system (ANR) h a s been developed for use in aircrew flying helmets (ref. 1) in which the icoustic noise field inside the ear defender is detected using a miniature iicrophone and an antiphase signal is fed back to a communications telephone Jithin the ear defender. The feed-back loop also affects the communications gignal which has to be compensated outside the loop in order to retain the )riginal transfer function from telephone amplifier to listeners ear.

The concept was not new at the outset of the work reported. In 1953 Olsen and May (ref. 2) investigated the possibility of creating a noiseless zone in €Pee field conditions, but the technique had not apparently been successfully spplied to headsets. However, the Procurement Executive, Ministry of Defence had outlined a method which proved the basis of this present work.

The major engineering development involved in integrating the technique into a conventional aircrew flying helmet was associated with the identification and characterisation of suitable microphones and telephones. At the time of ?revious publication (ref. 1 ) the feasibility of active reduction had been iemonstrated though tests with random noise had not been undertaken. Some instability problems still existed and the performance with a communications signal had not been assessed.

During the last two years the remaining engineering problems have been svercome and an extensive program of laboratory subjective trials has been completed. In-flight trials of the system are currently underway. Throughout this programme of work there has been active participation by the Royal Aircraft Establishment, Farnborough, and the development has been carried out with the support of the Procurement Executive, Ministry of Defence, U.K.

PRINCIPLES One channel of the complete system is shown schematically in Figure 1 .

The design is based upon the existing ear defender shell and seal, and incorpo- rates a high-fidelity type moving-coil telephone and a miniature electret mi crophone.

A complete analysis of the system shows that the pressure at the ear (Pel is comprised of two components due to the signal (Sv) and the external noise field without ANR (np) where A,B,K,T1 and T2 are the transfer functions of the amplifier, feedback loop, acoustic cavity, telephone and microphone, respectively.

It is apparent that the amplification of the two components is different and that if the signal is pre-emphasised, to compensate €or its attenuation Sv by the feedback loop, and returned to its original level then an improvement of a factor ( 1 + ABKT1T2) can be achieved.

The product of the terms is of fundamental significance as it KT1T2

represents the overall electrical transfer function of the telephone - cavity -

microphone system. To provide ANR, the product AB would ideally have a complementary variation with frequency. In practice the frequency response must be tailored to ensure that when the loop gain approaches unity the total loop phase shift does not approach 1 8 0 ' .

The systems ability to match this transfer function KT1T2 will determine the degree of noise reduction which is obtained.

EVALUATION The ANR system has been comprehensively tested in a series of laboratory trials, prior to flight trials by the Royal Aircraft Establishment.

Two basic methods of measuring ANR have been used throughout the project.

These are: a ) An electronic computation method An external broadband noise method b) The computer method used a single frequency excitation which was analysed by means of an analogue computer circuit that calculated the value of and continuously plotted its amplitude and phase as a function of 1 + ABKTITZ frequency.

In the second method, preferred because of its close approximation to the real life situation, subjects wearing the ANR modified flying helmet were 5 46 zxposed to an external noise field similar to that experienced by pilots in a iigh performance strike aircraft. Comparisons of attenuation and speech intelligibility scores, with and without the ANR system in o m d the modified helmet's attenuation was also checked €lying helmet, Objective measurements of helmet attenuatio nicrophones placed at the subject's pinnae, and o These two methods of measurement were carefully compared in a structured ieries of eight experiments to check that the more convenient laboratory mnputational method gave the same results as the noise excited experiment.

Che intelligibility testing was carried out using anglicised modified Rhyme Pest material comprisingfour sets of fifty initial or final consonant words blus twenty five central vowel words. The subject was required t o score his feception of a word by marking a score sheet showing all six rhyming alterna- :ives for a given word and, at the end of the experiment, the sheets were

karked by the experimenter. For each condition - ANR on or off - two of the

four word lists available were presented and the number of subjects was chosen :o give a fully balanced experiment with a satisfactory level of statistical The speech signal presented to the ear was matched, for all the iignificance.

ielmet configurations tested, to within rt 1 d B ( A ) overall and -L 1 dB in each ! I 3 octave band.

Eighteen subjects were tested twice in each configuration and the mean .mprovement in attenuation was found to be 12 d B ( A ) for the noise field used :Figure 2 ) .

Figure 3 shows the ANR performance as a function of frequency, together 6th a measure of the scatter of results. The comparison between mean helmet rttenuation values is shown in Figure 4 .

The mean improvement in speech intelligibility with the ANR system in tperation was 21% relative to a baseline of 51% when the system was switched )ff. The individual performances of subjects are shown in Figure 5 .

All those tested commented upon the increased comfort provided by the iigher attenuation of the ANR system, particularly at lower audio frequencies, rhere the aircraft's noise energy is concentrated.

CONCLUSIONS The use of active noise reduction in a laboratory trial simulating flight conditions has been shown to give encouraging objective and subjective results.

The system is capable of application to a wide range of situations where high attenuation (with or without voice communications) against low frequency noise is required. The response of the ANR system may be engineered, to an extent, to optimise attenuation for particular noise fields, such as are experienced in rotary wing and fixed wing aircraft, and for certain applications, the system may be fitted into a small, battery operated, electronics package which may be easily worn on the person.

REFERENCES 1. Dorey, A.P., Pelc, S . F . , and Watson, P.R. 1 9 7 5 , "An Active Noise Reduction System for use with Ear Defenders". 8th International Aerospace Symposium, Cranfield, 24-27.

2. Olsen, H.F., and May, E.G. 1 9 5 3 , J.A.S.A. 25 , No.6.

-

EXTERNAL \ NOISE

EAR D E F E N D E R \ \ .- - \

\ Figure 1.- A system block diagram.

n a PI a C Y 9) w v F 4 a * 90

&

a0 125 250 500 lk 2k 4k 1/3 octave centre frequency, Hz Figure 2 . - Simulated aircraft cockpit noise.

ean for 18 subjects

- .I--

e --

octave centre frequency, Hz Figure 3.- ANR performance.

ANR-modified

%

n z

3 20

EJ 10 H I * * 1 : - 1 . . 1 ' f ' * I lk 2k 4k 125 250 500 1/3 octave centre frequency, Hz Figure 4 . - Mean helmet attenuation.

ANR on - mean 72%

ANR off - mean 51%

a

8 60

c n @ 50 ae 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 SUBJECT Figure 5.- Word scores for 18 subjects, DESIGN OF HELICOPTER ROTORS TO NOISE CONSTMINTS Edward G. S c h a e f f e r and Harry S t e r n f e l d , Jr.

Boeing V e r t o l Company SUMMARY Results from the i n i t i a l phase of a r e s e a r c h c o n t r a c t , "Study of Design Con- s t r a i n t s on H e l i c o p t e r Noise," NAS1-15226, sponsored by t h e NASA Langley are p r e s e n t e d . A d e s c r i p t i o n of t h e o v e r a l l program i s in- Research Center cluded. Basic c a l c u l a t i o n s of nonimpulsive r o t o r harmonic and broadband hover n o i s e s p e c t r a , over a wide range of r o t o r d e s i g n v a r i a b l e s w e r e accomplished; and t h e s e n s i t i v i t y of PNL t o changes i n r o t o r desigfi parameters are presented.

Measured r o t o r n o i s e d a t a w e r e used t o c o r r e l a t e t h e c a l c u l a t i o n s i n v e r i f y i n g t h e p r e d i c t i o n methodology.

INTRODUCTION Increased emphasis on reducing t h e n o i s e generated by h e l i c o p t e r r o t o r s t o minimize a u r a l d e t e c t i o n t i m e s i n m i l i t a r y a p p l i c a t i o n s , and i n c r e a s e community acceptance d u r i n g commercial o p e r a t i o n s , now r e q u i r e t h e h e l i c o p t e r manufac- t u r e r t o c o n s i d e r n o i s e c o n s t r a i n t s of h i s product e a r l y i n t h e d e s i g n phase.

Impending n o i s e r e g u l a t i o n s , such as t h e FAA/ICAO p o s s i b l e n o i s e l i m i t s f o r c e r t i f i c a t i o n of h e l i c o p t e r s , are f o r c i n g d e s i g n e r s t o implement n o i s e c o n t r o l measures d u r i n g p r e l i m i n a r y d e s i g n performance s t u d i e s when t h e s i z i n g of r o t o r s is being dytermined.

Basic r o t o r d e s i g n parameters such as t o t a l t h r u s t , b l a d e t i p speed, d i s k l o a d i n g , number of b l a d e s p e r r o t o r , and r o t o r s o l i d i t y , which i n v a r i a b l y a f f e c t t h e n o i s e produced by t h e r o t o r , have g e n e r a l l y been decided long b e f o r e n o i s e r e s t r i c t i o n s are considered. One r e a s o n i s t h a t most p r e l i m i n a r y d e s i g n e r s do n o t have s i m p l i f i e d g u i d e l i n e s f o r p r e d i c t i n g r o t o r n o i s e which c a n be meaningful d u r i n g e a r l y r o t o r d e s i g n d e c i s i o n s s t a g e s . Consequently, most d e s i g n s 3 a r e semi-finalized, b e f o r e n o i s e estimates of t h e c o n f i g u r a t i o n can be made. Subsequent changes t h a t may b e r e q u i r e d i n reducing t h e n o i s e t o comply w i t h c e r t a i n r e g u l a t i o n s f i n d themselves i n c o n f l i c t w i t h d e s i g n s t h a t have a l r e a d y been set.

T h i s study, when completed, w i l l r e s u l t i n a g e n e r a l method, and sets of design c h a r t s , which w i l l p e r m i t e v a l u a t i o n s o f t h e n o i s e and performance t r a d e o f f s of s i n g l e r o t o r s d u r i n g t h e e a r l y d e s i g n s t a g e . The measure of performance w i l l b e t h e percentage of a v a i l a b l e r o t o r t h r u s t which must b e expended i n l i f t i n g t h e d r i v e system ( r o t o r b l a d e s , hub, and r o t o r t r a n s m i s s i o n ) .

Given a desired thrust and noise limit, the charts can be used to define the corresponding radius, chord, and tip speed for 2, 3, 4 , 5 or 6 bladed rotors.

The rotor which requires the lowest drive system weight is the optimum design.

Conversely, given a completely defined rotor the charts can be used to predict the noise.

Results from the completed initial phase of the study, which includes the cal- culation of both rotor harmonic and broadband, nonimpulsive hover noise and the relative importance of various rotor design parameters that influence changes in Perceived Noise Level (PNL) are diseussed in this paper.

j SYMBOLS AND ABBREVIATIONS Values are given in both SI and U.S. Customary Units. The calculations were made in U.S. Customary Units.

T thrust, N (lb) blade tip speed, m/sec (ft/sec) VT peak frequency, Hz fP blade area, m2 (ft2) Ab (J angle between centerline of rotor shaft and line to observer, deg one-third octave frequency band correction S j lift coefficient CL r distance to observer, m (ft) N/m Sound Pressure Level, dB (re 2 x SPL Perceived Noise Level, in PNdB PNL A-weighted network dBA C-weighted network dBC Broadband noise BB PNLT Tone-corrected Perceived Noise Level NOY Unit used in the calculation of Perceived Noise Level. It is the noisiness of a noise for which the Perceived Noise Level is 40 PNdB.

The noisiness of a noise that is judged by a subject to be n times that of a 1-NOY noise is n NOYS.

P R O G R A M he o b j e c t i v e of t h e program is t o provide a "handbook" f o r h e l i c o p t e r designers r o t o r s during t h e prelim- nd c o n f i g u r a t i o n managers t o e v a l u a t e t h e n o i s e of t h e e f f e c t on r o t o r payload.

nary design phase, and t o estimate t o o l t h a t can b e used during n o i s e n order t o produce an e f f e c t i v e d e s i g n e r ' s .nd performance tradeoff e v a l u a t i o n s , t h e t o t a l r o t o r n o i s e s i g n a t u r e h a s t o )e represented a c c u r a t e l y . A l l major sources of r o t o r n o i s e are included i n Figure 1 shows an example of .eveloping t h e design c h a r t s f o r t h e handbook.

:hese sources and t h e i r c o n t r i b u t i o n t o t h e o v e r a l l n o i s e s i g n a t u r e . The sub- (nonharmonic), jective weighting of t h e s e n o i s e sources; harmonic, broadband i d impulsive, which i s t h e prelude t o determining t h e PNL, are shown i n The engine n o i s e minor c o n t r i - 'igure 2 as t o t a l NOY values per octave band.

shown f o r completeness only. Examination of t h i s f i g u r e i n d i c a t e s w t i o n is :hat i n terms of annoyance, r o t o r impulse is t h e major f a c t o r ; b u t i f t h e :otor d i d not have an impulsive c h a r a c t e r i s t i c then broadband n o i s e predomi- iates t h e Perceived Noise Level (Figure 2) t o a much g r e a t e r e x t e n t than t h e Sound Pressure Level Spectrum (Figure 1).

t h e following phases: Che o v e r a l l study c o n s i s t s of L . Calculating t h e nonimpulsive r o t o r harmonic and broadband n o i s e s p e c t r a using e s t a b l i s h e d p r e d i c t i o n procedures recognized and used by i n d u s t r y and found i n open l i t e r a t u r e . The range of r o t o r physical parameters included i n t h e c a l c u l a t i o n s are: t h r u s t , 44 t o 356 kN 10 000 t o

;i

80 000 l b ) ; d i s k loading, 287 t o 575 N/m2 (6 t o 1 2 l b / f t ); s o l i d i t y , 0.04 eo 0.12; number of blades, 2 t o 6; and t i p speed, 152 t o 244 m/sec Calculations are f o r a s i d e l i n e d i s t a n c e of (500 t o 800 f t / s e c ) .

150 meters from r o t o r and a height of 150 meters (which corresponds t o Com- the measurement l o c a t i o n s being considered i n t h e r e g u l a t i o n s ) .

bining t h e n o i s e s i g n a t u r e s i n t o one-third octave frequency bands c a l c u l a t i n g PNL, dBA and dBC.

2. Appwqng impulsive c o r r e c t i o n s developed by t h e Boeing Vertol Co. and s u b j e e t i v e adjustments from Reference 1 t o a d j u s t dBA, dBC and PNL values t o a s u b j e c t i v e l y equivalent broadband' l e v e l .

Preparing a set of design c h a r t s t o permit d i r e c t determination of values 3.

of dBA, dBC and PNdB f o r range of r o t o r physical parameters.

An example of a p o s s i b l e design c h a r t format is shown i n Figure 3 f o r determining t h e PNdB i n hover and, providing a r a t i o n a l e showing t h e e f f e c t s o f r o t o r configuration on forward f l i g h t noise.

4. Evaluating t h e performance p e n a l t y f o r each main r o t o r , c o n f i g u r a t i o n and t i p speed combination. The r a t i o o f d r i v e system weight t o r o t o r t h r u s t s h a l l b e used as a n index bf t h e design e f f i c i e n c y .

RESULTS AND DISCUSSION Prediction of Nonimpulsive Rotor Hover Noise The harmonic rotation noise calculation was based on the method developed in Reference 2. This widely accepted rotor noise calculation includes the design variables of thrust, disk loading, tip speed, and number of rotor blades. The only change made to the equations of Reference 1 was that an airloads harmonic 1.3 was used instead of 2.0, as specified by the original decay exponent of authors. This modification reflects a more realistic airload harmonic decay of 15 dB per octave which has been measured by other researchers and provides better agreement with measured data in the higher harmonic range.

The broadband, or nonharmonic, rotor noise calculation used was from the unpublished semiempirical prediction made by Robert J. Pegg of the NASA Langley Research Center. The equation from this prediction,

= -240 log T + .746 VT + 786

fp

SPL = 10 log Ab 4- 60 log vT

10 log (COS20 + .I)

+ Sj -20 log r + f (CL) - 53.29

-

cL for E < .48

f ( C L ) = 10 log .4 -

f ( C L ) = .9 + 80 log for .48

.48 has as its design variables, thrust ( T ) , tip speed ( V , ) , blade area (Ab) and lift coefficient (CL).

A computer program was written to include all of the design variables and to provide an automatic calculation of both the harmonic and broadband noise, then combine them into one-third octave frequency bands and print-out the Figure 4 shows a sample of this output.

resultant dBA, dBC and PNL.

Nine hundred sixty computer cases were run during the initial phase of the program to provide adequate definition of the design variables for preparation of the "handbook" charts.

Prediction-Data Correlation Measured noise data, shown in Figures 5 and 6 , from a nonimpulsive and moder- ately impulsive rotor were directly compared to the calculated one-third octave SPL using the developed computer program. The agreement between predictions and measurement for the inonimpulsive case (fig. 5) are generally quite good, the discrepancy in the 500 Hz octave band is probably due to destructive inter- ference between the direct and first ground reflected waves which calculates to occur at 556 Hz. In the case of the impulsive rotor (fig. 6 ) good agree- ment is attained in the first two harmonics and higher frequency broadband noise since the harmonic noise prediction method does not account for the increase in mid-harmonic loading which typifies impulsive rotor noise.

Perceived Noise Level S e n s i t i v i t y t o Rotor Design To provide a n i n d i c a t i o n of t h e s e n s i t i v i t y of PNL t o changes i n design vari- a t i o n s , f i v e b a s e l i n e r o t o r designs r e p r e s e n t i n g d i f f e r e n t classes of h e l i - c o p t e r s w e r e i n v e s t i g a t e d .

For each b a s e l i n e c o n f i g u r a t i o n t h e r o t o r parameters of t h r u s t , d i s c loading, t i p speed and number of blades w e r e v a r i e d one a t a t i m e (at c o n s t a n t l i f t c o e f f i c i e n t ) and t h e r e s u l t a n t PNdB c a l c u l a t e d .

Figure 7 shows a n example of t h e c a l c u l a t e d nonimpulsive hover SPL f o r one p a r t i c u l a r case (-3-bladed, 89-kN (20 000-lb) t h r u s t r o t o r ) . Taking t h i s con- f i g u r a t i o n as a b a s e l i n e design and varying each of t h e parameters one a t a time r e s u l t s i n t h e PNL s e n s i t i v i t y c h a r t shown i n f i g u r e 8. Similar s t u d i e s have been done f o r f o u r o t h e r b a s e l i n e designs which cover a wide range of values and t h e r e s u l t a n t summary ( t a b l e I) i n d i c a t e s some rough g u i d e l i n e s which can be used pending release of t h e f i n a l design c h a r t s which w i l l r e s u l t from t h i s study.

CONCLUDING REMARKS The c a l c u l a t i o n of t h e nonimpulsive harmonic and broadband hover n o i s e f o r a wide range of r o t o r design v a r i a t i o n s w a s accomplished. The p r e d i c t i o n method- ology used c o r r e l a t e d w e l l with measured w h i r l tower d a t a . Application of t h e p r e d i c t i o n s t o v a r i a t i o n s i n r o t o r design ( t h r u s t , t i p speed, d i s c loading, and number of blades p e r r o t o r ) has shown t i p speed and t h r u s t as having t h e most e f f e c t on changing the PNL.

, REFERENCES 1. S t e r n f e l d , Harry, Jr.; and Doyle, Linda Bukowski: Evaluation of the Annoyance Due t o Helicopeter Rotor Noise. N A S A CR-3001, 1978.

2. Lowson, M.V.; and Ollerhead, J.B.: Studies of Helicopter Rotor Noise.

USAAVLABS Tech; Rep. 68-60, U.S. Army, Jan. 1969.

TABLE I.- INTERIM RESULTS SUMMARY O F SENSITIVITY OF PNL TO DESIGN PARAMETER VARIATION

Sensitivity *

Parameter Range 137 to 290 m/sec 2 to 5 PNdB per 30.5 m/sec Tip speed (450 to 950 ft/sec) (100 ft/sec) 11 121 to 358876 N 2 PNdB per doubling of Thrust (2 500 to 80000 lb) thrust 96.1 t o 574.6 N/m2 0.5 PNdB per 96.1 N/m2 Disk loading (2 lb/ft2) (2 to 12 lb/ft2) <0.5 PNdB per blade addition 2 to 6 Number of blades per rotor *Based on varying parameter under study while holding all others constant.

ALTITUDE 1 2 0 m (394 ft) GROSS WEIGHT i a 140 kg (40,000 Ib) TIP SPEED 2 3 4 m/sec (769 ft/sec) AIRSPEED 1 1 1 km/hr (69 mi/hr) SOUND PRESSURE

LEVEL - dB RE

2 X 1 OJ N/m*

OCTAVE BAND CENTER FREQUENCY - HZ

Figure 1 . - Helicopter noise source contribution during 6-degree approach.

ALTITUDE 120 m (394 ft) 6' GROSS WEIGHT 16 140 kg (40,000 Ib) TIP SPEED 234 m/sec (769 ft/sec) AIRSPEED 111 km/hr 4c (69 mi/hr) noys PNL - 107 PNdB 3c

OCTAVE BAND CENTER FREQUENCY - HZ

Figure 2.- S u b j e c t i v e weighting of h e l i c o p t e r n o i s e during 6-degree approach.

DISK LOAD RANGE

96.1 - 575N/m'

(2-1 2 Ib/ft*) TIP SPEED (VT) TIP SPEED (VT) RANGE 152-244 m/sec RANGE 152-244 m/sec (500-800 ft/sec) (500-800 ft/sec) PNdB V SOLIDITY RANGE 0.04-0.1 2 0.04-0.1 2

vTtvlIN { 7

L I I I I THRUST

RANGE 44 4 8 4 - 355 876 N (10,000-80,000 Ib)

Figure 3.- P o s s i b l e design c h a r t format f o r 2-, 3-, 4-, 5-, and 6-bladed r o t o r s .

ati T O T A L T O T A L C @ A oae C B C 7 5 . 7 7 6 . 9 8 5 . 9 I TJTAL ' n 1 5 ~ .

.,.... L'fiPt. R L ' G r , r l s E I ) . . . . I

................. l % , . ? * I l , : V ' L h . l % . . . . * . . . I . . . . . . . .

rwo FREU 0AhD +PL 0AlrD 5 P : wanrwic FHEO SPL .... .... 1 . .... .. ....

........ ...... *)....

a0 81.1 19.5 07.1 20 87.1 25 .U 1 5 55.1 25 27.0 39.1 PO97 31 .V 31 5b.1 31 34.1 5U.b 15.b 72.7 40 8 0 . I 40 80,7 *O 82.9 4 78.2 50 .U 50 5 1 . 1 5 0 5$.1 97.7 70.9 63 7 5 . n 63 61.1 69.4 63 75.0 6 117.3 8C 85.8 80 7 2 . t 80 7 3 . 1 7 13b.P 68.3 ion 7 0 , ' ) 71.6 100 1 G O 108.1 8 1 5 6 . 4 67.3 125 71.1 72.1 1 . 3 125 136.2 9 175.9 bb.4 1 b0 7 1 . .

1 9 5 . . 6 5 . 6 160 69.1 16C 171.6 10 205 6 P . 3 200 7 0 . 9 ilL.2 215.1, ba.9 2cc 1 1 71.7 7 7 2 . e 23a.q 66.3 61.5 251 25C 1 P 2 5 0 315 72..

7 4 3 . 2 354 * I 65.7 6 8 - w 315 315 13 e00 71.0 *cc .3?.5 1'1 273.6 61.2 400 66.'

70.2 24u.9 293.2 6 2 . 6 500 66.6 63.3 hRb.5 312.1 600 66.1 630 lb 62.2 630 8 6 6 . 9 800 61.1 67.5 332.2 b1.7 1000 .U 1000 6 6 . 0 1 c c c l : b 9 . 7 351.P 61.3 1x 1250 b5.B l Z 5 0 .U 1250 1373.9 19 371.3 60.9 1600 . u 1600 6 . ~ 3 16CG 1729.8 390.9 60.6 2c .U ?OOO b2.Y 2coc 2179.. $10.1 60.2 2000

." 2500 -1.2

430.0 2500 25cc P I * * . ' 22 59.9 .V 61.1 3150 3150 3k59.6 4 4 9 . 1 49.5 3150 60.8

." 4000

4ccc 4358.8 469.. 3Y.2 4000 5000 59.4 4 8 8 . 6 511.9 m o o .U 50GC 5,91.8 25 b3OC .O 6300 5 f . Y C3CC 6319.3 508.1 5 8 . 7 2 b 8OCC .U 8000 56.1 e c c c 8717.7 527.7 5a.a 10000 5 a . b 547.2 5 1 1 . 1 10000 .U lCCcC 10q83.6 28 566.8 29 57.5 5 8 6 . 7 30 57.6 31 605.9 5 1 . 4 6 2 5 . . 5 7 . 2 33 645.0 56.9 3r 664.5 56.7 35 614.3 56.S 703.5 3 b 5b.3 37 723.1 56.1 38 742.7 5b.9 762.2 s+.7 40 7 1 1 1 . 8 55.6 B Figure 4 . - Rotor noise calculation - computer program sample output.

.*.'.-.*,.. L-l*,>l'.r"

LEVLL - 1'8

5 9 . 8 1 5 . 8 6 3 13.1 11.6 1 C C 12.7 125' 11.4 1 6 0 1 0 . 9 6 . 7 9 2 C O 11.7 2 5 0 7 . 7 5 1 2 . 4 0 . 7 9 11.U 4 C O 8 . 5 5 1 0 . 2 I . 1 C 5 c c b9.3 6 3 C 7.63 b l . 5 b . ? * 8 C C b6.C 1 C C C O i i 5 0 3 . 8 1 E L C 0 . 5 : b Y . 3 l t C C 8 . C h 6 2 . 3 d . + l E C C C 51.2 2 5 C C 8 . 5 8 3 1 c c bl.1 9.13 b D . 9 4 c c c 9.31 5Y.4 sccc 7.55 .oo 37.9 6.30 * 00 63c; bb.1 4.58 * 00 B C C C 5 ' 4 . 6 l C 0 C I : 3 . 3 6 * 00 P h L T = 91.3

FhL - F 3 . S

' C e K L E C l l F h 1 - 5 5 C Figure 4 . - Concluded.

ROTOR THRUST 66.727 N ROTOR TIP SPEED ROTOR SOLIDITY DISK LOAD BLADE RADIUS 1 / 3 OCTAVE BAND CENTER FREQUENCY - HZ Figure 5.- C o r r e l a t i o n of c a l c u l a t i o n s with w h i r l tower nonimpulsive r o t o r n o i s e a t 152-m (500-ft) distance.

ROTOR THRUST 266,907 N (60,000 Ib) 9 0 ROTOR TIP SPEED 229 m/sec (750 ft/sec) ROTOR SOLIDITY 0 09 DISK LOAD 430 N/mz (9 Ib/ft') BLADE RADIUS 14m (46 ft) SOUND PRESSURE

LEVEL - dBRE 6o

2 X 10 -sN/mz 1/3OCTAVEBANDCENTERFREQUENCY- Hz Figure 6 . - C o r r e l a t i o n of c a l c u l a t i o n s w i t h w h i r l tower impulsive r o t o r n o i s e a t 152-m (500-ft) distance.

BASE POINT VALUES THRUST 88,969 N (20,000 Ib) TIP SPEED DISK LOADING SOLIDITY LIFT COEFF

Figure 7.- Calculated - nonimpulsive hover noise 3-bladed rotor at

152-m (500-ft) distance.

/ /' -6- / - -8 I I I I 1 I I I I I Pg 259 260 1Bs 188 (7601 laso) (SWl (m) ( S O 1 TIP WEED. m / m (fthd I I I I I 1 2 3 4 6 BLADES PER ROTOR Figure 8 . - Relative change in PNdB with design parameter variation.

THE COST OF APPLYING CURRENT HELICOPTER EXTERNAL NOISE REDUCTION

METHODS WHILE MA I NTAI N I NG REAL I STIC V EH I CLE PERFORMANCE*

MICHAEL A. BOWES

KAMAN AEROSPACE CORPORATION

SUMMARY

Analytical methods were developed and/or adopted for calculating he1 icopter

component noise, and these methods were incorporated into a unified total

vehicle noise calculation model. Analytical methods were also developed for

calculating the effects of noise reduction methodology on he1 icopter design, performance and cost.

These methods were used to calculate changes in noise, design, performance

and cost due to the incorporation of engine and main rotor noise reduction

methods. All noise reduction techniques were evaluated in the context of an

established mission performance criterion which included consideration of

hover ceiling, forward flight range/speed/payload and rotor stall margin.

INTRODUCTION

The helicopter, which has long been considered an essential military tool ,

is now enjoying rapidly expanding use in civilian applications. The size

o f the civil helicopter fleet, which numbered less than 1000 vehicles in

1960, has now grown to over 5000, and continued expansion is anticipated

(figure 1 ) . This growth in fleet size, coupled with a corresponding increase

in the type and number o f civil missions being performed, has caused a

o f and reaction to he1 icopter noise in the community.

heightened awareness

As a consequence of this, the Federal Aviation Administration (FAA) has

taken steps to formulate a helicopter noise certification rule, which will

limit the allowable noise of future design helicopters in much the same way.

that the existing Federal Air Regulation (FAR) Part 36 limits jet transport

noise.

In establishing such a rule, consideration must be given to the needs of the

Consequently, helicopter operator as well as the desires of the community.

definition of a reasonable specification requires knowledge of both the

communi ties' subjective acceptance of he1 icopter noise, and the technological

and economic aspects of helicopter noise reduction. The study which forms

the basis for this paper (reference 1) was directed towards the technological

and economic aspects .of the problem. Specifically, the ob tive of this

study was to determine the degree of noise reduction obtainable with current

helicopter noise reduction technology, and the cost of applying this

technology.

* This study was sponsored by the U. S . Department of Transportation, Federal

Aviation Admi ni stration , under Contract DOT-FA76WA-3791.

The technical effort included the development of a unified method for predicting helicopter vehicle noise including the noise contributions of the rotors, engines and drive system.

Analytical methods were also formulated for deter- mining the impact of noise reduction on vehicle design, performance and cost.

These tools were then used to estimate and compare the benefits and costs of alternative noise reduction methods, within the context of established vehicle performance criteria.

STUDY APPROACH The noise signature of a helicopter is composed of contributions from the rotors, engines and drive train. Well developed technologies exist for reducing the noise generated b y each of these components. However, because these contributions combine in a complex, spatially and frequency dependent manner it is not possible to evaluate these noise reduction methods on an isolated component basis. Component noise reduction methods, therefore, must be evaluated within the context of the total he1 icopter noise signature. This requires the use of a noise calculation method which, although capable of estimating the combined noise contributions of all components, still retains a high degree of detail for estimating the noise output of each individual component. This "systems" approach to helicopter noise modeling was applied in the present study.

The noise calculation approach discussed above provided the means for evaluating the potential for helicopter noise reduction. To apply these approaches real istically, however, it was necessary to determine the nature and extent o f changes in vehicle design and performance character- istics which must be made to incorporate noise reduction methodology.

This information was also required to assess the economic cost of heli- copter noise reduction.

.( The intent of the overall study effort was to determine how much noise reduction can be achieved in future design civil helicopters using existing noise reduction technology, and what changes in total life cycle cost will result from the achievement of this noise reduction.

Since the study concerned itself only with future design civil helicopters, it was necessary to make certain assumptions as to the nature of these vehicles and what their noise and cost characteristics would be if The effects of noise reduction was not considered in their design.

noise reduction could then be determined relative to these baseline char- acteristics. In the study program it was assumed that future design

civil helicopters will be required to perform similar missions to those

presently being performed. Since vehicle design is principally a function of required mission performance, it was further assumed that future civil helicopters will be similar in design to existing vehicles.

These assumptions lead directly to the use of existing civil helicopter

characteristics (table 1) as the baseline for determining changes due to the

incorporation of noise reduction methodology.

The basic premise of the study was that noise reduction of future civil

helicopters will be achdeved in addition to, rather than at the expense Df, required mission performance. Noise-reduced vehicles will fly as fast, as high, as far and with the same payload, although they may be heavier and more costly to own and operate. This concept of a constant mission performance requirement provided a realistic context within which the effects of helicopter noise reduction could be determined and assessed.

To apply this approach, the following mission performance criteria were established: 1. Constant payload.

2. Constant out-of-ground effect hover ceiling.

3 . Constant range (at the best cruise speed of the baseline vehicle).

4. Adequate (equal or greater) stall margin.

In general terms, these criteria were applied in the following manner, as

illustrated in figure 2, beginning with a baseline reference vehicle con-

figuration having known performance characteristics. First, the direct effect of the introduction of a noise reduction method was determined in terms of a change in vehicle gross weight at constant payload.

This new gross weight was then used to establish a new installed power requirement, and the consequent changes in engine weight and rated fuel consumption rate which result from this change in installed power.

Installed power, as well as engine weight and rated fuel consumption, were also changed to reflect any direct effects of noise reduction such as engine silencer losses. Weight and installed power changes were then iterated until a combination was arrived at which satisfied the base1 ine vehicle out-of-ground effect hover ceiling capability.

The above procedure resulted in a vehicle configuration which could operate at the same altitude with the same payload as the reference configuration.

Forward flight performance was then considered in order to satisfy the

Given the new vehicle established range and speed capability criteria.

gross weight determined by hover performance requirements, rotor stall margin was calculated and compared to that of the baseline vehicle.

If insufficient stall margin was indicated, changes in rotor design

were effected, which increased stall margin to that of the baseline configuration. Any changes in weight which resulted from these rotor design changes were calculated and accounted for, iteratively, through reconsideration of the .hover performance requirement. Once stall margin and hover performance were determined to be consistent with the established criteria, forward flight power required for the new vehicle configuration was calculated.

Forward flight power required was determined for flight at the best cruise speed of the baseline vehicle. This power was then used to determine the need for any change in fuel load required to maintain a maximum range equal to that of the baseline vehicle. If fuel load was changed, vehicle gross weight was adjusted accordingly and, again, compensated for through consideration of hover performance and stall margin criteria.

The new' vehicle configurations resulting from the above procedure were often substantially different in design from the baseline vehicles. In

general, these new configuations showed changes in gross weight, air-

frame weight, installed power, engine weight and fuel load. These

changes in vehicle design were in addition to, and were the direct result

of, one or more changes in vehicle design associated with the intro- duction of some noise reduction methodology. Since all of these changes had the potential for affecting the net noise reduction achieved with a given noise reduction methodology all were considered in the subsequent calculation of vehicle noise reduction. These design changes were also used to assess noise reduction cost.

ANALYTICAL METHODS The analytical methods developed and/or adapted for use in the present program fall into three general categories. These are: 1. Noise calculation.

2. Vehicle design and performance calculation.

3. Cost calculation.

With respect to noise, analytical models were either derived or adapted from exisping methods.

These enabled calculation of the rotor system, engine (turbine and reciprocating) and transmission noise components.

These component models were incorporated in a unified vehicle noise calculation method, which had the capability of generating 1/3 octave sound pressure level spectra as a function of time, at any observer location, for any steady state translational flight condition. This

method is illustrated in figure 3. These calculated 1/3 octave spectra

are automatically converted to effective perceived noise level (EPNL)

and instantaneous A-weighted sound pressure level (dBA) , overall sound

pressure level (OASPL) , perceived noise level (PNdB) and tone corrected

perceived noise 1 eve1 (PNLT) units.

A separate analytical method was developed to enable calculation

of the changes in he1 icopter design and performance characteristics which

result from the application of noise reduction technology to the various noise producing vehicle components. This method reflects the approach discussed in the preceding section and illustrated in figure 2.

The cost calculation method was developed from historical he1 icopter cost data, which relate the three elements of life cycle cost to the various vehicle design parameters (table 2). This model considers initial investment cost to be related to vehicle airframe weight and installed engine weight. Indirect operating cost is related to vehicle total empty weight. Direct operating cost is assumed to be a function of both empty weight and installed engine power. The cost calculation nethod permits determination of both absolute vehicle dollar costs and percentage changes in costs relative t o an established baseline helicopter design. Life cycle costs are calculated as a function of both annual usage rate and total useful l i f e .

The preceding analytical methods were used t o calculate baseline noise, performance and cost characteristics for several existing helicopter nodels, w i t h gross weights ranging from 8 kN t o 80 kN. Calculated EPNL's for these vehicles are shown i n figure 4. Investigation of the results of these calculations revealed that the main rotor and engines contributed most t o the vehicle EPNL and, consequently, subsequent noise reduction evaluations were directed a t these sources.

TURBINE E N G I N E NOISE R E D U C T I O N Helicopter turboshaft engine noise was found t o be dominated by exhaust Since these noise components may be effectively radiated components.

reduced through exhaust duct treatment, a study was performed t o evaluate the effect of such treatment on the total vehicle noise signature.

Three representative present generation helicopters were chosen for this study (table 1 ). A generalized exhaust silencer configuration, illustrated in figure 5, was established, and the normalized acoustic and aerodynamic performance characteristics of this duct treatment were derived (figures 6 and 7 ) .

Various levels of duct treatment were simulated, for each study vehicle, and estimates of the total vehicle noise reduction were made.

The average of fly-over and fly-by vehicle EPNL reductions achieved with engine silencing i s shown i n figure 8 as a function of silencer for each study vehicle. To provide a meaningful comparison, weight, silencer weight i s expressed as a percentage of vehicle gross weight.

On this basis, achieved noise reductions are roughly comparable, for comparable weight penal t i e s , for the three vehicles.

The additions t o vehicle gross weight and reductions in engine per- formance indicated in figures 7 and 8 do n o t reflect the total impact of engine silencing, and these changes alone do n o t represent an adequate basis f o r estimating changes in vehicle cost due t o silencer use. To provide this basis, the changes in vehicle design necessary t o accommo- date these direct penalties were determined.

Incorporation of an engine silencer increases vehicle gross weight by an amount equal t o the sil'encer weight. Vehicle airframe weight, however, must also increase, t o carry the added silencer weight. This change in air- frame weight further increases gross weight, requiring additional engine power and, consequently, increased engine weight. These three weight changes increase the fuel load required t o maintain constant range capa- Additions t o fuel load and engine weight further increase airframe bility.

weight, gross weight and power required. The ultimate gross weight, air- frame weight, engine power and weight and fuel load can be calculated through an iterative solution of the individual weight and power relation-

ships involved in the analytical method. Additional effects of engine

silencing are decreased available power and increased specific fuel consumption. These direct penalties result in the need for increased installed power and added fuel load, and these factors were also taken into account in redesigning the vehicle.

The net effects of incorporating exhaust duct treatment are illustrated in figure 5, which compares induced vehicle design changes to vehicle

EPNL reductions. Significant changes are shown in all

the design parameters considered, with the magnitude of change increasing sharply with noise reduction.

As might be expected, installed engine power is most greatly affected, with a 6% to 10% engine power growth shown for a 3-3.5 EPNdB reduction in EPNL.

The changes in vehicle design shown in figure 9 have been interpreted

in terms of changes in vehicle costs. Cost changes have been calculated

in terms of percentage changes in the basic cost elements of initial

investment cost, indirect operating cost and direct operating cost, as well as total life cycle cost. These calculations have been made using the parametric he1 icopter cost model discussed previously, with the direct silencer cost added to initial investment cost.

Change in investment and indirect operating costs due to engine exhaust

silencing are given in figure 10, with direct operating cost and life cycle cost changes shown in figure 11. The life cycle cost data shown refer to a useful life of 15 years, with an annual usage of 1500 hours.

The magnitudes of the cost increases shown are best illustrated b y con- sidering these changes in absolute terms. Considering an S-61 vehicle, for example, a 3 EPNdB noise reduction obtained through engine silencing would raise initial investment cost from $1.779 million per aircraft to $1.846 million per aircraft, an increase of $67,000. Indirect costs, on a yearly basis, would rise by over $5000 per year, from $147,00O/year to $I52,000/year. Direct operating cost, initially at $272 per hour would go up to $281 per hour, an increase o f over $8 per hour. Taken together, and assuming a useful life of 15 years with a usage rate of 1500 hours/year, total cost to own and operate this aircraft would increase by $293,000, from a baseline of $10.111 million to $10.404 million. This represents an annual cost increase o f nearly $20,000.

MAIN ROTOR NOISE REDUCTION

Evaluation of the significance of the various helicopter noise components indicated that the main rotor contributes substantially to the total vehicle noise signature.

Consequently, analyses were performed to deter- mine the extent of vehicle noise reduction obtainable through the applica- tion of rotor noise reduction methodology.

Methods considered in these analyses consisted of changes in gross rotor design parameters only, including increased rotor radius, blade chord and blade number and reduced rotor speed. The effects of these changes were evaluated in terms of the net vehicle noise reduction obtainable, considering potentially offsetting i n - induced changes in vehicle design, for constant performance. These duced design changes were also interpreted in terms of changes in vehicle cost, which were then compared t o anticipated vehicle noise reductions.

The base1 ine vehicles used for the preceding engine noise reduction study, the Hughes 500, Bell 205 and Sikorsky S-61, were also used in the per- formance of the main rotor noise reduction study.

Performance of the main rotor noise reduction evaluation was predicated on the same performance criteria used i n evaluating turbine engine noise reduction. In this regard the geometric rotor design parameters, including rotor radius, blade chord and number of blades, were treated as independent variables, and the effects of increasing each of these relative t o baseline vehicle values was evaluated separately. This could be done because changes in these parameters could be compensated for by iterating the vehicle design without violating the basic performance criteria. The remaining rotor design parameter, rotor t i p speed, was n o t evaluated independently since a reduction in rotor speed leads directly t o a reduced rotor stall margin, and this cannot be compensated for through the type of vehicle reconfiguration considered in the design analyses. Reduced speed can, however, be achieved w i t h o u t sacrificing s t a l l margin, i f a compensating increase i n blade o r disk area i s affected, since these area changes tend t o increase stall margin. Consequently, i n the present study, rotor speed variation has been considered only i n con- junction w i t h appropriate blade o r disk area changes.

The design implications of changes in rotor radius, blade chord and number of blades are illustrated i n figures 1 2 and 13. These curves shaw the changes i n gross weight and installed power which result from increasing rotor radius, blade chord and number of blades. Data are included for both constant and reduced rotor speed, w i t h r o t o r speed changes in accordance w i t h the constant s t a l l margin curve of figure 14.

While the data of figures 12 and 13 pertain t o the S-61 baseline vehicle only, similar results were obtained for the other study vehicles.

In figure 1 2 , gross weight i s seen t o increase with r o t o r radius, blade chord and blade number, with identical trends shown for chord and blade number. Rotor radius increases gross weight most quickly and the trend indicated i s nonlinear, with increasing slope. This i s due t o the fact t h a t r o t o r radius growth necessitates an increased fuselage size, in addition t o increased structural weight due t o .load require- ments. The maximum 25% increase in S-61 rotor radius results in a 6.6% increase in vehicle gross weight.

The trend of gross weight w i t h either chord or blade number i s linear and less steep than the trend with rotor radius. In this case, airframe weight only increases due t o the added rotor system weight and the added structural weight needed t o support the heavier rotor. Only a 4.1% gross weight increase i s indicated for a 25% blade area change, whether due t o chord or blade number increase.

Doubling the chord or number of blades causes a 16.4% increase i n gross weight.

The trends of installed power w i t h chord, blade number, and rotor radius a r e g i v e n i n figure 13.

Installed power is shown t o increase l i n e a r l y w i t h both chord and blade number, b u t t o decrease nonlinearly w i t h rotor radius, i n this case w i t h decreasing (absolute) slope. A 9.1% installed power reduction is indicated for the maximum 25% rotor radius increase. Installed power increases 6.3% f o r a 25% increase i n blade area, whether due to blade chord o r blade number. Doubling chord or number of blades increases installed power by 24%.

The data of figures 12 and 13 show only an insignificant difference i n the effects of rotor geometry changes evaluated alone and evaluated i n conjunc- tion w i t h rotor speed reduction. The magnitude of rotor speed reduction con- sidered i n these data is, however, relatively small, a s indicated i n figure 14. T h i s figure r e l a t e s rotor t i p speed t o change i n blade area, and the curves shown represent lines of constant s t a l l margin. A s shown, only a 3.1% reduction i n rotor t i p speed can be accommodated by a 25% radius increase. A 3 . 7 % reduction i n rotor speed is indicated for a similar 25% blade area increase, accomplished by increasing chord or blade number. A 12% t i p speed reduction can be obtained by doubling either blade chord or blade number.

The magnitude of vehicle gross weight increase associated w i t h the various rotor system changes strongly suggested t h a t noise reductions anticipated t o r e s u l t from the rotor system changes would tend t o be o f f s e t by increases i n noise due t o rotor thrust growth. Based on this indication, i t was decided to use a simplified rotor noise cal- culation method to determine the approximate magnitude of achievable net rotor noise reduction and, based on the r e s u l t s of these calcula- tions, decide whether t o proceed w i t h the more involved rotor and total vehicle noise calculations. T h i s approach was arrived a t based on the premise t h a t unless significant rotor noise reductions were shown through the simple analysis, no worthwhile reductions would be calculated f o r the total vehicle u s i n g the detailed analysis.

The simplified rotor noise calculation method chosen f o r use was obtained from reference 2. T h i s method r e l a t e s the magnitude of the h i g h frequency random component of rotor noise t o rotor speed squared, thrust squared and blade area.

T h i s method was used t o estimate the maximum possible rotor system noise reduction obtainable w i t h the various rotor system parameter changes considered.

The r e s u l t s of these calculations a r e summarized i n table 3 f o r the three study vehicles. Also given are the changes i n cost associated w i t h each of the rotor system var i a t i ons .

Comparison o f the cost and approximate rotor noise reduction data of ;able 3 reveals that the cost of reducing helicopter rotor noise levels is very high. Considering the 5-61 study vehicle, for example, increasing rotor size b y 25%, raises life cycle cost by over 5.7%, For a 1500 hour per year use rate, and the cost differential is greater

For lower annual use rates. In absolute terms, the 25% greater rotor

"adius increases life cycle cost by almost $.6 million dollars, or more than $38,00O/year. This rotor design change reduces rotor noise by

less than .5dB which, in all probability, would produce no measurable

Zhange in total vehicle noise.

The most beneficial rotor design change, doubling the number of blades and reducing rotor speed by approximately 12%, raises life cycle cost by slmost 30%. This translates into a $3.03 million dollar life cycle cost increase, or in yearly terms, over $200,000 added annual cost.

In terms of total vehicle noise, as discussed previously, the 2.8dB

rotor noise reduction associated with this design change, would probably

only result in a 1.6dB reduction in vehicle noise.

Because of the high cost to benefit ratios determined for the selected rotor noise reduction methods, it was concluded that these methods are not practical means for reducing helicopter noise, and that further analyses of these methods was not warranted. Consequently, these methods were not evaluated with the more involved noise calculation tech- niques originally intended for use. However, a small number of rotor

noise reduction design changes were subjected to further evaluation in

order to verify the appropriateness of the approximate noise calculation

method. In all cases studied, the involved noise calculation technique

indicated noise reductions similar in magnitude to those obtained with the approximate method.

CONCLUSIONS The results of the present study indicate that small , but meaningful ,

reductions in helicopter noise can be obtained b y treating the turbine

engine exhaust duct. Furthermore, these reductions do not result in excessive life cycle cost penalties. Currently available main rotor

noi se reduction methodol ogy, however, was shown to be inadequate and

excessively costly.

This result strongly suggests the need for additional he1 icopter rotor noise research, which should be directed at developing more efficient methods for reducing rotor noise.

As with any study of this nature, the results of the present effort should be interpreted only within the context of the study groundrules.

In this regard, two such groundrules are particularly important.

First, the vehicle design analysis used in this study considered only

current he1 icopter design and fabrication technology. Improvements in

these technologies, particularly those which result in better structural

efficiency, rotor performance and engine efficiency, could improve the effectiveness of current noise reduction methods, b y minimizing the extent of offsetting vehicle design changes.

The second study groundrule which must be considered i n evaluating the study results r e l a t e s t o the use of the constant performance concept.

While this approach provides a r e a l i s t i c framework f o r evaluating the cost of noise reduction, other approaches might also be equally valid. One such alternative approach would be the specification of m i n i m u m induced design change, w i t h variable performance capability.

In this context, the major impact of noise reduction would be interpreted i n terms of performance penalties, which would then be related t o cost d i f f e r e n t i a l s . T h i s approach is equally valid, although i t is somewhat more d i f f i c u l t t o apply and interpret than the constant performancc method.

REFERENCES 1. Bowes, M. A. : He1 icopter Noise Reduction Design Trade-off DOT/FAA Report FAA-AEQ-77-4, January 1977. Study.

2. King, R. J . and R. G. Schlegel: Prediction Methods and Trends f o r Helicopter Rotor Noise. CAC/AVLABS Symposium Procedings, June 1969.

TABLE 1. STUDY VEHICLE CHARACTERISTICS

Gross Instal 1 ed Study

F1 y-Away Cost

Weight Power Vehicle Ma nu f a c t u rer

( 8 )

(kN) (kW) S-67 Si kors ky 86.3 2237 1.8m 6- 2U5 Bel 1 42.2 1043 .6m H-500 Hughes 10.7 236 .12m TABLE 2. COST MODEL

0 INITIAL INVESTMENT COST - GI

DIREET OPERATING COST - DOC

o Airframe = f (Airframe WeignLj o Maintenance and spares = f {Empty Weight) o Engine = f (Installed Power) o Fuel and Oil = f (Instal 1 ed Power) o Initial Spares = f (Empty Weight) o Crew = f (Empty Weight) o Avionics = f (Empty Weight)

INDIRECT OPERATING COST - IOC o LIFE CYCLE COST - LCC

o Insurance = f (Empty Weight) LCC = GI + DOC(NA Lu) + IOC(LU) o NA = Annual Usage o Lu = Useful Life P

TABLE 3. VEHICLE COST AND NOISE CHANGES FOR

MAXIMUM NOISE REDUCTION CONFIGURATIONS -

S-61 A Investment Parameter A EPNL Varied Hr/Yr (EPNdB) +25% Radiu: +50% Chord 14.86 -1.08 +5 Blades 30.99 -1.67

1 I I

+25% Radiu!

14.02 13.98 .73 10.88 5.96 I - .68

-3% Q R b50% Chord 24.04 23.30 8.81 20.15 -7.2% Q R

6 Blades I

50.05 48.12 17.61 41.6 -2.82 -11.9% Q R

29*98 I

6 - 5 -

0 /’

4 , FLEET SIZE, 3 THOUSANDS

LINEAR TREND - - - - -

I 1 I I 1960 1965 1970 1975 1980 YEAR Figure 1.- Domestic c i v i l h e l i c o p t e r f l e e t growth.

U CRUISE RANGE

VEHICLE GROWTH

PLUS NOISE REOUC- - > VEHICLE

TION MODIFICATION GROSS WEIGHT CONFIGURATION

rlul-/ ENGINE S I Z E

~ HOVER CEILING ~ STALL MARGIN Figure 2,- Vehicle design methodology.

LOCATION, FLIGHT PERFORMANCE CONDITION & A/C RETURN Figure 3 . - Noise c a l c u l a t i o n methodology.

T c P ’ - 0 FLY-OVER, 91 M ALTITUDE

=@a” -=

A FLY-BY, 152 M TO ST ‘BD 0 FLY-BY, 152 M TO PORT

8o t

6 8 10 1 5 20 30 40 60 80 100 GROSS WEIGHT, KN F i g u r e 4 . - Calculated v e h i c l e n o i s e .

EXHAUST 7

-

F L O W

.I

V E LAYER (POROUS)

ABSORBENT M A T E R I A L

Figure 5.- Exhaust silencer configuration.

LID

25r

2o t

INSERTION Loss, dk?

50 100 200 500 1000 2000 5000 10000

FREQUENCY - HZ

Figure 6.- Exhaust silencer acoustic performance.

- LYCOMING T53-~-13B

--- GE T58-140-1

--

I I I I J 0 ' 5 6 1 2 3 4 L/D Figure 7.- Exhaust silencer aerodynamic performance.

3,5

I- /

3,O VEHICLE 2,O EPNL REDUCTION EPNdB 1.0 0 1 2 3 4 5

SILENCER WEIGHT - % OF GROSS WEIGHT

Figure 8.- Exhaust silencer weight.

CHANGE I N CHANGE I N GROSS WEIGHT, INSTALLED % POWER, % 0 1 2 3 4 0 1 2 3 4 VEHICLE EPNL REDUCTION - EPNdB

VEHICLE EPNL REDUCTION - EPNdB

F i g u r e 9.- Design changes due t o exhaust s i l e n c i n g .

'r

A DIRECT A L I F E OPERAT I NG CYCLE X COST, COST, %

VEHICLE EPNL REDUCTION - EPNdB VEHICLE EPNL REDUCTION - EPNdB

F i g u r e 10.- Direct o p e r a t i n g c o s t and l i f e c y c l e c o s t of exhaust s i l e n c i n g .

5 80 7 , 7, 6 - A INDIRECT A INITIAL 5 - OPERAT I NG INVESTMENT COST, % COST, % 4, 3 - 3 2 - 1- 1 0 1 2 3 4 0 1 2 3 4

VEHICLE EPNL REDUCTION - EPNdB

VEHICLE EPNL REDUCTION - EPNdB

F i g u r e 11.- I n i t i a l investment c o s t and i n d i r e c t o p e r a t i n g c o s t O f exhaust s i l e n c i n g .

A GROSS W E I GHT , % AT CONSTANT

-

ROTOR SPEED

----- WITH REDUCED

0 20 40 60 80 100

A BLADE AREA - %

F i g u r e 12.- Gross weight w i t h b l a d e area.

+30 t-

+20 BLADE NUMBER O R A INSTALLED +10 POWER, % AT CONSTANT

-

ROTOR SPEED WITH REDUCED

- ----

ROTOR SPEED -10 0 20 40 60 80 100 A BLADE AREA - % Figure 13. - Installed power with blade area.

I N A T I P SPEED, 4 : BLADE NUMBER O R -9 '- 0 20 40 60 80 100

A BLADE AREA - %

Figure 14.- Constant stall margin.

HELICOPTER CABIN NOISE - M E T H O D S O F SOURCE AND P A T H

IDENTIFICATION AND CHARACTERIZATION Bruce S. Murray and John F. Wilby Bolt Beranek and Newman Inc.

SUMMARY The e f f e c t i v e quieting of helicopter cabins requires t h a t t h e weight and :pace of t h e treatments be minimized. The application of these treatments ;herefore requires t h a t t h e paths by which t h e noise a r r i v e s at t h e cabin, :oupled with t h e r a d i a t i n g surfaces and t h e sources of o r i g i n be i d e n t i f i e d as Aearly as possible. The techniques described i n t h i s paper have been employed t s p a r t of comprehensive helicopter quieting programs which have achieved nota- ) l e reduction of cabin noise on s e v e r a l e x i s t i n g helicopter designs.

INTRODUCTION The continued expansion of and competition f o r t h e helicopter market, par- ;icularly i n t h e 8-20 seat capacity range, has l e d t o much,attention being siven t o t h e provision of quiet and comfortable passenger accommodation. The i e l i c o p t e r i n t h i s s i z e range when f i t t e d with t h e best "standard" i n t e r i o r a i l 1 t y p i c a l l y exhibit a cabin noise l e v e l of 93-95 dBA when f l y i n g at i t s 3esign speed, whereas a commercial transport w i l l have a t y p i c a l noise l e v e l ranging from 75 t o 82 d B A at i t s c r u i s e speed. Figure 1 summarizes t h e t y p i c a l i o i s e l e v e l s t h a t can be expected with various i n t e r i o r configurations and shows t h a t , even with t h e best available treatment, t h e noise l e v e l s a r e above those f o r a commercial transport. O f p a r t i c u l a r note are t h e tone l e v e l s which control t h e l e v e l s i n many portions of t h e spectrum and are of course more annoying than t h e same l e v e l s of random noise which tend t o predominate i n commercial a i r c r a f t . The s e l e c t i o n of noise control treatments t o reach t h e 1owest.helicopter l e v e l i n Figure 1 requires t h a t t h e noise sources and t h e i r paths t o t h e cabin be c a r e f u l l y measured s o t h a t minimal additional mass i s applied and t h a t t h e space occupancy is not affected. I n t h i s l a t t e r regard the passenger headroomds often c r i t i c a l as t h e cabin roof may w e l l be a major r a d i a t i n g area.

The mechanical design of a helicopter presents a p a r t i c u l a r challenge t o noise control engineers i n t h a t it i s usually a t i g h t l y coupled s t r u c t u r e en- cased with l i g h t r i g i d panels, which r a d i a t e sound e f f i c i e n t l y , plus r e l a t i v e - l y t h i n s i d e w a l l s and windows, which have a poor transmission loss. The presence of machinery located on t h e s t r u c t u r e plus t h e l o c a t i o n of t h e passen- gers within t h e main s t r u c t u r e f u r t h e r compound t h e d i f f i c u l t y .

The magnitude of t h e noise control t a s k can be appreciated by reference t o Figure 1 which shows t h a t even an engineering program which produces 20-30 dB reduction i n cabin noise l e v e l s above 500 H z from t h e bare i n t e r i o r i s insuffi- c i e n t t o meet commercial fixed wing j e t a i r c r a f t l e v e l s . This is a formidable challenge f o r any noise control project even without t h e constraint of minimum weight. This unusually high t a r g e t must therefore be m e t with t h e best possible understanding of t h e d e t a i l s of t h e noise mechanisms and transmission paths per- t a i n i n g t o t h e p a r t i c u l a r a i r c r a f t i n question.

Cabin noise reduction may be achieved by modification t o t h e source l e v e l s , t h e sound path, and t h e receiver environment. It i s not often possible t o re- duce t h e source l e v e l s i n helicopters as, f o r example, t h e impact of reducing terms of t h e gear noise by a l t e r i n g t h e mesh forces has wide implications i n r e l i a b i l i t y and load carrying capability. It i s thus a matter not l i g h t l y undertaken. Accordingly t h e reduction of cabin noise on e x i s t i n g helicopters i s concentrated on modification of t h e paths between source and receiver.

For t h e purposes of f u r t h e r discussion it is convenient t o consider sepa- r a t e l y t h e major groups of i n t e r n a l noise sources i n a helicopter. These may be c l a s s i f i e d as propulsion machinery comprising engine and transmission, and turbulent boundary l a y e r e f f e c t s .

The intrusion of main and t a i l r o t o r noise i n t o t h e cabin w e have not found t o be s i g n i f i c a n t . Although some t a i l r o t o r blade r a t e harmonics may appear i n a narrow band spectrum taken i n t h e cabin, t h e treatment applied t o t h e i n t e r i - o r t o correct other source l e v e l s a l s o leads t o reduction of these harmonics.

The broadband turbulent boundary layer blade noise may be considered as a s e r i e s of d i s t r i b u t e d dipoles which do not r a d i a t e e f f e c t i v e l y i n t h e direction from blade t o cabin. d I

PROPULSION MACHINERY NOISE - STRUCTUREBORNE

The t i g h t s t r u c t u r a l coupling and proximity of t h e main gearbox and leads t o t h e dominance of machinery noise i n t h e engine(s) t o t h e cabin usually cabin. Indeed, t h e main reduction gear mesh tone i s usually a predominant sen- should be noted t h a t work i s being performed on t h e s a t i o n i n t h e cabin. It f e a s i b i l i t y and techniques f o r mesh noise reduction a t t h e source; however t h i s i s beyond t h e scope of t h i s paper. Figure 2 i s a representation of t h e ways i n

which machinery noise a r r i v e s a t t h e cabin - other paths are possible but i n our

experience, t h e i d e n t i f i c a t i o n of t h e paths outlined i s s u f f i c i e n t f o r a l l prac- t i c a l purposes.

I n an o v e r a l l sense, we attempt t o measure d i r e c t l y t h e individual contri- bution of t h e machinery vibration and i t s case-radiated noise t o t h e sound f i e l d i n t h e cabin. W e then analyze individually t h e paths f o r t h e vibrations and noise. Consider as an example t h e methods used i n analyzing t h e structureborne noise from a main r o t o r gearbox. A very simplified diagram of t h e helicopter as it relates t o t h e structureborne noise from t h e gearbox i s shown i n Figure 3.

There are always more than two mounts. The f l e x i b i l i t y and damping of each i s taken as t h e t o t a l t h a t e x i s t s between t h e transmission attachment and t h e s t r u c t u r a l frame attachment. Also shown is t h e usual honeycomb overhead panel hich forms an i n t e g r a l p a r t of t h e s t r u c t u r e and, at t h e same t i m e , presents a arge noise r a d i a t i n g area i n t o t h e cabin. The w a l l s of t h e s t r u c t u r a l frame ill a l s o usually form p a r t of t h e cabin and may be s i g n i f i c a n t r a d i a t o r s .

The measurement program t y p i c a l l y includes t h e i n s t a l l a t i o n of accelerome- ers at t h e points located by A i n Figure 3 which are chosen t o t r a c k t h e vi- (passenger). Flight tests are then performed r a t i o n from source t o receiver ver t h e required operating range t o obtain t h e magnitude of these vibration evels which are recorded on seven or fourteen t r a c k tape machines. I n t h i s egard it i s important t o obtain as much coincident t i m e h i s t o r y of t h e accel- r a t i o n l e v e l s as possible t o allow t h e use of correlation analysis l a t e r on.

'he a c t u a l d i r e c t i o n and precise location of t h e accelerometers i s a matter f o r he experienced judgement of t h e experimenter as it i s simply not f e a s i b l e t o over a l l t h e possibly relevant vibration locations. The locations selected .re c l e a r l y influenced by h i s appreciation of t h e most l i k e l y s t r u c t u r a l paths, n d w i l l . p r e f e r a b l y involve t h e contribution of t h e s t r u c t u r a l designer.

I n t h e f l i g h t tests, several cabin microphones w i l l be employed t o monitor ,he t o t a l noise as well as t o located obvious acoustic 'hot s p o t s ' .

A s e r i e s of ground vibration t e s t s are conducted with a l l a i r c r a f t systems ;hut down i n which vibration i s applied t o t h e transmission case from a shaker iystem. The object here i s t o reproduce only t h a t phenomenon which one wishes ,o study without any other acoustic interference. Notwithstanding t h e t i g h t l y :oupled s t r u c t u r e of t h e helicopter it is most desirable t o attempt t o shake ;he gearbox i n such a way t h a t t h e correct d i s t r i b u t i o n of vibration l e v e l s i s jeen i n t h e mounts as occurs i n f l i g h t . If t h e noise control program i s being Ierformed on j u s t one helicopter then it may be necessary t o adjust t h e shaker -ocation and d i r e c t i o n t o obtain t h e desired mount vibration d i s t r i b u t i o n . If iowever, it i s possible t o perform t h e s t a t i c t e s t s on another sample then an t l t e r n a t i v e method i s t o provide l o c a l i z e d e x c i t a t i o n at each mount pad. This )pportunity w i l l a l s o allow t h e measurement of mount impedance looking i n t o t h e structure. This information, i n conjunction with t h e measured vibration i n ? l i g h t , allows determination of t h e vibratory power flow i n t o t h e s t r u c t u r e .

Given t h a t one can achieve t h e c o r r e c t d i s t r i b u t i o n of vibration on t h e r t a t i c t e s t a t t h e mount pads, w e then take readings of t h e induced noise i n ;he cabin at t h e selected locations. This allows us t o e s t a b l i s h t h e t r a n s f e r h n c t i o n : "Transmission Vibration-Cabin Noise". During t h i s test measurements a r e made at t h e selected s t r u c t u r a l frame and honeycomb panel locations t o arrive at t h e proportion of t o t a l vibration at these points t o t h a t induced by the transmission vibration alone.

The f i n a l s t e p i n t h e procedure i s t o apply t h e derived t r a n s f e r function to t h e a c t u a l measured i n f l i g h t vibration levels t o arrive at t h e contribution to cabin noise l e v e l s from t h e transmission vibration alone.

Figure 4 represents a t y p i c a l r e s u l t obtained on a l i g h t helicopter and I n t h e example shown i n Figure 4, t h e t r a n s f e r i l l u s t r a t e s t h e methodology.

function w a s based on average mount acceleration versus cabin noise and not on the a c t u a l transmission acceleration. The r e s u l t a n t computed cabin SPL due t o t h e structureborne transmission noise agrees creditably with t h e o v e r a l l result as w i l l be seen later. There i s however some discrepancy i n t h e high frequency range which i n d i c a t e s t h a t a more searching analysis could be worthwhile depending on t h e methods of noise control being contemplated.

There are a number of t r a p s f o r t h e unwary i n t h i s approach of which probably t h e most s i g n i f i c a n t i s t h e implied assumption of l i n e a r i t y . A s most shakers are unable t o drive t h e mechanical elements at frequencies and levels indenti- cal t o full-scale conditions, one must resort t o t e s t i n g at lower vibration and acoustic levels and applying l i n e a r scaling. This method i s usually satis, factory f o r m e t a l l i c a l l y mounted machinery components, but may be unsuitable f o r elastomeric mountings which exhibit a nonlinear load deflection curve, hence changes i n i s o l a t i o n performance, p a r t i c u l a r l y rear mount resonance may occur unless correct loading of t h e mounts i s achieved. Depending on t h e par- t i c u l a r helicopter design, it may be desirable t o measure t h e i s o l a t i o n provided by t h e mounts during f l i g h t and compare t h i s t o t h e ground t e s t runs with a view t o v e r i f i c a t i o n of t h e mount performance.

Although constant bandwidth analysis i s useful i n determining t h e major contributors t o t h e cabin noise spectrum, w e f i n d t h a t t h e l i n e density i s nor mally so high t h a t it i s easier t o analyze i n 1 / 3 octave bands, r a t h e r than tr; t o account f o r each individual d i s c r e t e frequency. The use of c o r r e l a t i o n i n view of t h e highly techniques has been used i n some of t h e analyses, but correlated vibration signatures which appear at t h e mounts i t s success has bee limited to, a general overview of t h e contributions of transmission vibration t cabin noise. W e f i n d t h a t t h e method presented here y i e l d s adequate accuracy and allows .us t o determine t h e most appropriate methods of noise control which may consist of mount modification, i s o l a t e d i n t e r i o r panels, or a combination of these.

PROPULSION MACHIWRY NOISE - AIRBORNE

The methods w e employ t o determine t h e contribution of case radiated nois t o t h e cabin noise l e v e l s are similar i n p r i n c i p l e t o those described above.

Reference again t o Figure 4 w i l l show our assumed acoustic paths f o r airborne noise which may be grouped as e i t h e r d i r e c t acoustic leaks through holes be- tween t h e machinery and passenger compartments or as t h e transmission loss of t h e w a l l s separating these compartments. The former case is usually e a s i l y spotted by v i s u a l inspection as well as by l o c a l i z i n g hot spots with a roving microphone i n t h e cabin. The l a t t e r case i s investigated by measuring t h e acoustic l e v e l s i n t h e machinery compartment at a number of locations dur- ing f l i g h t operations.

S t a t i c ground tests are then conducted t o evaluate t h e t r a n s f e r function between t h e noise i n t h e machinery compartment and t h e cabin SPL.

Single or multiple loudspeakers are placed i n t h e machinery compartment t o generate a acoustic f i e l d d i s t r i b u t i o n similar t o t h a t observed during f l i g h t and s i m u l - taneous recordings are made of machinery and cabin noise from which t h e trans- fer function i s derived.

586’ Figure 5 demonstrates a t y p i c a l r e s u l t of such a t e s t and t h i s may be com- pared w i t h the t o t a l noise i n the cabin shown i n a subsequent figure.

TURBULENT BOUNDARY LAYER NOISE One source of broadband noise I n t h e cabin i s t h e turbulent boundary l a y e r which is present over t h e e x t e r i o r of t h e helicopter. A t low forward veloci- ties boundary l a y e r noise w i l l be negligible compared t o other sources, but as helicopter speeds increase there i s a likelihood that such noise will become important.

The turbulent boundary l a y e r w i l l e x c i t e the cabin s t r u c t u r e and t h e win- However, since t h e s t r u c t u r e w i l l be covered by insulation material and dows.

i n t e r i o r t r i m , t h e windows w i l l be t h e important surfaces radiating boundary l a y e r noise i n t o t h e cabin. I n order t o estimate t h i s contribution t o cabin noise l e v e l s , it i s necessary flrst t o estlmate t h e vibration of the window and then the acoustic radiation.

For t y p i c a l helicopter speeds and window thicknesses, the acceleration power s p e c t r a l density Sa(f) of a window pane can be estimated using statisti- c a l energy analysis, under t h e assumption t h a t resonant response i s dominant.

The window vibration can be estimated using S p ( f ) S a ( f ) = f where Sp(.f) i s t h e boundary layer pressure excitation at frequency f , Uc is the pressure convection velocity, Q t h e panel l o s s f a c t o r , CL t h e longitudinal wave velocity i n the panel, M , t h e panel surface mass density and K t h e radius of gyration. Acceleration spectra have been estimated f o r t h e cabin windows assu- ming 11 = .01 and Uc = 0.8 Uo ( U i s t h e forward f l i g h t speed).

The above equation has been used t o estimate one-third octave band l e v e l s for windows, on a helicopter flying at i t s normal c r u i s e condition. The re- s u l t i n g spectrum is shown i n Figure 6 where it i s compared w i t h l e v e l s measured on three window panels, The agreement is quite good. The forward door windows show higher l e v e l s than predicted, which are probably due t o increased turbu- lence, as these windows a r e j u s t downstream of the most extreme bends of t h e fuselage contour and also may Be affected by increased turbulence due t o rotor downwash. The passenger door window vibration i s considerably greater than predicted at frequencies below 400 Hz as t h i s panel responds i n a resonant fashion t o t h e r o t o r pressure f i e l d .

The radiated sound pressure l e v e l s can be estimated using an equation of t h e form where Si(-f') i s t h e power s p e c t r a l density of t h e i n t e r i o r sound f i e l d , i s

t h e transmitting area, & i s t h e absorbing area with average absorption coeffi-

c i e n t a, po i s t h e air density, co i s t h e speed of sound, and CI i s t h e acoustic r a d i a t i o n e f f i c i e n c y of t h e transmitting s t r u c t u r e .

A s an upper l i m i t , CI can be assumed t o be unity.

The value of S a ( f ) can b e calculated, as indicated above, or obtained from measured vibration levels. An example of .the l a t t e r case i s shown i n Figure 7 , where t h e spectrum represents t h e acoustic power radiated by a l l windows of t h e helicopter cabin. I n t h i s p a r t i c u l a r case, t h e predicted sound l e v e l s r e s u l t - ing from turbulent boundary l a y e r e x c i t a t i o n were below thdse predicted f o r other sources. However t h e boundary layer contribution w i l l become more impor- t a n t as helicopter speeds increase and noise control techniques are applied t o other sources.

OVERALL RESULTS Summation of t h e individual contributors calculated from t h e noise source diagnosis r e s u l t s should y i e l d a value close t o t h a t measured i n f l i g h t . Fig- ure 8 shows t h e individual contributors determined f o r one model of helicopter i s a summary of data presented e a r l i e r with t h e addition of engine struc- and tureborne noise. Addition of these contributors y i e l d s t h e s o l i d l i n e shown i n Figure 9 and by contrast t h e d i r e c t l y measured l e v e l i n t h e cabin i s shown dashed.

The general agreement i s q u i t e good considering c e r t a i n simplifying as averaging of t h e gearbox mount vibration l e v e l s . It i s assumptions such c e r t a i n l y adequate f o r t h e design of i n t e r i o r noise treatments although f u r t h e r i f it i s desired t o change t h e machinery mounting refinements a r e necessary arrangements so as t o modify t h e vibratory power flow or i t s d i s t r i b u t i o n i n t o t h e s t r u c t u r e .

The estimated l e v e l above 2000 Hz i n Figure 9 i s about 2 dB below t h a t measured and may indicate t h a t an important contributor has been missed. How- ever, given t h a t t h e spectrum shapes are similar, it i s more l i k e l y t h a t t h e r e i s an e r r o r i n t h e t r a n s f e r function determination. A more searching analysis is expected t o resolve discrepancies of t h i s order. A similar s i t u a t i o n , al- though reversed, e x i s t s below 1200 Hz and may be caused by e r r o r s i n t h e experi- ment where t h e same path contributes t o two source mechanisms.

A s an example, tests on t h e machinery, t h e casing w i l l when performing structureborne noise radiate noise which can a l s o a r r i v e at t h e cabin v i a t h e acoustic path as w e l l as the s t r u c t u r a l path under study. Experimental care and anticipation of t h i s effect w i l l ensure that nasty surprises a r e avoided.

CONCLUDING R E M A R K S W e have shown t h a t by using r e l a t i v e l y simple concepts together with care- ful experimental work it i s possible t o generate r e l i a b l e data on which t o base the design of high performance noise control treatments.

A s an indication of t h e weight penalties associated with the noise control treatments derived from a thorough study of the source paths w e have been able t o achieve t h e best levels shown i n Figure 1 f o r an added weight of some about 40 kg (100 l b ) over t h a t f o r the unfurnished i n t e r i o r .

From t h i s can be sub- tracted the weight of a normally furnished i n t e r i o r which would produce the levels shown i n Figure 1. It has also been our experience t h a t it i s possible t o redefine some i n t e r i o r furnishing arrangements, which have noise control b u i l t in, so that no weigh% penalty i s incurred. This i s achieved by remov- ing and relocating or redesigning t h e noise control treatments t o obtain t h e best efficiency.

REFmENCE 1. Wilby, John F. and Smullin, Joseph I.: I n t e r i o r Acoustic Environment of STOL Vehicles and Helicopters.

Presented a t NOISE-CON 77, 17-19 October, 1977, Hampton, Virginia.

ONE-THIRD OCTAVE BAND SOUND PRESSURE LEVEL (dB re 2 ~ 1 0 ' ~ N / r n * 1 ONE-THIRD OCTAVE BAND CENTER FREOUENCY (Hi?)

Figure 1.- Typical helicopter i n t e r i o r noise.

CASING VIBRATION CASING RADIATED NOISE

k-J r

MOUNTING ACOUSTIC FEATURES RAD IAT ION

I

EXCITATION OF OF STRUCTURAL SURFACES Figure 2.- Paths of engine and gearbox noise t o t h e cabin.

MAIN ROTOR TRANSMISSION MOUNT tj STRUCTURAL FRAME CABIN OVERHEAD HONEYCOMB PANEL (RIGIDLY ATTACHED TO THE FRAME) I I @ DENOTES ACCELEROMETER LOCATIONS @ DENOTES MICROPHONE LOCATIONS @ DENOTES SHAKER LOCATIONS @ DENOTES IMPEDANCE MEASUREMENT LOCATIONS Figure 3 . - Simplified diagram for transmission noise structureborne into cabin.

d CABIN SPL per lg rms - 1 1 0 - ACC'N AT MOUNTS - -0 ONE-THIRD - ACC'N OCTAVE BAND COMPUTE0 CABIN SOUND PRESSURE (dBre 1g rmsl s P L LEVEL

I 1

( dB re 2 x IOm5 N/m -10 \0 \ -20 \ MOUNT ACCELERATION

i

7 5 &w2501 5 b O I IlObO' I2d001 kobo' 'ado; & ! O O

k ! '

ONE-THIRD OCTAVE BAND CENTER FREQUENCY ( Hz) Figure 4 . - Typical result of structureborne transmission noise test.

120 ~ ~ ~ ~ 1 ~ ~ 1 1 ~ ~ 1 ~ 1 MEASURED MACHINERY NOISE

l -

- COMPUTED NOISE ONE-THIRD loo - OCTAVE BAND CABIN S P L REDUCTION NOISE SOUND PRESSURE REDUCTION LEVEL MACH'Y TO -

(dB re 2 x I O + N/m2 1 go -

30 PASSENGER COMPARTMENT dB 80 - - 70 I 1 1 1 1 1 I I I I I 1 1 I I I I I 1 1 1 1 I 1 0 63 125 250 500 1 0 0 0 2000 4000 8000 16,000 Figure 5.- Computation of cabin noise due to airborne machinery noise.

' O PASSENGER DOOR WINDOW ONE-THIRD /PILOT DOOR WINDOW OCTAVE BAND SOUND PRESSURE L E V E L (dB re 2 ~ 1 0 ' ~ N / m 2 1 -20

-'F

Figure 6 . - Cabin window vibration levels - computed contribution of

turbulent boundary layer.

1 1 0 I I I I 1 1 1 1 1 1 I I I I I I I I I I I I 7

-

- -

-

ONE -THI RD OCTAVE BAND SOUND PRESSURE L E V E L -

-

I d B re 2 ~ 1 0 - ~ N / r n ~ 1

-

-

I I I I I I I I I I I I I I I I I '

63 125 250 500 1 0 0 0 2000 4000 8000 1 6 , 0 0 0 F i g u r e 7.- P r e d i c t e d c a b i n n o i s e level induced by t u r b u l e n t boundary e x c i t a t i o n of c a b i n windows.

TRANSMISSION ONE-THIRD 8o OCTAVE BAND SOUNDPRESSURE LEVEL (dB re 2 ~ 1 0 - ~ ' N / r n ~ 1 70 NOISE

6o t

50 r I ~ ' I I I I I ' I ' ' ' ' ' ' ' ' ' ' ' " ' I 63 1 2 5 250 500 1 0 0 0 2000 4000 8000 16,000 3 ONE-THIRD OCTAVE BAND CENTER FREOUENCY (Hz) Figure 8 . - Composite of n o i s e s p e c t r a i n c a b i n unfurnished i n t e r i o r .

MEASURED LEVEL ONE-THIRD OCTAVE BAND SOUND PRESSURE LEVEL SUM OF CONTRIBUTORS ON FIG 8

70 t

125 250 500 1000 2000 4000 8000 16,000 ONE-THIRD OCTAVE BAND CENTER FREQUENCY (Hz) Figure 9.- Comparison of measured and computed cabin noise.

A PRACTICAL A P P R O A C H TO HELICOPTER INTERNAL NOISE PREDICTION Larry S . Levine and Jon J. DeFelice Sikorsky Aircraft Division United Technologies Corporation SUMMARY A p r a c t i c a l and w e l l correlated procedure f o r predicting helicopter i n t e r n a l noise i s presented. The development of t h e method w a s supported by N A S A Contract NAS1-14446. It accounts f o r the propagation of noise along multiple paths on an octave-by-octave basis. The method is s u f f i - ciently general t o be applicable t o conventional helicopters as w e l l as other a i r c r a f t types, when the appropriate s t r u c t u r a l geometry, noise source strengths, and material acoustic properties are defined. A guide i s provided f o r t h e prediction of various helicopter noise sources over a wide range of horsepowers for use when measured data are not available.

The method is applied t o t h e prediction of t h e i n t e r i o r levels of

the NASA/Sikorsky Civil Helicopter Research Aircraft (CHRA) , both w i t h

and without soundproofing installed. Correlation with measured l e v e l s was very good. Speech Interference Level (SIL) w a s predicted within 1.5 dB a t a l l conditions. A sample problem i s a l s o shown i l l u s t r a t i n g t h e use of t h e procedure. T h i s example calculates t h e engine casing noise observed i n the passenger cabin of the CHRA.

INTRODUCTION Design efficiency has become an increasingly important characteristic i n the helicopter industry as manufacturers s t r i v e t o improve a i r c r a f t performance and operators s t r i v e t o hold operating costs down. As cabin i n t e r n a l noise requirements become more stringent f o r increased passenger comfort, soundproofing weight becomes an important issue. k designer needs t o know t h e acoustic environment i n a bare a i r c r a f t cabin t o be able t o define an effective soundproofing configuration. It would be a waste of valuable a i r c r a f t weight-empty t o carry many pounds of sound- proofing i n areas where they are not needed. But it would also be a serious e r r o r i f an i n t e r i o r failed t o meet the design requirement because not enough soundproofing was used. What i s needed i s a practical, i n t e r n a l noise prediction method that can provide the distribution and spectral content of the cabin noise signature. The goal is t o substitute an analytic evaluation of an i n t e r i o r configuration i n t o the noise model t o determine t h e most e f f i c i e n t placement of treatment weight.

.

There are many challenges i n attempting t o formulate and predict helicopter i n t e r n a l noise.

The noise observed i n a helicopter cabin is produced by several, widely varying source types. For example, rotor noise is generated by t h e aerodynamic forces of lift and drag.

Trans- mission noise i s generated by t h e mechanical forces transmitted i n gear tooth contact. Further complicating t h e formulation, a given source's noise may be propagated v i a d i f f e r e n t means: airborne, structure-borne, o r both.

The noise sources t h a t , i n general, make a s i g n i f i c a n t contribution t o cabin levels are t h e main transmission, engines, r o t o r s , and boundary l a y e r . Transmission noise i s composed of d i s c r e t e tones at gear clash frequencies and t h e i r harmonics and i s propagated as both airborne and structure-borne noise. It dominates cabin noise f o r several reasons.

airframe which F i r s t , t h e transmission i s coupled t o t h e l i g h t l y damped acts as an e f f i c i e n t radiator. Second, primary clash frequencies are range of peak s e n s i t i v i t y : 500-2000 high and often f a l l w i t h i n . t h e ear's Ha. F i n a l l y , d i s c r e t e tones are more annoying than broadband noise of t h e same l e v e l . Engine noise is broadband, peaking at 250-500 Hz, except f o r t h e compressor s h a f t r o t a t i o n a l and i n l e t vane passage tones. Its contributes t o propagation i s both airborne and structure-borne and it t h e cabin noise spectrum i n p r a c t i c a l l y every octave. Rotor noise har- monics are found at t h e lowest octaves. Their propagation i s airborne, soundproofing i s not very e f f e c t i v e at these frequencies , but because r o t o r harmonics are d i f f i c u l t t o keep out of the cabin. Boundary layer noise i s broadband and high i n frequency. For t h i s reason it is r e l a t i v e l y easy t o treat.

A n i n t e r n a l noise prediction method must, therefore, be many-faceted.

It must i n t e g r a t e ideas from such diverse areas as aerodynamic and structure-borne noise, radiation and propagation e f f e c t s , material transmission l o s s and absorption, and r o t o r , engine, and gear noise source strength prediction. Even though t h e method must represent a complex physical system, it must a l s o be simple enough t o apply within r e a l i s t i c c o n s t r a i n t s of t i m e and e f f o r t . Ideally, it should be suffic- i e n t l y f l e x i b l e t o be useful i n preliminary design as w e l l as i n t h e d e t a i l design stage with s p e c i f i c (possible measured) data.

The method discussed i n t h i s paper has those c a p a b i l i t i e s . While a more d e t a i l e d development and exercise of t h i s method i s presented i n Reference 1, t h i s paper w i l l serve as a guide t o i t s development and use.

The user can assemble t h e noise model i n whatever d e t a i l desired. When t h e s p e c i f i c t e s t data i s not available f o r individual noise sources, generalized prediction methods w i l l provide a reasonable guide. These methods predict r o t o r , engine, gear, and boundary l a y e r noise source strengths over a wide range of helicopter operating parameters. A means of estimating panel transmission loss based on t h e barrier's m a s s , stiff- ness, and physical dimensions i s a l s o included t o supplement t h e user's data base.

The approach of t h e method i s t o follow t h e propagation of each noise source t o t h e observer's location. ' h i s may occur v i a more than one path so each must be dealt with separately.

The properties of a l l intervening barriers and c a v i t i e s are considered and applied t o t h e sound transmitted through each. i s allows one t o account for t h e room acoustics and t o show t h e relative importance of t h e d i r e c t radiated i-ersus t h e reverberant f i e l d s . The frequency breakdown uses t h e nine ?referred octave bands from 31.5 Hz t o 8000 Hz (Reference 2 ) . Most % i r c r a f tnoise specifications a r e i n octave f o r we t h e three. SIL or Speech Interference Level 2000 Hz. These frequencies tend t o govern an ioise environment. Not only is human hearing mt the voice also appears primarily within t The method considers two forms of sound propagation, a structure-borne, Conventional room acoustics i s used t o represent air- borne propagation. For t h e frequencies and cabin dimensions involved, this approach i s reasonable. While cabin standing waves cannot be pre- dicted with t h i s technique, it i s assumed t h a t noise measurements are spacially averaged. Only at the lower frequencies, where wavelength and cabin dimensions a r e of t h e same order, do the assumptions begin t o break down. The propagation and radiation of structure-borne noise i s a d i f f i - c u l t process t o analyze i n an exact form. So many unknowns e x i s t i n describing t h e properties (mass, s t i f f n e s s , and damping) of complex structures t h a t a very simplified approach has t o be employed t o obtain a workable solution. Some gross assumptions are made concerning the block- age of structure-borne noise at turns, s t r u c t u r a l breaks, and at heavy frames. These assumptions serve t o define the extent and shape of the radiating areas. Skin panels and frames are assumed t o radiate uniformly within t h e areas bounded by s t r u c t u r a l breaks and equally well at a l l frequencies. While t h i s i s not e n t i r e l y accurate, present knowledge of the subject i s insufficient t o permit a more specific analysis. It w i l l remain for' further work i n the areas of s t r u c t u r a l impedance and radiation efficiency t o formulate an exact solution of t h e program f o r a r b i t r a r y structures.

SYMBOLS h rotor t o t a l blade area, f t *b directivity/distance factor C , Co , C 1 ¶ . .

Cavg average directivity/distance factor C speed of sound (sea l e v e l s t d ) , 340 m/sec (1116 f t / s e c ) ; blade chord, ft rotor broadband noise peak frequency, Hertz f S boundary layer peak frequency f O h progected blade th:ckness, f t acoustic intensity w a t t s / m 2 '3 'obs reference intensity 10-13 w a t t s / m 2 IO L radiating surface dimension, ft harmonic number m N number of c a v i t i e s 9 dynamic pressure room constant; distance t o observer, ft R , R o , R 1 , . . .

r near-field distance f ar-f i e l d distance r2 t o t a l area ST source area sS b a r r i e r area S, So, S 1 , . . .

T t h r u s t , pounds

t blade thickness, f t urn f r e e stream velocity r o t o r t i p speed, f t / s e c vt

w, w i acoustic power, w a t t s

t o t a l acoustic power, w a t t s wT X distance from leading edge a absorption coefficient

-

a average room absorption coefficient

n non-dimensional fluctuating pressure l e v e l s

6 boundary layer moment& thickness 6" boundary l a y e r displacement thickness E % leakage 0 blade l i n e a r equivalent t w i s t ; degrees density of air, i n t e n s i t y r a t i o P , P i

c SUm

radiating surface area r a t i o transmission coefficient effective transmission coefficient radiating surface power r a t i o CONVERSION FACTORS To convert from To Multiply by ft m 0.3048 f t 2 m 0.0929 horsepower k W 0 7457 lbm 0.4536 l b f 4.448 l b f / f t 2 47.88 slugs / f t 3 515 4 FORMULATION O F THE MODEL Approach There are two types of acoustic processes t h a t must be expresged mathematically t o predict t h e i n t e r n a l noise environment.

They include t h e airborne propagation of sound through several cavities and structure- borne propagation over large radiating surfaces, shown i n Figure 1.

These two propagation modes are t h e paths by which sound e n e r a reaches t h e cabin.

It i s convenient t o break t h e cabin down i n t o elementary radiating surfaces. The d i r e c t i v i t y characteristics of these simple radiators a given observer position. The determine t h e d i r e c t sound f i e l d f o r standard equations of room acoustics are used t o establish t h e reverberant f i e l d . The two f i e l d s a r e combined t o form t h e complete i n t e r n a l noise environment .

Propagation and Radiation Formulas Consider an arbitrary surface of some area S t o be radiating noise The t o t a l energy emitted by t h e p l a t e i s W (watts). (Figure 2 ) . The sound power l e v e l (PWL) of t h i s radiating p l a t e r e l a t i v e t o a standard source radiating 10-12 w a t t s (wO) i s

PWL = 10 log w/(w,)

(1) The i n t e n s i t y I (watts/m2) on the surface of t h i s p l a t e (outgoing energy per unit area) i s related t o the p l a t e ' s area, S. Since

W = I . S or I = w/s

then the Intensity Level ( I L ) i s W 1

-

I L = 10 Log - = 10 Log w/wo - 10 Log s

S wO o r

I L = PWL - 10 Log s

( 2 ) The sound pressure l e v e l (SPL) actually observed a t an arbitrary point i n a room i s given by the sum of both the reverberant and direct radiated components. Different types of acoustic sources radiate differ- ently depending on t h e source type and i t s extent i n space. The point source radiates i t s power symmetrically about any sphere with t h a t source at i t s center. Because of spherical spreading, the SPL observed drops off 6 dB per doubling of: distance from the source. However, for f i n i t e or surface sources t h i s decay w i t h distance i s not the same. The l i n e works of E. 2. Rathe (Reference 3) and R. B. Tatge (Reference 4 ) deal e x p l i c i t l y with t h i s subject. Close t o a l i n e source, SPL drops off a t a slow r a t e , -3dB/doubling. A s t h e distance t o the observer becomes much greater than t h e source length t h e roll-off increases t o the rate of a point source, -6dB/doubling. Tatge developed several curves which plot the roll-off i n observed sound pressure l e v e l w i t h distance for one c i r c u l a r and several rectangular sources of different aspect r a t i o s .

They are non-dimensionalized with respect t o distance i n terms of the source dimensions and assume t h a t the observer i s e i t h e r over the center of t h e source o r i n its plane.

Recalling equation (2), it would be convenient t o determine the observed SPL j u s t by adding a correction factor. This factor would account f o r source shape and the distance and orientation of the observer t o it. Rewriting ( 2 ) :

SPL = PWL - 10 log s + Correction

( 3 ) O n t h e radiating surface, t h e term would account f o r pressure doubling.

Far away fromthe source it would show a 6 dB reduction w i t h each doubling D f distance. This i s exactly what Tatge does i n h i s work. The curves are derived f o r radiatYon normal t o a surface's center or in-plane from the mid-point of one edge (See Figure 3). Tatge's curves a r e i n terms of dB. These curves were replotted i n terms of t h e i r antilogs:

Corr . / l o

c = 10

Equation (3) i s then redefined as

(3A) SPL = P W L - 10 Log S + 10 Log C

Figures 3-7 p l o t C f o r rectangles of 1:1, 2:1, 4:1 and 8:1 aspect r a t i o s and f o r a c i r c u l a r disk, respectively. Distance i s non-dimensionalized i n terms of t h e source's long dimension. One may interpolate between curves t o f i n d values of C f o r intermediate aspect r a t i o s . When t h e observer point i s not d i r e c t l y over t h e center of t h e radiating surface, t h e distance from t h e point t o the source center may be used as an approx- imation. Equation 3A formulates the d i r e c t f i e l d due t o elementary radiating surfaces and may be used with Figures 3-7 f o r a given source type and observer location.

When describing how sound i s attenuated i n passing through a b a r r i e r , we speak of t h e b a r r i e r ' s transmission l o s s ( T L ) . Transmission loss r e f e r s t o t h a t portion of t h e incident pressure t h a t i s dissipated or reflected. The w a l l ' s sound transmission l o s s (STL) can be expressed i n terms of a transmission coefficient T:

STL = 1 0 Log 7 ( 4 )

The transmission coefficient T i s r e a l l y t h e fraction of the incident pressure t h a t is transmitted through the b a r r i e r : ( S T L / ~ O ) ( 5 ) T = 10- AS T goes t o 1, a l l of t h e pressure i s transmitted and STL goes t o 0.

The effective transmission loss (or t h e effective transmissibility T e f f ) of a b a r r i e r made up of more than one section l i e s somewhere between t h e TL of each i n d i v i d u d section. For a b a r r i e r of t o t a l area S , composed of N sections

whose T'S are given by,Tl, ' c z . . . T ~ , t h e net ETL i s given by

ETL = 10 Log S = 10 Log U T e f f

s T +s T +... s T

1 1 2 2 N N This r e l a t i o n can be used t o model the presence of an acoustic leak i n t h e panel. Consider a hole (where T = 1) of some fraction E of a panel's t o t a l area. Asstune a uniform transmission loss over t h e r e s t of t h e area Then t h e e f f e c t i v e transmission loss of the e n t i r e panel i s given (T~).

by T = (&€)TI + E e f f Take E = .01 (1% leakage) as an example. The quantity 1--E i s .99 s 1.0; then = T + .01 T e f f 1 a high transmission loss panel, T i s very s m a l l . The Note t h a t f o r l i m i t i n g value of T i s .01, meaning t h a t t h e maximum E;TL possible is eff 20 dB. Should t h e area of t h e acoustic leak be anything l a r g e r than .I% of t h e t o t a l area, t h e net ETL of t h e panel W i l l be dramatically reduced.

Figure 8 i l l u s t r a t e s t h e importance of leakage t o panel transmission l o s s performance. It p l o t s the actual TL obtained from a panel designed f o r a s p e c i f i c value but suffering from leakage e f f e c t s . From t h e figure, it i s seen t h a t a panel designed f o r 40 dB attenuation at a s p e c i f i c octave A n e f f i c i e n t i n t e r i o r w i l l y i e l d only 20 dB i n t h e presence of 1% leakage.

i s usually designed f o r 0.1% leakage; The sound f i e l d inside an a i r c r a f t cabin or any other cavi$y i s composed of both d i r e c t radiated and reverberant components. Direct radiation, as t h e name implies, involves only one path between t h e source and receiver. Reverberation, on the other hand, involves multiple paths because of r e f l e c t i o n s from the cavity w a l l s . The a b i l i t y of a w a l l t o absorb sound i s t h e same quality t h a t distinguishes an acoustically "live" (reverberant) room from an acoustically "dead" one. The absorption coefficient a represents t h a t f r a c t i o n of t h e incident pressure t h a t i s dissipated when t h e w a l l would otherwise be a perfect r e f l e c t o r . When very l i t t l e energy is dissipated at each r e f l e c t i o n , t h e sound pressure i n t h e cavity due t o a source can build up t o many times the corresponding free-field'value. In a highly r e f l e c t i v e room, these l e v e l s can grow by 10 dB or more. In t h i s case, it i s highly advantageous t o add absorption.

To reduce t h e i n t e r n a l l e v e l s by 10 dB through blockage (transmission l o s s ) might require adding several hundred pounds of i n t e r i o r weight.

w i l l reduce t h e The addition of absorptive material i n t o t h e cabin reverberant sound f i e l d by dissipation at each reflection. To achieve t h e same reduction using absorption, only a small weight penalty i s required since acoustically absorbant materials such as fiberglass b a t t i n g and polyurethane foams are extremely l i g h t .

a room depends upon the room constant R.

The acoustic livenejjs of It i s i n u n i t s of length and represents the area of perfectly absorbent

material present i n t h e room. Embelton (Reference 5 - Chapter 9 ) derives

R as follows: The absorption coefficient of a w a l l at one octave i s given by a. I f t h e t o t a l surface area of a cavity i s given by i=l ien t h e cavity's average absorption coefficient is defined as N nd, subsequently, t h e room constant becomes The Equations of Multibarrier Acoustics W e now have a l l t h e t o o l s required t o derive the equations governing Figure he propagation of sound through an a r b i t r a r y number of cavities.

The formula t h a t shows a source cavity (0) and an observer cavity (1).

xpresses t h e impinging sound pressure l e v e l SPL on the o u t l e t w a l l i n erms of t h e source power l e v e l P a s , source area S room constant Ro, S ' s d d i r e c t i v i t y / distance term Co i s given by

SPLo = PwLs - 10 Log Ss + 10 Log (Co + 4 / R O )

( 9 ) :quation ( 9 ) repeats quantitatively that the SPL observed at an arbitrary koint i n a cavity is t h e sum of both t h e d i r e c t l y radiated and reverberant :omponents.

The sound pressure l e v e l transferred through barrier 0 i n t o cavity 1 ;PLol is given by t h e impinging sound pressure l e v e l SPL minus t h e Iarrier's sound transmission loss STL or 0'

(10) SPLol = SPLo - STLO

The power l e v e l Recall STLO i s given by 10 Log l / ~ ~ i n equation ( 4 ) .

entering cavity 1 (PWL ) i s the i n t e n s i t y times the barrier area, o r (11) PWLOl = SPLOl + 10 Log so

(1lA) P W L o l = P W L s - 10 Log Ss + 10 Log (Co + 4/RJ- 10 Log l / T o

+ 10 Log so i s now t h e sound power source i n cavity 1. Again applying Equation

pwLol

( 9 ) , t h e SPL impinging on the observer i n cavity 1 (SPL ) i s given by

(12) SPLl = P W L o l - 1 0 Log So + 10 Log ( C

+ 4/R1) By s u b s t i t u t i n g equation(1l.A) SPL may be expressed i n terms of the o r i g i n a l source power l e v e l 1

(13) SPLl = P W L s - 1 0 Log Ss + 10 Log (Co + 4 / R O )

+ 10 Log T 0 + 10 Log (C1 + 4/R1) A s more c a v i t i e s are added, t h e terms can be grouped together. For N c a v i t i e s , t h e sound pressure l e v e l i n cavity N i s given by

( 1 4 ) SPLN = P W L s - 1 0 Log Ss + 1 0 Log [ ( C o + 4/RO)(C1 + 4/R1)

. (cN + 4 / ~ ~ ) ]

It i s i n t e r e s t i n g t o consider some l i m i t i n g cases. A s STL+O, ~+l.

If t h e r e were no b a r r i e r s , t h e product of a l l t h e T ' S would be 1. Since Log(1) = 0 , t h e r e would be zero barrier attenuation. A s t h e w a l l s become e i t h e r more transparent or absorptive, R grows l a r g e r and t h e component due t o reverberant build-up disappears. Comparing t h e r e l a t i v e magnitude of t h e terms C and 4 / R w i l l show whethen d i r e c t radiation or reverberation dominates t h e sound f i e l d . The proper treatment f o r noise reduction becomes apparent. Adding absorption w i l l reduce t h e reverberant component only, while adding transmission loss w i l l reduce t h e d i r e c t radiation.

This points out t h e need f o r a balanced treatment, i n t h a t over-treating one component does nothing for t h e other. Should t h e source impinging on S be i n t h e open a i r , such as with r o t o r noise, SPL reduces t o 0 0

(15) SPLo = P W L s - 10 Log Ss + 10 Log Co

Hence, when t h e r e i s no source cavity and t h e sound pressure l e v e l impinging on w a l l S i s already known, t h e terms i n equation ( 1 4 ) t h a t are on t h e r i g h t hang side of equation ( 1 5 ) can be replaced simply with SPLo.

S t r u c t u r a l Radiation Radiation by s t r u c t u r a l members i s d i f f i c u l t t o predict without Point (attachments measured values of s t r u c t u r a l impedance o r mobility.

l i n e (frames, s t r i n g e r s ) , and surface (skin) sources Contribute t o t h e Rather than attempt t o i d e n t i 0 t o t a l p i c t u r e of s t r u c t u r a l radiation.

Whole sur- each component, a more s t a t i s t i c a l approach should be used.

faces are assumed t o radiate instead of d i s c r e t e p a r t s . The power fed ito t h e cabin by t h e source i s d i s t r i b u t e d among t h e individual radiating vfaces The r e s u l t i n g i n t e n s i t y d i s t r i b u t i o n on each surface (power/unit *ea) i s assumed t o be uniform.

When more than one surface within a cavity radiates, t h e expression )r t h e SPL at the observer can become very complex. Some simplification in be accomplished by expressing t h e radiation i n t e n s i t y of t h e surfaces I terms of t h a t of one surface. The t o t a l i n t e n s i t y observed at one If x i t i o n i s t h e sum of t h e i n t e n s i t i e s propagated from each surface.

nere a r e N surfaces of area S ... S radiating power W 13 W2’ - 0 .

1’ s23 N then t h e t o t a l i n t e n s i t y observed i s !l here C1, represent t h e d i r e c t i v i t y f a c t o r s f o r t h e radiating

C2, ... C

N

urfaces. Define a radiation i n t e n s i t y r a t i o pi = - . wl/sl Equation (16)

an be rewritten w i /Si

I hen t h e l e v e l i s given by

(18) SPLobs = PWLl - 10 Log S 1 + 1 0 Log [C1 + C2/p2 +. . .+ CN/pN]

The r a t i o p i i s r e a l l y the r a t i o of t h e i n t e n s i t y of w a l l 1 t o the .ntensity of w a l l i. If w e assume t h a t t h e i n t e n s i t y l e v e l drops 6 dB at 1 corner or i n t e r s e c t i o n with a heavy frame, as i l l u s t r a t e d i n Reference i , Chapter 11, we are saying t h a t the i n t e n s i t y i s cut i n half. I n other rords, t h e radiation r a t i o l / p i equals 1/2. This greatly simplifies t h e ?quation f o r t h e t o t a l i n t e n s i t y observed. A generalized expression f o r structural radiation t h a t includes panel transmission loss and room kcoustics follows :

(19) SPLobs = PwLl - 10 Log s1

T ~ ( C ~ + 4 / R ) + 4 / R ) + N

+ ... +

p2 PN SOURCE STRENGTH PREDICTION OF MAJOR C O M P O N E N T S A guide t o t h e prediction of noise generated by t h e major helicopter noise sources i s presented. Trends are shown over a wide range of oper- a t i n g parameters. Specific examples are given based on t h e prediction of noise i n t h e NASA/Sikorsky C i v i l Helicopter Research Aircraft (CHRA), a modified CH-53 A/D. The CHRA i s a s i x bladed, s i n g l e main r o t o r helicopter i n t h e 15 900 kg (35,000 I b ) weight c l a s s , powered by two G.E. T-64 engines.

Transmission Noise The prediction of gear noise remains t h e most challenging of a l l t h e aspects of helicopter i n t e r n a l noise. Recent work by Grande et a1 (Reference 6 ) has demonstrated consistent trends with such variables as horsepower, s p e c i f i c tooth load, pitch-line velocity, manufacturing tolerances, and gear type. Other s t u d i e s (Reference 7 ) attempt t o orre- NASTRAMd and l a t e with t h e noise f i e l d surrounding t h e gearbox casing.

other f i n i t e element models are j u s t now being used at acoustic frequencies t o a n a l y t i c a l l y predict frequency response and acoustic radiation (Reference It would be a d i f f i c u l t enough problem i f t h e noise r a d i a t i o n 8 ) .

stopped at t h e gearbox casing f e e t : a problem i n d i r e c t radiation.

However, t h e casing i s mounted t o an a r b i t r a r y airframe which i s driven by t h e casing's foot motions: a problem i n structure-borne noise.

Differing airframe geometries, casing designs, and gearbox mounting techniques add a new s e t of variables t o t h e noise problem. Figure 10 shows t h e t r e n d i n bare cabin gear clash tones with a v a r i e t y of gearbox mounting types. The curves show t h e influence of t h e propagation path from t h e primary gear clash source t o t h e r a d i a t i n g airframe. S c a t t e r i n t h e data can be a t t r i b u t e d t o l o c a l (but s i g n i f i c a n t ) frame resonances.

How, then, can t h e designer determine t h e cabin noise l e v e l s generated by t h e main transmission i n an a r b i t r a r y helicopter? Some assumptions must be made about t h e propagation of structure-borne noise along t h e airframe. When driven by a gearbox foot, a heavy s t r u c t u r e such as a forging w i l l tend t o radiate along i t s e n t i r e length. Intersections between heavy s t r u c t u r e and l i g h t s t r u c t u r e (skins and s t r i n g e r s ) tend t o r e j e c t structure-borne noise because of t h e impedance mis-match.

Beraneck considers t h e problem of t h e attenuation of structure-borne noise at corners and i n t e r s e c t i o n s (Reference 5, Chapter 11). H e assumes t h a t , with t h e same s t r u c t u r a l properties on e i t h e r side of a 90 t u r n , a 3 dB attenuation i n power w i l l be observed.

Similarly, crossing over a heavy frame i s assumed t o give a 3 dB reduction. O n t h i s basis, t h e analyst can examine t h e a i r c r a f t s t r u c t u r e and e s t a b l i s h t h e r a d i a t i n g areas.

A s an example, figure 11 i l l u s t r a t e s t h e r a d i a t i n g areas used t o model t h e CHRA structure-borne noise induced by t h e main transmission.

a review of t h e CHRA N A S T R A N work performed by M.

The model r e s u l t e d from W. Dean (Reference 9 ) . The primary area of radiation i s t h e c e i l i n g from t a t i o n 282 t o s$ation 442. The frame at STA 442 i s a major forging and erves t o carry t h e landing gear loads. The frame at STA 282 i s a l s o a a j o r s t r u c t u r e and c a r r i e s t h e engine support loads. Beyond these t a t i o n s , a 3 d B reduction i n i n t e n s i t y is assumed. These areas are ounded forward by t h e cockpit bulkhead (STA 162) and aft by t h e t a i l c o n e ntersection (STA 522). Radiation from t h e side xtend from t h e c e i l i n g (waterline 191) down t o on t waterline 132. The presence of t h e f u t i n g t r u c t u r e within t h e sponson i s assumed t u f f i c i e n t l y t o eliminate i t s r a d i a t i o n of structure-borne noise.

In t h e absence of a detailed prediction scheme, some trends can be eveloped based on measurements taken i n untreated CH-53 a i r c r a f t .

igure 12 p l o t s t h e observed r e l a t i o n between t h e t o t a l radiated acoustic ower within t h e cabin at each gear c l a s h fundamental and t h e consumed P. This p l o t was derived from t y p i c a l CH-53 data measured at Sikorsky.

'here a r e separate l i n e s f o r t h e t h r e e major gear types i n t h e main ransmission. For t h e CH-53D, t h e phased 2nd stage planetary gears ;enerate nearly 15 dB less acoustic power than t h e unphased first stage d a n e t a r i e s . The main bevel i s another 5 d B down from t h e second stage )lanetary l e v e l .

A s discussed i n t h e section on s t r u c t u r a l r a d i a t i o n , t h e observed Iound pressure l e v e l s can be determined from t h e r a d i a t i o n i n t e n s i t y of

,he dominant surface (PWL - 10 Log S ) , t h e room constant R and

1 1, l i r e c t i v i t y / d i s t a n c e f a c t o r s C (equa-kion 22). The value of t h e sound i )ewer l e v e l radiated by t h e dominant overhead region can be expressed i n ,erms of t h e t o t a l power (From Figure 11) and t h e area r a t i o s of t h e :econdary regions. If t h e t o t a l power radiated by t h e s t r u c t u r e i s given )Y

(20) WT = w + w +...+ WN

1 2 md i f t h e i n t e n s i t y r a t i o between t h e dominant (1) and secondary surfaces :i) i s assumed t o be ir 'i 'i si 1 - = - - = - w1 s1 ;hen t h e t o t a l power radiated i n t o t h e cabin by N surfaces can be expressed (21) WT = znd ' N

1 2 2 ) PWLT = PwLl + 10 Log (1 + "2 + . * . + - >

2 2 Finally, t h e sound power l e v e l radiated by t h e dominant area (Pa1) i n terms of t h e t o t a l power from Figure 12 (PWLT) is This equation expresses t h e acou d by t h e dominant c e i l i n g area i n terms of t h e t o t a l ac i a t e d with t h e gearbox. The gear noise octave spect ed by calculating t h e harmonic frequencies of each gear and applying t h e generalized harmonic spectrum of Figure 13. This spectrum w a s developed from t h e findings of Grande, e t al. i n Reference 6 and agrees w e l l with observed CH-53D data.

Once t h e harmonic frequencies and l e v e l s are determined, t h e t o t a l octave l e v e l s can be summed according t o t h e bands i n t o which t h e harmonics fall.

For example, t h e C H - 5 3 consumes 3.7 Mw (5000 Kp) (approximately) i n both 150 knot c r u i s e and hover at sea level, operating near 15 900 kg (35 000 l b ) gross weight. Figure 12 i n d i c a t e s t h a t t h e acoustic power radiated at t h e fundamental gear clash frequencies a r e 134, 118.5, and 115 dB f o r t h e first stage p l a n e t a r i e s , second stage p l a n e t a r i e s , and main bevel gear, respectively. Gear noise harmonics occur at multiples of t h e first stage gear clash frequency of 527 Hz, and bevel clash frequency of 2710 Hz.

The following table of power l e v e l s summarizes t h e construction of ~ t h e octave spectrum:

TOTAL ACOUSTIC P O W E R -

PWLT

Octave Level - dB

Gear 500 Hz 1000 H z 2000 H z 4000 H z 8000 H z 2nd P1. 118.5 118.5 103.5 103.5 93.5

Bevel -- -- 115 5 115 5 100.5

100.5 S W n 118.5 134.5 134.5 123.3 113.0 The above table refers t o t h e t o t a l structure-borne acoustic power Only a small radiated i n each octave because of t h e main transmission.

percentage of t h i s t o t a l power i s radiated by t h e transmission casing t s e l f due t o i t s r e l a t i v e l y s m a l l exposed area and t h e f a c t t h a t it i s mered by a rubber d r i p pan. Most of it i s radiated by t h e cabin c e i l i n g , idewalls, and frame m e m b e r s . S u b s t i t u t e P W L T i n t o equation ( 2 3 ) t o etermine P W L S u b s t i t u t e P W L i n t o equation (19) t o determine cabin 1 ' 1 o i s e l e v e l s .

Engine Noise A review of turboshaft engine noise radiated by i n l e t , casing, and Figure 1 4 shows t h e xhaust reveals consistent trends with horsepower.

ound power l e v e l r a d i a t e d by s e v e r a l types of turboshaft engines over a ide range of horsepower. This chart, derived from Reference 6 and d d i t i o n a l Sikorsky data, follows t h e r e l a t i o n P W L = 10 Log(KP) -t 108 dB.

o s t of t h e noise i s radiated by t h e casing and exhaust except when t h e ngine blade passage frequency falls within t h e 8000 H z octave. Then n l e t noise w i l l dominate t h e uppermost octave. With t h e current CH- 3A/D engine i n s t a l l a t i o n , i n l e t noise i s not heard i n t h e passenger w a s not considered a major source i n t h e abin. Therefore? i n l e t noise HRA cabin. By overlaying t h e spectra of t h e engines onto one p l o t A conserva- ormalized by HP, a close correspondence i n shape w a s noted.

i v e , generalized s p e c t r a l shape w a s averaged through t h e upper l i m i t of he data f o r casing, exhaust, and i n l e t power levels. These a r e p l o t t e d n Figures 15, 16, and 17. B y adding 10 Log (HP) t o these non-dimension- .lized spectra, a close estimate of t h e engine octave power l e v e l s can be Ibtained. Note t h a t t h e T-64 i s quieter i n t h e lower octaves than t h e ;eneralized spectrum. To c a l c u l a t e t h e engine noise radiated i n t o t h e . i r c r a f t cabin, t h e r a d i a t i n g areas must be determined. The f i r e w a l l 'ransmits t h e casing noise while t h e a i r c r a f t skin aft of t h e engine lacelle transmits t h e exhaust noise. The calculation procedure f o r !ngine casing noise i s demonstrated i n t h e sample problem later i n the )aper. Note t h a t t h e Exhaust noise contours a r e p l o t t e d i n Figure 18.

.evels f a l l o f f sharply i n t h e near-field: 1 5 dB down at a distance of ,wo exhaust diameters.

Rotor Noise Few methods e x i s t f o r t h e prediction of near-field r o t o r noise.

h t h e r l a n d and Brown (Reference 10) provide an excellent technique f o r letermining blade passage harmonic l e v e l s within a radius o f one r o t o r tiameter. It has been applied successfully t o t h e prediction of both nain and t a i l r o t o r noise. The method i s adequately explained i n Reference LO and w i l l not be reproduced here. What i s required i s a means of ?redicting near-field r o t o r broadband noise i n terms of octave bands. Many zomputer programs e x i s t f o r t h e prediction of r o t o r noise i n t h e far- Field. The method employed at Sikorsky Aircraft i s based on t h e Lowson wd Ollerhead r o t o r r o t a t i o n a l noise program (Reference 11) modified by 2 . L. Munch t o a l s o predict r o t o r broadband noise (Reference 12). The nethod has since been updated t o include t h e e f f e c t s of blade t w i s t based D n extensive whirlstand t e s t i n g . A graphical representation of t h i s nethod w a s prepared by W. Bausch of Sikorsky Aircraft and w a s included in a comprehensive V/STOL noise prediction report authored by B. Magliozzi (Reference 13). The approach taken w i l l be t o use t h i s method t o c a l c d a t e both r o t o r r o t a t i o n a l and broadband noise i n t h e f a r - f i e l d and correct it back t o near-field l e v e l s . The observer p o s i t i o n w i l l be on-axis t o f u r t h e r simplify t h e equations.

Broadband Noise Rotor broadband noise i s a function of t i p speed (V ), t h r u s t ( T ) ,

t o t a l blade a r e a (%) , blade l i n e a r equivalent t w i s t ( 0 f , and distance

t o t h e observer ( R ) . O n t h e r o t o r axis, t h i s r e l a t i o n i s given by

sP$ = 20 Log v 20 Log T - 10 Log A b - .56 ( 0 )

b t

- 20 Log R + 21.9

This represents t h e o v e r a l l r o t o r broadband noise l e v e l . To obtain t h e octave l e v e l s , a generalized spectrum shape based on t h e r o t o r Strouhal frequency i s used. Rotor broadband noise has long been associated with t h e unsteady vortex shedding at t h e a i r f o i l t r a i l i n g edge. The exact mechanism i s not c l e a r but a scaling with Reynolds number i s observed, hence t h e association with vortex shedding. The frequency of peak broad- band noise ( f ) i s well predicted i n terms of blade chord ( c ) , thickness S (t) and v e l o c i t y

f s = .28 v

.7 hl where h , = c s i n o + t cos o - 7 - 7 and V and 0 represent t h e velocity and blade angle of attack at t h e 70% r d i u s , r6gpectively.

octave center It i s convenient t o set t h e peak frequency t o t h e closest it equal t o 250 frequency. For example, i f f i s found t o be 225 Hz, set Hz. Using t h e generalized oczave spectrum o f Figure 19, t h e individual octave l e v e l s , SPLocT, are determined by adding t h e corresponding band l e v e l SPLBL t o t h e o v e r a l l broadband l e v e l , SPLOA SPLOCT = SPLOA + SPLBL Rotational Noise The harmonics of blade passage f o r a hovering r o t o r can be determined from t h r u s t , torque, t i p speed, and t w i s t . Because t h e t h r u s t t e r m dominates on-axis, t h e torque component w i l l be neglected.

Obtain t h e p a r t i a l l e v e l SPLT from Figure 20 corresponding t o t h e r o t o r t h r u s t ( T ) .

For each harmonic, m, c a l c u l a t e mB and f i n d t h e corresponding p a r t i a l l e v e l SPL from Figure 21. Find t h e correction f o r blade t w i s t , SPLo from Figuge 22.

The swn of these p a r t i a l l e v e l s f o r one harmonic, S P k , epresents t h e t o t a l on-axis l e v e l for a hovering r o t o r at a distance of 00 f t . A t any distance, R S P k = SPLT + SPL, + SPL0 + 20 Log (-) R The t o t a l r o t o r noise octave spectrum i s obtained by adding t h e o t o r harmonics i n each octave, logarithmically, and combining these with he respective broadband l e v e l s . In general, r o t o r noise dominates only he lowest octaves. This i s because of t h e rapid roll-off of r o t o r armonics and t h e s m a l l amount of transmission l o s s provided by s t r u c t u r a l iaterials at low frequencies.

The problem now i s t o t r a n s l a t e far-field octave band data i n t o t h e .ear-field. be accomplished by using t h e d i s t a n c e / d i r e c t i v i t y This can urve for a c i r c u l a r source from Figure 7. By taking t h e noise measured o r predicted) on t h e r o t o r c e n t e r l i n e at 10 r o t o r diameters, and then loving t h e l e v e l s i n on t h e curve t o t h e desired distance below t h e r o t o r lead, an approximation t o t h e near-field r o t o r spectrum i s obtained. One .mportant assumption i s made f o r r o t o r noise very close t o t h e r o t o r disk.

jecause a r o t o r i s not a s o l i d surface r a d i a t o r , it i s assumed t h a t C can ,e no l a r g e r than 1.0. R e a l i s t i c a l l y , t h e r e i s no pressure doubling i n ;pace as t h e r e i s near a w a l l . In equation form, There r i s t h e f a r - f i e l d distance, r i s t h e near-field distance, and L ts t h e r o t o r diameter. A s an example?; l e t L = 72 f t . If t h e octave Level measured at 720 ft i s 100 dB and t h e l e v e l 7.2 f t under t h e r o t o r is desired, then C (7.2/72) = 100 + 1 0 Log SPL C(720/72) near loing t o curve 1 of Figure 7 , C(.1) = 1.8 and C(10) = .001. However, C :annot be g r e a t e r than 1.0, s o take C(.1) = 1.0.

Then 10 Log (1.0/.001) 30 and SPL = 100 + 30 = 130 near Note t h a t using a simple 6 dB/doubling of distance r e l a t i o n would have given 140 dB f o r t h e near-field l e v e l .

This technique can a l s o be used for t a i l r o t o r s when in-plane l e v e l s are required by obtaining C from curve 2 of Figure 7.

Boundary Layer Noise In high speed f l i g h t , boundary l a y e r noise can be a s i g n i f i c a n t p a r t o f t h e noise observed i n t h e a i r c r a f t cabin.

Airframe noise is generated by any p a r t of t h e a i r c r a f t s t r u c t u r e protruding i n t o t h e flow. It can be e s p e c i a l l y intense when generated by s t r u t s , f l a p s , doors, or open c a v i t i e s . However, t h e d e t a i l e d prediction of airframe noise is beyond t h e scope of t h i s report and w i l l not be dealt with*. The CH-53D a i r c r a f t i s r e l a t i v e l y clean i n t h e airframe sense, with no major sources. Only t h e noise generated by t h e turbulent boundary l a y e r w i l l be considered.

The method used f o r calculating boundary*layer noise i s due t o B i e s (Reference 1 4 ) . H i s work i s a summary of wind tunnel and a i r c r a f t measure- m e i n t s made of turbulent boundary l a y e r pressure fluctuations over a wide range of Reynolds and Mach numbers.

The procedure i s summarized as follows. Calculate t h e o v e r a l l fluc- t u a t i n g pressure l e v e l s from t h e equation = 20 Log q + 84 dB (25) FPLover a l l 1/2pUz. The boundary l a y e r where q i s the f r e e stream dynamic pressure displacement thickness i s approximated by

6" = 0.0016 x

(26 1 where X i s t h e distance from t h e leading edge (assumed greater than 10 f t ) . The c h a r a c t e r i s t i c frequency i s determined from

f = 0.1 u,/s+

( 2 7 ) 0 A i n Figure 23 p l o t s t h e non-dimensionalized f l u c t u a t i n g pressure l e v e l s 1 Hz octave bands. To determine t h e dimensional octave l e v e l s , use t h e : following equations

- -

SPL octave where t h e octave bandwidth ( b w ) i s given b y & t i m e s t h e octave center frequency. 2 The following example calculated t h e boundary layer noise at s t a t i o n CHRA fuselage at 150 knots (253.5 f't/sec). The dynamic pressure 342 on t h e q i s = 76.5 l b / f t 2 q = 1/2 (.00238) (253.5) = 20 Log (76.5) + 84 = 121.7 dB FPLove rail The boundary Layer momentum thickness i s 6 = 0.0016 (21) = 0.0336 f t is then given by fo = 0 . ' 1 (253.5/0.0336) = 754.5 H z he following t a b l e lists t h e values of A obtained from Figure 23, t h e alues of SPLlHz9 and t h e octave l e v e l s : %fo A 10 Log(bw) SPL Octave SPLIHz octave frequency 31.5 .04 -3 15 10 5 63 .08 -3 18 108 2 1 1 1 1 125 17 -3 250 .33 -4 113 500 * 67 -5 27 115 1000 1.34 -7 30

1.67 - 10

2000 33 113

4000 5.30 - 1 4 36

10.60 -20 8000 39 109 The SPL's of t h e above t a b l e indicate t h e f l u c t u a t i n g pressure l e v e l s on t h e outside skin induced by t h e turbulent boundary layer.

INTEGRATED METHOD Checkli s t A step-by-step checklist i s presented below t o summarize t h e process of t r a n s l a t i n g an external source strength i n t o an i n t e r n a l noise l e v e l .

1. Define a i r c r a f t s t r u c t u r e and geometry.

2. Identify t h e major sources and paths.

3. Define t h e source strengths at t h e required operating parameters.

Engine noise from Figures 14-18 and Equation ( 1 4 ) . a.

b. Gearbox noise from Figure 12 and Equations (23) and (19).

c. Rotor noise from near-field data and Figure 7.

d. Boundary l a y e r noise from Equations (25)-(29).

4. Translate t h e source strengths i n t o t h e cabin using Equation (14) or (19) as required: a. Determine C ' s from r a d i a t i n g surface s i z e and distance i t o t h e observer v i a Figures 3-7.

b. Obtain transmission loss data from a data base.

c. Calculate room constant data from cabin dimensions and Equations ( 6 ) - (8).

d. Apply Equation ( 1 4 ) or ( 1 9 ) octave by octave.

Tabulate and sum t h e octave data from each source and path t o 5.

obtain t o t a l observed sound pressure l e v e l .

Application t o Engine Casing Noise The use of t h e equations of multibarrier acoustics w i l l be demon- strated by working through t h e calculation of engine casing noise i n t h e Figure 25 presents an i d e a l i z a t i o n of bare a i r c r a f t cabin (Figure 24).

t h e engine/nacelle arrangement. Having engineering drawings of t h e i n s t a l l a t i o n i s e s s e n t i a l i n determining t h e physical dimensions required for t h e calculations. The complicated arrangement can be simplified by making a f e w assumptions. F i r s t , l e t t h e nacelle be a rectangular struc- Next, l e t t u r e with ;he f i r e w a l l corresponding t o t h e o u t l e t w a l l b 0' t h e source be a f l a t surface with dimensions equal t o t h e engine's average cross-section. Faces ao, e and f can be combined since they represent 0' 0 t h e c y l i n d r i c a l nacelle f a i r i n g . Face d reduces t o an open annulus around t h e t a i l pipe extension t h a t serves as t h e engine cooling air e x i t . Face c includes t h e surface area of t h e engine (which a c t s as an inner w a l l ) as well as t h e f i b e r g l a s s engine intake duct.

The nacelle f a i r i n g i s made of reinforced f i b e r g l a s s approximately 0.838 mm (0.033 i n . ) t h i c k . There i s one f i r e - f i g h t i n g access hole i n t h e f a i r i n g . The f i r e w a l l , i n t e g r a l with t h e a i r c r a f t s k i n , i s made of titanium The presence of numerous s t i f f e n e r s and doublers makes it necessary t o use two average values of f i r e w a l l thickness. Engineering p r i n t s i n d i c a t e t h a t 0.457 and 1.09 mm (0.018 and 0.043 i n . ) are appropriate values. The propor- t i o n s are 59% and 41%, respectively. Geometric parameters come from t h e engineering drawings. Some allowances must a l s o be made f o r leakage through access holes and j o i n t s . One percent has been found t o be a good approximation, Consider t h e n a c e l l e cavity at t h e 31.5 H z octave. Table I summarizes t h e data and shows how t h e room constant i s calculated f o r t h a t octave.

Wall a . (nacelle Most of t h e w a l l s are made up of two d i f f e r e n t materials.

f a i r i n g ) i s b u i l t of .033 f i b e r g l a s s (69.8 f t 2 area) but a l s o has a fire- This serves t o reduce t h e net trans- f i g h t i n g access hole (.34 f t 2 ) i n it.

ission loss of t h e w a l l .

To c a l c u l a t e t h e e f f e c t i v e transmission l o s s (ETL)

e a composite panel, use t h e r e l a t i o n

1:= ST

T S 1 ~ l + S2r2 nd ETL = 10 Log 1 : T Then The absorption coefficient a for f i b e r g l a s s at 31.52Hz i s .02.

Multiplying by t h e f a i r i n g area (70.1 f t ) then S(T + a ) + 01 i s 0.72.

Repeating t h e process f r t h e other w a l l s , t h e room constant R s 50.4 ft2.

t 31.5 Hz i s calculated t o be 101 f t Following t h e path of transmission i n t o cavity 1, the d i r e c t i v i t y / i s t a n c e function C must be determined t o f i n d t h e l e v e l impinging on The engine i s approximately a 3 x 1 rectangle with long u t l e t w a l l bo.

There i s The distance r/L i s 1.9/6.7 = 0.28.

i d e dimension 6.7 f t .

o chart for C f o r a 3/1 rectangle, so t h e average of t h e 2 / 1 and 4 / 1 Table I shows that t h e ETL of t h e alues must be taken. Cav i s 0.371.

All t h a t remains is t o 'irewall (bo) i s 9.7 dB an6 T = .lo8 at 31.5 Hz.

t o account f o r t h e propogation 'ind t h e d i s t a n c e l d i r e c t i v i t y f a c t o r C Figure 25 shows If t h e casing noise from t h e Sirewall ko t h e cabin center.

,hat t h e f i r e w a l l i s a 6 x 1 rectangle w i t h long dimension 8.3 f t .

, e t t i n g t h e observer be positioned under t h e main transmission, r/L

Iecomes 4.4/8.3 = 0.53. Taking the average between t h e 4/1 and 8/1

*ectangular source curves, C1 i s 0.085. The work sheet of table I1 :ummarizes a l l of t h e data f o r each octave including t h e casing power .eve1 f r o m t h e source s t r e n g t h data base. Source 0 on t h e work sheet i s All of the required ;he engine i t s e l f , while Source 1 i s t h e firewall.

The .nformation i s known and can now be entered i n t o Equation ( 1 4 ) .

?esulting SPL's are l i s t e d i n t h e last column.

CORRELATION The method w a s applied t o the prediction of CHRA i n t e r n a l noise f o r 30th t h e t r e a t e d and untreated cases. The model included main r o t o r , t a i l r o t o r , engine, main transmission and boundary l a y e r noise.

The noise l e v e l s predicted i n the bare cabin agree very w e l l with the measured d a t a as shown i n Figures 26-27. The best c o r r e l a t i o n e x i s t s i n t h e middle octaves which are dominated by gear noise. The SIL is predicted within 1.2 d B . The lower octaves are high by 2-4 d B i n hover. This i s due t o an overprediction of main r o t o r noise which controls the l e v e l of these octaves. This points out the shortcomings of using a uniform, c i r c u l a r source t o approximate a r o t o r disc. The c o r r e l a t i o n improves i n cruise.

The upper octaves are underpredicted by as much as 10 dB i n both hover and cruise. A review of t h e narrow-band s p e c t r a at these f l i g h t conditions confirmed t h e existence of these high broadband l e v e l s over t h e gear c l a s h harmonic tones. It i s s i g n i f i c a n t t h a t t h i s occurs at such high frequencies. Because of t h e l a r g e amount of transmission loss associated with most materials at t h e upper octaves, these upper octave levels should be very low. This implies that t h e r e is some d i r e c t r a d i a t i o f r o m t h e skin surface or through a leak. The one source t h a t could provide t h e necessary l e v e l s at t h e upper octaves i s t h e engine. It i s l i k e l y f i r e w a l l and surrounding t h a t engine-induced vibration w a s being fed t o t h e frames v i a t h e engine mounts and forced s t r u c t u r a l radiation i n t h i s area.

Engine-induced structure-borne noise was’ not considered i n t h e calculations because of a l a c k of t h e appropriate data.

Correlation of t h e treated cabin l e v e l s shows some i n t e r e s t i n g e f f e c t s (Figures 28 and 29). Predicted and measured l e v e l s agree very w e l l i n t h e upper octaves, unlike t h e bare a i r c r a f t case. This tends t o confirm t h e contention t h a t engine-induced vibration i s being radiated by This 500 Hz octave i s underpredicted by 6 dB.

t h e skins or frames. The i s due t o t h e f a c t t h a t t h e aft bulkhead w a s r a d i a t i n g a s i g n i f i c a n t amount of structure-borne noise. This w a s confirmed during f l i g h t t e s t s of t h e t r e a t e d CHRA when a lead-vinyl c u r t a i n w a s placed over t h e bulkhead.

The l e v e l s observed i n t h e gear noise-dominated octaves dropped 2-6 dB.

The l e v e l s i n t h e 125-500 Hz octaves are underpredicted. This frequency region i s dominated by main r o t o r and engine casing noise implying t h a t t h e r e w a s some s o r t of leakage or panel resonance. It i s probable t h a t these sources entered t h e ECU ducting behind t h e valances. The treatment i s not continuous over t h e frames where t h e valances are attached. Since t h e r e i s not treatment within t h e ECU ducts, any noise entering would be free t o propagate along t h e length of t h e ducting. The contribution of t h e ECU system was not included i n t h e noise prediction method because an adequate model w a s not yet available.

Overall, t h e c o r r e l a t i o n of predicted l e v e l s with measured data i s excellent. Comparisons indicate t h a t t h e method could be improved by adding t h e e f f e c t s of engine-induced structure-borne noise and developing a procedure t h a t would account f o r t h e ECU ducting.

CONCLUSIONS The integrated method presented provides an e a s i l y workable and 1.

correlated procedure f o r t h e prediction of helicopter i n t e r n a l noise. The method i s s u f f i c i e n t l y general t o be applicable t o helicopters and other a i r c r a f t types when t h e appropriate s t r u c t u r a l geometry, noise source strengths, and material acoustic properties are defined.

2. CHRA c o r r e l a t e w e l l with The l e v e l s predicted by t h e method f o r t h e measured data i n both hover and cruise. The hover SIL w a s predicted within 1.2 dB f o r t h e bare a i r c r a f t and 0 . 1 dB f o r t h e treated air- c r a f t . I n c r u i s e , t h e SIL correlated within 0.2 dB f o r t h e bare a i r c r a f t and within 1.2 dB f o r t h e treated case.

3. An accurate definition of t h e problem i s e s s e n t i a l f o r good correla- t i o n between measured and predicted levels. This includes the l o s s and abso near-field r o t o r noise over-predict t h e levels very close t o the rotor disc. A more exact model i s required t h a t accounts f o r the f a c t t h a t most of the acoustic energy i s generated by the outboard sections o disc and is not uniformly distributed.

REFERENCES Levine, L. S. and DeFelice, J. J., "Civil Helicopter Research Aircrafl 1 .

Internal. Noise Prediction" , N A S A Contractor Report , N A S A CR-145146

April 1977.

USA Standards I n s t i t u t e : USAS-SI. 6-1967 Preferred Frequencies and 2.

Band Numbers f o r Acoustical Measurement , 1967

Rathe, E. J., "Note on Two Common Problems of Sound Propagation," 3.

Journal of Sound and Vibration, 10( 3) , 472-479, 1969

Tatge, R. B. , "Noise Radiation by Plane Arrays of Incoherant Sources, 4.

JASA Volume 52, Number 3 (Part I), 1972 Beranek, L. L., "Noise and Vibration Control", McGraw-Hill Book 5.

Company, 1971.

Grande, E. and Brown, D . , "Small Engine Noise Prediction," Technical 6.

Report AFAPL-TR-73-79, Volume 11, December 1973 BOwes, M. A., Giansante, N ; "Helicopter Transmission Vibration and 7.

Noise Reduct ion Program", USAAMRDL-TR-77.14 , June 1977.

8. Howells, R. W. and Sciarra, J. J., "Finite Element Analysis Using N A S T R A N Applied t o Transmission Vibration/Noise Reduction", Paper presented at Fourth N A S T R A N Users' Colloquium at NASA-Langley Researc Center (NASA-TMX-3278) , September 1975.

Dean, M, W. , "Correlation of an Extended CH-53 Helicopter Nastran

9.

Model with F u l l Scale Aircraft Shake T e s t Data," N A S A Contractor Report, N A S A CR-145012, J u l y 1976 10. Sutherland, L. C. and Brown, D . , "Prediction Methods for Near Field Noise Environments of VTOL Aircraft ,I1 AFFDL-TR-71-180, May 1972 11. Lowson, M. V. and Ollerhead, J. B., "Studies of Helicopter Rotor Noise," USAAVWS Technical Report 68-60, January 1969 12. Munch, C. L., "Prediction of V/STOL Noise for Application t o Commit Noise Exposure ,11 Dept. of Transportation Report DOT-TSC-OST-73-19, May 1973 Magliozzi, B., "V/STOL Rotary Propulsion Systems Noise Prediction 13.

and Reduction," Dept. of Transportation, FAA Systems Research and Development Service Report FAA-RD-76-49, May 1976 Bies, D. A., "A Review of Flight and Wind Tunnel Measurements of Boundary Layer Pressure Fluctuations and Induced S t r u c t u r a l Response,

N A S A Contractor Report, N A S A CR-626 , October 1966

TABLE 1.- C A V I T Y AND WALL NOMENCLATURE D

I E T L i 11.5 114.51 1 . 5 I 0 I I

.7 .035 .7 I 1 I

r l

.34 6.5 63 ' 1

ETL2

0 8 27 I

- .158 .0012

T 2 1 1 I

A - OUTROARD

I ETL 11.5 19.7 I 10.11 0 I 1

R - ItlBO&?D

r e f i - .7 .lo81 .097 1 I I

C - FOR\IARD

Q .02 .02 .02 0 I

D - AFT

;?eff+a .72 .128 .117 1

E - UPPER

S f r ~ + a ) r 5 0 . 4 1.41 8 . 5 1 .62 I

F - LOHER

t ft2 1

I 154

NOTE: For tone dominance use transmission loss a t tone frequency, I ft2 =.093 m 2 DIMENSIONS LISTED IN FEET FOR COMPATI BlLlTY WITH Sa' BARRlE R EQUATIONS

I R=-

I- a v z l- a w V z W U W LL W U

*

/ --

% / M A I N TRANSMISSION \ / . . - - - -. . - / \ AIRBORNE STRUCTURE-BORNE \

NOISE 7 A' f

I I / / Figure 1.- Modes of propagation for cabin noise.

TOTAL POWER RADIATED - W (WATTS )

POWER /UNIT AREA

INTENSITY - I (

OBSERVED SOUND PRESSURE

LEVEL - SPLoes

I

I L = IOLOG -

IO PWL = I L + IOLOG S

SPL = PWL-10 LOG S + I O LOG C

Figure 2.- The relation of PWL, I L , and SPL for a plane source.

& W )-I U c) td W h a u u r l t d 3 4 *r( a U c ) 8 a & & t d .rl 3 I o l d m &

i 2 l

rl F a k cw l a I- o a L 2 .

t I- o W E Figure 7.- Directivity factor for a circular disk.

4 0 3 0 I O I Figure 8 . - Effect of leakage on panel transmission loss.

CAVITY 0 CAVITY I OBSERVER SPL, Figure 9.- Two-cavity problem.

a a, U (d *rl a (d k x P k (d a, bo w w a, w w m a, U a, k r L I 0 0

2 s 8

SllWM 1,-01 ' P C J ,lMd HSW13 W39 0321lWWCJON Figure 11.- Radiating surfaces for transmission noise based on the CHRA NASTRAN model.

I I 1 t . I . , . ! . I 140..

e

0 - 1370 H Z - - 4 dB/DOUBLINO

UNPHASED PLANETARY

-, 130.-

-3 - c - - --

aa1.4

---

W > w -I m a W -3 d8lOOUBUNG 120..

m #-527 H t . m a A m 0 L A PHASED PLANE TAW @A

--,I-- - - -- -- A

0.4.0

2 l

4 . 2 dB/DOUBLING & ! IIO-. . A

A - BEVEL 2710 H I . 7-------

I- a 0 s 2.6 a L .

I I * I . I * I . I I Figure 12.- Total radiated sound power levels versus consumed horsepower f o r CH-53 A/D aircraft.

f -20

t

Figure 13.- Genetalized gear noise harmonic spectrum.

I60 I50 JPTO-II

GTCP - 660-4

I 3a I2C I I I I I I I I I I I I I I ( 1 200 400 600 IK 2K K 3K 4 K SHAFT HP I HORSEPOWER : 0.7457 KILOWATTS Figure 14.- Overall sound power l e v e l v s . horsepower for several engine types.

1 1 0

t BO

A i:

8 BO

a

! ! 70

Figure 15.- Composite engine casing noise.

QITIn PASS BANGS IN HERTL 100 1000 lW0a FREQUENCY IN HERTZ Figure 16.- Composite engine exhaust noise.

ACOUSTIC DATA - OCTAVE SPECTRUM REDUCED INLET NOISE SPECTRUM I30 I 2 0 m 110 -I h IO0 N ;.1 t

e

a FREQUENCY I N HERR Figure 17.- Composite engine i n l e t n o i s e .

-I a

E

ai m & m a -

Bo

& w h a- .

aJ a b - U fn a a

w 1

x I- a w n ' I

i !

e - - x & . ! A l l 8 4 2 1 2 4 8 16 32 64 128 BAND CENTER FREQUENCY REFCPEAK FREQUENCY f , F i g u r e 19.- Rotor broadband n o i s e o c t a v e spectrum shape.

ROTOR THRUST, LB I POUND-FORCE -4.448 NEWTONS F i g u r e 20.- H e l i c o p t e r r o t a t i o n a l n o i s e p a r t i a l level based on t h r u s t .

I IO I 1 I I I I I I I 1 1 I I I I I I IO0 m U

-

I J a 90 v) J W a 80 a I- C a n 4 6 8 I O 20 40 60 80 100 2 m B , HARMONIC ORDER X NO. BLADES Figure 21.- On-axis thrust component SPL vs. harmonic order as a function of rotor tip Mach number MT.

m -0

-

BLADE TWIST Q) J a v) -10 J W J -20 I I I I I I I I I 1 I I I I -30 I 2 4 ' 6 8 I O 20 40 60 80 mB HARMONIC ORCER X NO. BLADES Figure 22.- Effect of twist on rotational noise harmonic thrust component.

Iy (d k U

ki

-

a, k 4B m v1 a, k a k a,

B

rl R N Iy k t (d a a W E I P M E *rl U

s

d U m 9) k %I U k u N I m N a, k M d Ezl 0 0 <u Y)

I T

P

I PHYSICAL P R O B L E M PWLO (SOURCE 0 r /1 I A - I I I &* I I

L S , ( SOURCE

I I I

J+I I

I I I C A V I T Y 0 C A V I T Y 1 IDEALIZED P R O B L E M Figure 24.- Engine casing n o i s e i n bare cabin.

SOURCE ENGINE/NACELLE ARRAEIGEMENT \ I N SOURCE 1.4 ft\/' /"

8.3 ft -1

S I M P L I F I E D ARRANGEMENT Figure 25,- Source cavity representation.

low ' FREQUENCY IN HERTI Figure 26.- Correlation of measured and predicted bare aircraft noise levels. Middle cabin; OGE hover.

FREQUZNCV IN HERTZ Figure 27.- Correlation of measured and predicted bare aircraft noise levels. Middle cabin; 150 knots indicated air speed.

63 7 OCTAVE PASS BANDS UI HERR n 3 80 \

z

b

! 4

~ 70 a -1 y 60 W -1 W a 2 50 P 4 0 S I , I , , I 100 loo0 1ww FREQUENCY IN HERTZ F i g u r e 28.- C o r r e l a t i o n of measured and p r e d i c t e d t r e a t e d a i r c r a f t n o i s e levels. Middle cabin; OGE hover.

IO0 F i g u r e 2 9 . - C o r r e l a t i o n of measured and p r e d i c t e d t r e a t e d a i r c r a f t n o i s e levels. Middle cabin; 150 k n o t s i n d i c a t e d a i r speed.

HELICOPTER INTERNAL NOISE CONTROL - THREE C A S E HISTORIES

Bryan D. Edwards and Charlie R. Cox BELL HELICOPTER TEXTRON SUMMARY Three case h i s t o r i e s are described i n which measurable improvements i n t h e cabin noise environments of t h e B e l l 214B, 2 0 6 ~ ,and 222 have been realized.

These case h i s t o r i e s t r a c e the noise control e f f o r t s followed i n each vehicle.

Among t h e design approaches considered, t h e addition of a f l u i d pulsation dam- per i n a hydraulic system and t h e i n s t a l l a t i o n of elastomeric engine mounts a r e highlighted. It i s concluded t h a t substantial weight savings result when the major i n t e r i o r noise sources a r e controlled by design, both i n a l t e r i n g the noise producing mechanism and interrupting t h e sound-transmission paths.

INTRODUCTION Owners and operators of helicopters today expect comfort l e v e l s comparable t o those of other transportation vehicles i n which they t r a v e l . The emphasis on passenger comfort r e f l e c t s recent market trends. For t h e t h i r d consecutive year, business use of helicopters reached a record high i n 1977 with 1219 corporations/executives operating helicopters. This is an increase of 12.7 percent more f i r m s using helicopters than during 1976. Additionally, the num- ber of' c i v i l government agencies operating helicopters increased by 9.5 percent over t h e previous year. The helicopter manufacturer has responded t o these trends by quieting derived versions and incorporating noise source control i n new designs.

The noise environment within a helicopter cabin is made up of contributions

from many sources including t h e rotors, engine( 8 ) , gearing, accessories, and

The r e l a t i v e amplitude of each of these sources may be, aerodynamic turbulence.

and. often is, different f o r each helicopter type. As a general r u l e , the high

frequency sound components of t h e engine(s) , gearing and accessories are t h e

most disturbing.

For existing helicopter designs and t h e i r derivatives, t h e means used t o identify t h e dominant sources include narrowband spectral analysis of noise

and vibration data, detailed mapping of sound transmission paths , and sampling

of production vehicles. Peaks i n t h e spectral analyses are related t o t h e physical properties (rotational speed, number of gear t e e t h , number of blades, etc. ) of each potential noise source Determiningthe path by which t h e sound e n t e r s the cabin i s often d i f f i - Airborne sound can e n t e r t h e cabin through t h e basic s t r u c t u r e , around c u l t .

poorly sealed doors and windows, and through openings, cutouts and ducts.

Conventional soundproofing treatment i s usually e f f e c t i v e i n t h e control of airborne sound, and i t s b e n e f i t s vary as a function of t h e mass o r density and thickness of t h e treatment. O n t h e other hand, sound transmitted t o the cabin v i a t h e s t r u c t u r e of'ten bypasses or flanks t h e soundproofing. The structure- borne path may a l s o amplify t h e o r i g i n a l sound and create'resonances i n panels, air c a v i t i e s and furnishings.

If a resonant o r near-resonant condition e x i s t s , either at the source o r i n t h e s t r u c t u r a l path, l a r g e v a r i a t i o n s i n cabin noise l e v e l s can be experien- ced. In f a c t , such a v a r i a t i o n indicates t h e probability of resonance. Samp- l i n g of a number of production vehicles i d e n t i f i e s the magnitude and extent of t h i s problem. The need f o r corrective action then has j u s t i f i c a t i o n , particu- l a r l y i f t h e action n e c e s s i t a t e s a production change and/or a r e t r o f i t .

For new designs, experience gained i n previous helicopters can be of s i g n i f i c a n t benefit p a r t i c u l a r l y i f t h e general layout , s t r u c t u r e and sound sources are similar. Acoustically weak spots and flanking paths can be designed out and modifications made t o t h e major sources and paths. For designs s i g n i f i - cantly d i f f e r e n t from current experience, t h e manufacturer must r e l y on in-house developed prediction methods and estimating techniques. A n industry-wide need e x i s t s f o r reliable, generally accepted noise prediction methods.

T h i s paper describes three case h i s t o r i e s i n which t h e techniques discussed Figure 1 above have been applied t o t h e B e l l 214B, 2 0 6 ~ and 222 helicopters.

i l l u s t r a t e s t h e general configuration and i d e n t i f i e s the dominant i n t e r n a l noise source of each helicopter p r i o r t o treatment. A s can be seen, t h e offend- i n each of t h e three designs. For the two derived i n g noise source i s d i f f e r e n t versions, noise from t h e hydraulic systems dominates i n t h e 214B, whereas t h e 2 0 6 ~ i n t e r i o r i s predominantly influenced by engine gearbox noise. In the i s minimal and t h e newly-designed 222, noise of the hydraulic system and engine l e v e l i s controlled almost s o l e l y by t h e main transmission.

CASE HISTORY #1: 214B HELICOPTER Configuration The B e l l 214B is a 15-place single-turbine transport helicopter of 6260 T t i s derived from t h e 214A helicopter and received kilograms gross weight.

F A A c e r t i f i c a t i o n i n January 1976.

I n i t i a l Noise Control The i n i t i a l noise control e f f o r t involved evaluation of three prototype soundproofing configurations. These treatments consisted of conventional blankets of d i f f e r e n t thickness and density attached t o t h e cabin roof, aft bulkheads, and side doorposts. Details of t h e t h r e e configurations are i l l u s t r a t e d i n Table 1.

Figure 2 shows t y p i c a l octave band spectra i n t h e aft cabin f o r t h e t h r e e 2onfigurations. I n comparison t o t h e austere treatment , t h e u t i l i t y i n t e r i o r reduces high frequency noise 1 0 t o 15 decibels, f o r doubling of t h e treatment Jeight. I n the VIP i n t e r i o r , additional thicknesses of material a r e used, increasing treatment weight by 40 percent.

This i n t e r i o r provides additional reduction i n t h e very high frequencies, but has l i t t l e e f f e c t on sound below 3bout 2000 H e r t z .

This i l l u s t r a t e s a p r a c t i c a l l i m i t often encountered when using conventional

blanket soundproofing - a point i s reached where additional treatment weight

no longer y i e l d s corresponding noise reduction. Another l i m i t i n g f a c t o r i s t h e d i f f i c u l t y i n s t i t c h i n g together t h e l a r g e number of layers.

Sound Sources

Concurrent w i t h t h e evaluation of various i n t e r i o r treatments , a program

was i n i t i a t e d t o control t h e noise at t h e source. The primary sound sources inside t h e 214B a r e i d e n t i f i e d i n t h e narrowband s p e c t r a l analysis of Figure 3.

Main and t a i l r o t o r noise t y p i c a l l y dominates t h e frequency range below about 500 Hertz. Above 500 Hertz, t h e spectrum contains a number of pure tones r e l a t e d t o gear meshes within t h e main transmission and d r i v e t r a i n , and t o t h e dual hydraulic system.

Hydraulic System Noise Reduction Because of t h e amplitude and number of high-frequency tones produced by t h e hydraulic system, means t o reduce t h i s source were fnlr;estigated. The pump of hydraulic system #1 i s driven by an accessory gear on t h e lower trans- I n mission case and generates noise at a fundamental frequency of 787 Hertz.

a similar fashion, t h e pump of system #2 i s driven from t h e upper case and has a fundamental frequency of 832 Hertz. Both pumps generate a number of harmonic tones r e l a t e d t o these fundamental frequencies.

Noise from each pump i s transmitted t o t h e cabin by f l u i d pressure o s c i l l a - t i o n s , referred t o as pressure r i p p l e , i n t h e hydraulic l i n e s . Pressure r i p p l e Its is s e t up as each piston i n t h e hydraulic pump passes t h e pressure p o d .

magnitude i s on t h e order of 20 kilograms per square centimeter,approximately 1 percent of t h e steady hydraulic pressure of 210 kilograms per square centimeter.

The pressure r i p p l e transdts high-frequency vibratory energy i n t o t h e structure v i a t h e f l e x i b l e hoses, bypass valves, and hard l i n e s clamped t o t h e aft cabin bulkheads and roof. This structure-borne vibration, i n t u r n , generates sound.

To reduce hydraulic system noise, several approaches were considered.

These included vibration i s o l a t i o n of a l l hydraulic system attachment points, pump modification, and t h e addition of flow-smoothing devices i n t h e f l u i d l i n e s t o reduce t h e pressure ripple. O f t h e t h r e e approaches, reduction of the pressure r i p p l e proved t o require t h e l e a s t design e f f o r t and development.

Vibration i s o l a t i o n would have required replacing a l l t h e e x i s t i n g l i n e and hose clamps, and developing elastomeric mounts f o r bypass valves and possibly t h e reservoirs. Modification of a pump with variable-spaced piston cylinders, t o d i s t r i b u t e t h e f l u i d o s c i l l a t i o n s over a random frequency range, would have necessitated.extensive prototype design and t e s t i n g . Reduction of t h e pressure r i p p l e w a s t h e most promising approach.

Laboratory Tests Laboratory t e s t s were conducted t o evaluate t h e flow-smoothing capability of f i v e devices, each of which w a s i n s t a l l e d i n t h e f l e x l i n e i m e d l a t e l y downstream of t h e hydraulic pump o u t l e t . The following devices were evaluated: 1) a pulsation damper consisting of a 300 cubic centimeter spherical volume, a pneumatically charged accumulator , 2 ) a f l u i d f i l t e r normally used for p a r t i c l e f i l t r a t i o n , 3) an "acoustic f i l t e r " which provides a dual path f o r f l u i d flow,

4)

introducing interference e f f e c t s , and 5 ) a variable length hose.

O f t h e devices t e s t e d , t h e pneumatically charged accumulator and t h e pulsa- t i o n damper were t h e most effective. The accumulator reduced pressure ripple by a factof of s i x . However, i t s i n s t a l l a t i o n i n a helicopter would have required a maintenance item t o periodically check t h e pneumatic pressure.

The pulsation damper, which reduced pressure r i p p l e by a factor of f i v e , required no such maintenance and was selected as t h e most p r a c t i c a l flow-smoothing device. The r e l a t i v e amplitude of t h e pressure r i p p l e before and after i n s t a l l a - t i o n of t h e pulsation damper can be seen i n Figure 4. Also schematically i l l u s t r a t e d i s + ? n e e f f e c t o f t h i s pressure r i p p l e reduction on noise inside a simulated passenger cabin.

i ght Te st F1 The pulsation dampers were then i n s t a l l e d i n t h e dual hydraulic system of t h e 214B and evaluated i n f l i g h t . The i n s t a l l a t i o n i s i l l u s t r a t e d i n Figure 5 .

#1, t h e pulsation damper i s mounted t o a transmission support member I n system and connected t o t h e pump and hard l i n e s by means of f l e x i b l e hoses. I n system #2, t h e u n i t i s i n s t a l l e d d i r e c t l y at t h e pump o u t l e t . In both cases, only minor changes are required i n t h e hydraulic hoses and f i t t i n g s . The e n t i r e i n s t a l l a t i o n weighs approximately 2.2 kilograms.

Figure 6 depicts t h e noise reduction realized with t h e pulsation damper.

Sound l e v e l s measured with and without t h e damper a r e compared. I n system #1, the pump fundamental i s reduced by about 13 decibels. The first, second and t h i r d harmonics a r e correspondingly lowered by 3 t o 6 decibels.

In system # 2 , the quieter of t h e two systems, t h e pump fundamental i s reduced by approximately 3 decibels and the first and second harmonics a r e lowered by 3 t o 6 decibels.

Reduction of t h e hydraulic system noise measureably improves t h e cabin noise environment. The dominance of a number of pure tones i s removed, reducing the objectionability and improving speech i n t e l l i g i b i l i t y . The A-weighted A maximum sound pressure l e v e l i n t h e cabin i s reduced by an average of 4 dBA.

Speech reduction of 6 dBA i s realized i n t h e aft passenger seat locations.

Interference Levels a r e decreased by an average of 6 decibels.

Based on the above improvements, t h e pulsation dampers are currently being i n s t a l l e d on all production 214B's. Additionally, the damper concept h a s been successfully applied t o t h e B e l l 212. Due t o differences i n routing o f hydraulic l i n e s i n t h e 212, t h e damper i s required i n only one of the two hydraulic systems. Flight t e s t s have confirmed t h i s and a reduction of about 1 4 decibels i n t h e pump fundamental has been measured. Modest reductions, an average of 2 decibels, i n t h e A-weighted noise levels and the speech interference levels were also realized.

CASE HISTORY #2: 2 0 6 ~ HELICOPTER Configuration The 2 0 6 ~ i s a five-place single-turbines corporate, business and u t i l i t y It is derived from the helicopter w i t h a design gross weight of 1451 kilograms.

2 0 6 ~ helicopter and received F A A c e r t i f i c a t i o n i n August 1971.

Sound Sources The 206 s e r i e s helicopters have a history of high-frequency cabin noise originating from the engine gearbox. One unusual characteristic of t h i s noise Figure 7 i l l u s - is that it varies considerably from one vehicle t o another.

'trates t h i g variation. Cabin noise l e v e l s sampled inside 167 production vehicles are shown. Levels i n t h e 4000-Hertz octave vary from 84 t o 95 decibels for t h e majority o f t h e sample. A t t h e extremes, however, levels as low as 77 decibels a r e possible and as high as 101 decibels.

The 5000-Hz Cabin noise sources of t h e 2 0 6 ~ a r e identified i n Figure 8.

tone which dominates t h e audible spectrum is traced t o t h e mesh frequency of t h e power takeoff (PTO) gear and torquemeter (TM) gear inside the engine output gearbox. Other sources that can be traced include t h e main transmission input pinion gear mesh at 1900 Hertz and t h e planetary stage gear mesh at 1300 Hertz.

The engine and i n t e g r a l gearbox a r e located above and behind the passenger cabin, supported by three s e t s of bipod legs r i g i d l y attached t o t h e gearbox The three engine mounting points are on the and airframe as shown i n Figure 9 .

engine gearbox housing.

Gear mesh vibrations propagate down the support legs d i r e c t l y i n t o a bulkhead aft of t h e passenger seats.

Once into t h e structure, t h e vibratory energy radiates as noise inside t h e cabin.

6 4 3 Engine Gearbox Noise Reduction Two approaches have been taken t o reduce engine gearbox noise. The first The second approach involves means of i s o l a t i n g t h e engine from th9e airframe.

consists of an investigation by t h e engine manufacturer aimed at reducing t h e gearbox vibration at t h e mounting points .

Engine Mount I s o l a t i o n Three engine mount i s o l a t i o n concepts were investigated. The first involves replacing each o f t h e s i x engine support l e g s with a new l e g made up of concen- t r i c m e t a l tubes separated by an elastomer. The second c o n s i s t s of a c i r c u l a r at each of t h e t h r e e engine mounting steel/elastomer washer assembly placed a washer assembly, but with a rectangular points. The t h i r d concept a l s o uses flange which provides greater elastomer area.

Hardware f o r each mount configuration w a s fabricated, i n s t a l l e d i n a t e s t helicopter, and evaluated i n f l i g h t . The concentric metal tubes proved unsuccess- f u l . During ground run, engine motion w a s excessive and f u r t h e r evaluation w a s aborted. However, tests of both t h e c i r c u l a r and t h e rectangular washer assemblies were successful. Both concepts measurably reduced t h e 5000-Hertz gearbox tone and caused no excessive engine motion. Two types of elastomer were evaluated: neoprene rubber and s i l i c o n . The rectangular washer assembly with s i l i c o n elasto- mer provided t h e m a x i m u m attenuation.

The i n t e r n a l s t r u c t u r e and i n s t a l l a t i o n d e t a i l s of t h e rectangular washer assemblies are shown i n Figure 10. Vibrations introduced at t h e engine pad transmit through a 2 millimeter thickness of elastomer before reaching t h e metal of t h e bipod legs. Since t h e engine must be somewhat r i g i d l y retained, the elastomer thickness i s kept t o a minimum. However, t h e frequency of i n t e r e s t i s s u f f i c i e n t l y high (5000 Hertz) t h a t t h i s r e l a t i v e l y t h i n elastomer provides s i g n i f i c a n t i s o l a t i o n .

The flanges of t h e washer assembly ?re elongated i n t o a rectangular shape t o provide as much elastomer shear area as possible within t h e physical c o n s t r a i n t s of t h e e x i s t i n g mount s t r u t s .

Cabin noise measurements with t h e elastomeric washers i n s t a l l e d show The noise 7-10 decibel reduction of t h e 5000-Hertz tone i n t h e aft cabin area.

measured at each passenger location before and a f t e r i n s t a l l a t i o n i s shown i n Figure 11. A t t h e l e f t passenger location, which has t h e highest amplitude before i n s t a l l a t i o n of t h e improved mounts, t h e tone i s reduced by 1 0 decibels.

Levels i n t h e center and r i g h t hand seat location a r e lowered by 9 and 7 decibels, respectively. With t h e improved mounts t h e noise i s f a i r l y constant across t h e aft cabin.

I n s t a l l a t i o n of t h e elastomeric washers i s r e l a t i v e l y simple, requiring enlargement of t h e b o l t hole i n each support l e g , and machining down t h e shoulder of t h e trunnion. The washer design i s such t h a t t h e engine i s w e l l supported even i n t h e event of elastomer f a i l u r e or burnout, and no c r i t i c a l misalignment of t h e engine and driveshafts is possible. Flight tests have shown t h a t engine motion i s w e l l within t h e design l i m i t s and t h e s i l i c o n elastomer i s not susceptible t o chemical or environmental erosion. A s h i p set of t h e mounts weighs less than 0.5 kilograms. Laboratory tests are now being conducted t o I determine the service l i f e of the improved mounts. They are expected t o be f u l l y q u a l i f i e d by J u l y 1978.

Gearbox Vibration Reduction The second approach involves studies and tests of gearbox vibration reduc- t i o n being conducted by Detroit Diesel Allison, manufacturer of the engine.

Figure 12 i s a schematic of the gas producer and power turbine gear t r a i n s , showing the r e l a t i v e positions of the torquemeter (TM) and t h e power takeoff (PTO) gears. This gear t r a i n provides a two-stage speed reduction, converting t h e 33,290 R P M of the power turbine t o 6016 R P M at the power output shaft.

The T M and P T O gears are the primary load carrying gears i n the output drive t r a i n .

The 5000-Hertz e x c i t a t i o n is generated at t h e mesh of these two gears.

Analytical studies i n d i c a t e t h a t both t h e PTO and T M gears have modes of vibration close t o t h e 5000-Hertz meshing frequency. This possible resonant condition would increase t h e vibratory energy transmitted t o t h e bearings, t o t h e gearbox housing, and f i n a l l y through t h e mounting system.

Hardware changes t o the e x i s t i n g gear t r a i n a r e being evaluated on an ex- perimental basis. The modifications and changes under consideration are l i s t e d i n Table 2. Gear tooth p r o f i l e modification o f f e r s the p o s s i b i l i t y of reducing The the excitation by providing a smoother loading/unloading of each tooth.

damper r i n g and the spray applied t o the gear web are intended t o damp out the vibrations transmitted from the gear t e e t h t o the s h a f t . Changes i n the gear resonant frequency by adding m a s s , the mesh frequency by adding gear teeth, and the gear support s t i f f n e s s are a l l designed t o reduce any coincidence e f f e c t s between e x c i t a t i o n and resonant frequencies.

This experimental program i s currently i n progress and f i n a l results are not available.

It i s anticipated t h a t one o r more of t h e above modifications t o t h e An engine w i l l lower t h e high-frequency vibration induced i n t h e engine gearbox.

8 t o 1 0 decibel noise reduction is expected. Coupled with t h e improved engine This w i l l remove mount, a cumulative reduction of 16 t o 20 decibels i s possible.

t h e engine gearing as a dominant noise source i n t h e 2 0 6 ~ helicopter and w i l l re- duce t h e wide v a r i a t i o n i n noise l e v e l from one vehicle t o another.

'CASE HISTORY #3: 222 HELICOPTER The two previous case h i s t o r i e s p e r t a i n t o derived versions and deal w i t h solutions t o e x i s t i n g noise problems, i d e n t i f i e d after t h e helicopter i s i n production. I n a new design, many of t h e s e problems can be avoided i f a t t e n t i o n i s paid t o noise control throughout t h e concept, preliminary design and develop- ment stages. Such i s t h e case f o r t h e B e l l Model 222.

Configuration The 222 is a 6-8 passenger, twin-turbine helicopter designed s p e c i f i c a l l y for the c i v i l market. It is powered by two AVCO/Lycoming turboshaft engines driving a two-stage s p i r a l bevel, single stage planetary main transmission.

Design Features A number of design features a r e incorporated t o reduce cabin noise l e v e l s .

Double roof construction separates t h e primary drivetrain noise sources and the cabin area. Provisions a r e made f o r a continuous layer of soundproofing below t h e lower roof. This treatment has a minimum of constrictions o r openings.

I n t h e hydraulic system, a low-noise pump i s specified. A l l hydraulic l i n e s art kept as short as possible and clamping of l i n e s t o panels i s avoided. A s p a r t of t h e basic suspension system, t h e nodalized pylon incorporates elastomeric bearings. These bearings prevent t r a n s f e r of structure-borne sound from t h e H i g h contact r a t i o t o o t h p r o f i l e s are main power t r a i n t o t h e cabin roof.

Finally, vendor-purchased used extensively i n t h e main transmission.

accessories such as o i l cooler fans, vent/defog blowers and t h e ECU meet stringent noise specifications or are designed t o t h e lowest p r a c t i c a l .

noise l e v e l s .

These design features r e s u l t i n a well-balanced cabin noise environment The prototype soundproofing requiring only minimal conventional soundproofing.

treatment weighs only 1 0 kilograms.

T t consists of foam/lead foil/foam sheets attached t o t h e inner roof and aft bulkhead. Roof t r i m panels of 4 centimeters aluminum sheet extend f r o m t h e aft bulkhead forward and provide a continuous closure over t h e soundproofing treatment. The treatment density varies along Densities of 4.88 kilograms per square meter are used i n t h e aft the roof.

portion d i r e c t l y beneath t h e main transmission. A l i g h t e r density, 2.9 kilograms per square meter, i s used in t h e forward roof.

Figure 13 i s a narrowband frequency spectra of t h e noise i n t h e aft passenger cabin of t h e 222. I n t h e frequency range above 500 Hertz, t h e major sound components emanate f r o m t h e main transmission. The two input pinions' gear mesh i s 3200 Hertz. Gear mesh of t h e planetary stage's spur gears i s 1050 H e r t z . Harmonics of these gear meshes, lower i n amplitude, are also present. Other secondary sources include t h e hydraulic system and other tones not i d e n t i f i a b l e at t h i s t i m e .

The forward passenger s e a t s a r e slightly quieter than t h e aft ones, but i n general t h e noise f i e l d i s uniform throughout t h e cabin. Table 3 compares t h e A-weighted sound l e v e l and t h e Speech Interference Level (SIL) for each seat location. The sound l e v e l i n t h e forward row of passengers averages 85 dBA; 87 dBA i n t h e middle row, and 86 &BA i n t h e aft row. SILs a r e 76, 77, and 78 db, respectively. These l e v e l s vary l i t t l e with air- speed and gross weight. Speech i n t e l l i g i b i l i t y is excellent and passengers can e a s i l y converse with each other.

CONCLUDIEJG REWKRKS The above case h i s t o r i e s i l l u s t r a t e use of available techniques t o control helicopter i n t e r i o r noise levels. Different techniques, it i s shown, are required Existing o r derived designs with noise problems can often f o r each type design.

be improved and require detailed knowledge of t h e source characteristics and New designs can often benefit fromthese experiences, particularly sound paths.

i n savings of weight required for soundproofing treatments.

Figure 1 4 illustrates t h e weight savings benefit. Cabin noise l e v e l s Maximum of the 214B, 206B, and 222 w i t h different i n t e r i o r s are compared.

l e v e l s of t h e three designs with no soundproofing a r e approximately t h e same.

Less The "best seat" l e v e l s , however, are lower i n t h e 222 by 6-8 dBA.

soundproofing weight (10 kilograms) i s required i n t h e 222 t o reach A-weighted levels of 84 t o 89 d B A and STLs of 75 t o 81 dB. The percentages of useful load required for soundproofing t o reach equivalent cabin noise inside t h e 214B, 2 0 6 ~ and 222 a r e 1.6%, 3%, and 0.7%, respectively.

The low soundproofing weight penalty of t h e 222 r e f l e c t s the early applica- t i o n of noise control i n t h e design. Another important benefit i s t h a t future improvements i n noise l e v e l appear t o be possible f o r modest increases i n t h e i n t e r i o r weight.

64 7 TABLE I. 214B PROTOTYPE INTERIOR TREATMENTS Blanket Composition Roof Aft Bulkheads Aft Bulkheads Roof TABLE 11. 206B ENGINE GEARBOX EXPERIMENTAL MODIFICATIONS Reduce Excitation

. Attach damper rings to gear webs

Change Resonant . Add mass to TM and PTO gears

. Change mesh frequency by increasing

number of teeth on both gears

. Increase stiffness of gear case

TABLE 111. PROTOTYPE 2 2 2 CABIN NOISE AND SPEECH INTERFERENCE LEVELS ENGINE OUTPUT .-:&’& . GEARING Figure 1.- Dominant i n t e r n a l n o i s e sources of t h e B e l l 214B, 206B and 222 h e l i c o p t e r s .

. . . .

In

110 T

l=l 100 .',Y H U / < ' W * . / / a .

90 -

B

fr: W rJ3 H

80 -

%

-

1 1 1 I 1 I I I I 31.5 63 125 250 500 1000 2000 4000 8000 OCTAVE CENTER FREQUENCY, HERTZ Figure 2.- E f f e c t of 214B prototype i n t e r i o r treatments.

FUNDAMENTALS

ROTOR NOISE - 4 M A I N TRANSMISSION

I00 UPPER PLANETARY W a H 80 HARMONICS M A I N TRANSMISSION INPUT PINION W rJ3 H REDUCER GEAR z 0 HYDRAULIC SYSTEM #1 HYDRAULIC SYSTEM #2

2o t

I I 8 1 1 1 t I I 10 50 200 1000 5000 20000 20 100 500 2000 10000 FREQUENCY I HERTZ Figure 3.- Frequency s p e c t r a of-214B cabin noise.

PULSATION HYDRAULIC

DAMPER Yu* i

SIMULATED V " \ / W PRESSURE PRESSURE RIPPLE RIPPLE -OUT DAMPER WITH DAMPER F i g u r e 4 . - Laboratory s e t u p and e f f e c t of p u l s a t i o n damper on h y d r a u l i c system p r e s s u r e r i p p l e .

TRANSMISSION ULSATION DAMPER F i g u r e 5.- P u l s a t i o n damper and i n s t a l l a t i o n schematic.

WITHOUT PULSATION DAMPER

0 WITH DAMPER

1 PUMP

1 S T 2ND 3RD FUNDAMENTAL HARMONIC HARMONIC HARMONIC (a) System 81.

4 WITHOUT PULSATION DAMPER

WITH DAMPER PUMP 1 S T 2ND 3RD HARMONIC FUNDAMENTAL HARMONIC HARMONIC (b) System #2.

Figure 6 . - Effect of pulsation damper on hydraulic system noise sources.

E 80 85 90 95 100

MAXIMUM CABIN NOISE LEVEL - 4000 HERTZ OCTAVE, DECIBELS

F i g u r e 7.- Noise sampling of 206 series h e l i c o p t e r s .

TRANSMISSION a 100- INPUT BEVEL t--7 W

-

H z

-

-

4 0 201 I t I I I I I I 1 10 50 200 1000 5000 20000 20 100 500 2000 10000 FREQUENCY, HERTZ F i g u r e 8.- Frequency s p e c t r a of 206B c a b i n n o i s e .

GEARBOX HOUSING ENGINE STRUTS Figure 9 . - 206B engine mount assembly.

(a) Exploded view of mount assembly.

- TRUNNION MOUNT

(b) Cross section of installed mounts.

F i g u r e 10.- 206B improved engine mounts.

STANDARD ENGINE MOUNTING WITH IMPROVED ENGINE MOUNTS

U

RIGHT LEFT CENTER SEAT SEAT SEAT Figure 11.- Effect of engine mounting on 206B cabin noise l e v e l s .

POWER TAKEOFF GEAR Figure 12.- Relative positions of engine torquemeter (TM) and power takeoff (PTO) gears.

- ROTOR NOISE _L_._l

HYDRAULIC

r SYSTEM

n I \ m

t;: /

ffl 80 U W ca @ 60 W 8 4 w m 1 I 1 I I I I I I 10 50 200 1000 5000 20000 20 100 500 2000 10000 FREQUENCY, HERTZ Figure 1 3 . - Frequency spectra of 222 cabin n o i s e .

0 NO TREATMENT

UTILITY TREATMENT WITH HYDRAULIC PULSATION DAMPER 0 10 20 30 40 50 SOUNDPROOFING WEIGHT, KILOGRAMS Figure 14.- Cabin n o i s e l e v e l s versus soundproofing weight.

AN ANALYTICAL EIETHOD FOR DESIGNING LOW NOISE

HELICOPTER TRANSMI ss IONS*

Robert B . Bossler, J r . , and Michael A. Bowes Kaman Aerospace Corporation A1 1 en C. Royal Appl ied Technology Laboratory U.S. Army Research and Technology Laboratories (AVRADCOM) SUMMARY The internal noise levels i n most c i v i l and military helicopters a r e excessive.

In general, the dominant source of noise is the geared Power transmission sys- tem which operates a t a h i g h power level and i s i n proximity t o the cabin. Within this system, vibratory excitations a r e produced a s a by-product of the gear meshing process. These excitations r e s u l t i n vibration of the transmission housing and airframe which ultimately radiate noise i n t o the cabin. This paper discusses the development and experimental Val idation of a method for analytically modeling this noise mechanism. T h i s method can be used within the design process to predict i n t e r i o r noise levels and t o investigate the noise reducing potential of a1 ternative transmission design details. Examples are discussed.

I NT RODUCT I ON The nature and extent of the internal noise problem i n Army helicopters will be i l l u s t r a t e d first. In present day c i v i l and military helicopters, the transmission i s mounted i n a p o s i t i o n close t o the cockpit. A s the single most dominant source of internal noise, i t i s imperative t h a t the noise produc- i n g characteristics of this component be understood. A hypothesis t h a t was investigated and verified under Army-sponsored e f f o r t s is that noise i s gener- ated by the transmission case a s a r e s u l t of nonuniform transfer of torque from p i n i o n t o gear due t o t o o t h profile errors or t o the e l a s t i c deformation of gear teeth under load. T h i s nonuniform transfer of torque produces a dynamic force a t the gear mesh frequency and i t s multiples, resulting i n a coupled torsional/lateral vibration response of the gear shaft. The l a t e r a l vibration (bending) produces displacements a t the bearings which i n t u r n cause the case t o vibrate, t h u s producing noise. Figure 1 presents allowable and predicted noise levels for the Army Heavy-Lift Helicopter. I t i s clear t h a t the noise levels will be unacceptable if a means of controlling the acoustic energy a t i t s source i s not found.

* Performed under Contract DAAJ02-74-C-0039 t o the Applied Technology Labora-

tory, U. S. Army Research and Technology Laboratories (AVRADCOM) In order t o meet MIL-A-8806, he1 icopter manufacturers have been using increas- ing amounts of noise attenuation blanketing with l i t t l e success and with the penalty of extra weight offsetting the technology advances i n light weight gearboxes. See figure 2.

This paper presents the results o f a program (reference 1) t h a t i s part of 16 years of Army-sponsored programs aimed a t understanding and control1 ing heli- copter internal noise and vibration and their effects on personnel and system components.

MODELING METHOD Problem A helicopter transmission such as the SH-2D transmission of figure 3 i s a complex dynamic system comprised of many interconnected mechanical elements.

This system responds, under the influence of periodic forces produced a t the various gear meshes, causing vibration of a l l of the individual elements and noise radiation from the transmission housing. The mechanical elements involved i n this response include a l l of the gears/gearshafts, their support bearings, and the transmission housing. Analytical prediction o f the trans- mission response, and consequently i t s vibration and noise generation charac- teristics, requires knowledge of the dynamics of each mechanical element, the nature and extent of coupling among these elements, and the characteristics of the gear mesh induced forcing functions.

The problem of transmission response prediction i s compounded by the fact t h a t the gear meshing forces involved occur a t high frequency, typically in the hundreds and thousands of Hertz. Since the frequency for which accurate response prediction can be made i s a direct function of the degree of detail in which the individual mechanical elements are modeled, very detailed elemen- system tal models are required. If the direct approach i s taken and the total such as with f i n i t e element modeling methods, the i s modeled as a whole, in an excessively large, requirement t o model in great detail quickly results complex model which i s n o t amenable t o ease of manipulation and use.

Approach With the method described herein, the transmission i s modeled as a fully r o t a t i n g gearshafts, the shaft supporl coupled dynamic system consisting of a1 1 bearings, and. the housing (fig. 4). Individual mechanical elements are modeled separately, in detail, then combined t o produce the complete system model. The concept of component synthesis i s used t o simplify the dynamic modeling task and t o reduce the size and complexity of the ultimate system model. Gear mesh v i bratory excitations, in terms o f relative deflections between mating gear teeth, are calculated independently through consideration of appropriate time-dependent tooth compliances and gear errors. These excita- t i o n s are introduced in the system model a t the proper gear mesh coordinates, and responses are calculated in terms of shaft and housing displacements and radiated sound power 1 eve1 .

Component Synthesis The technique of component synthesis may be used t o calculate the dynamic response of a linear complex structure and modification of i t a t a r e l a t i v e l y smal 1 number of discrete frequencies ( r e f . 2 ) . The component synthesis tech- The f i r s t of these r e l a t e s t o the nique w e use has two important features.

The analysis of each basic component i s reduction i n degrees of freedom.

carried out w i t h a s many degrees of freedom a s is necessary f o r a valid analysis. When the resulting analytical model i s used, however, the number of degrees of freedom may be d r a s t i c a l l y reduced and must include only: (1) Those which interface other components ( 2 ) Those which a r e t o be affected by changes ( 3 ) Those a t which a force i s applied or dynamic response i s specifically desired T h i s reduction in the number of coordinates i s performed only once a t each frequency of i n t e r e s t w i t h no loss i n the validity of the analytical model, regardless of the extent of this reduction.

The other important feature r e l a t e s to the ease w i t h which changes may be studied. Structural modifications such as local mass o r s t i f f n e s s changes, the addition of springs o r dampers between components, addition of vibration absorbers, and changes i n boundary conditions may be exactly modeled a t virtu- a l l y no computer cost and without performing a new modal analysis f o r each change.

A linear structure i s often represented i n the frequency domain as an imped- ance matrix. The s t a r t i n g p o i n t for the analysis requires a valid impedance matrix f o r each substructure f o r each frequency of i n t e r e s t . The c r i t e r i o n f o r a valid impedance matrix i s t h a t i t correctly predict the motion a t the coordinates of interest a t each frequency of i n t e r e s t . I t can be shown t h a t the c r i t e r i o n f o r a valid impedance matrix i s t h a t the elements of i t s inverse T h i s correctly represent the true response characteristics of the structure.

argument leads t o a d i r e c t method of obtaining a valid reduced impedance matrix as follows: Perform a structural analysis u s i n g conventional methods t o obtain a valid, f u l l size, impedance matrix a t each frequency of i n t e r e s t , Z ( W ) .

Invert Z a t each w t o obtain valid, f u l l s i z e mobility matrices Y ( w ) , Alternately, a modal approach or a direct integration technique may be used t o obtain Y ( w ) .

Select elements from Y a t each w corresponding t o the coordinates t o be retained. These elements a r e then formed into a new reduced mobil i t y matrix, Y,(w).

(4) The reduced impedance matrix i s then formed by inversion of YR: I t is t o be noted that ZR i s valid only a t the frequency a t which i t has been computed. However, ZR represents a physical system which, a t the fre- quency w behaves precisely as the system under study.

The reason for specifically obtaining the impedance matrix of the reduced sys tem is t h a t the impedance o f a complex structure is obtained by simply adding the impedance matrices of the separate components a t coordinates where the deflections are common. The substructures m u s t be modeled a s i f they were unrestrained a t the interface coordinates.

There are several considerations involved i n applying this technique t o prac- tical analyses : I t i s not important how the reduced mobilities are computed as long as they are valid.

For the method t o be practical, the number o f reduced coordinates must not be so large that matrix inversions become prohibitive.

Local impedance changes due t o addition of spring-mass systems o r boundary condition changes are simply added t o the reduced component impedances.

W h e n adding impedance matrices, the corresponding elements must represent deflections in the same direction.

The impedance elements add when their deflections are equal; t h u s , when components are separated by spri ng-damper devices, the impedance of this device must be added t o one o f the substructures prior t o synthesis.

The ability t o accommodate structural modifications including local mass o r stiffness changes, the addition of springs o r dampers between components, addition of vibration absorbers or changes i n boundary conditions i s a valuable feature of a structural dynamics program. With the impedances o f each element a t frequencies of interest stored in a data bank, variation in structural parameters of the individual components can be readily made. The technique i s described i n reference 2.

Application of component synthesis methods resulted i n substantial reduction the size of the impedance matrix ultimately used t o represent the trans- in mission system. The SH-2 transmission system consists o f four gearshafts and a planetary system.' The total number of degrees of freedom used t o model these shafts was 445. The case was modeled with 44 degrees of freedom, f o r a t o t a l of 489 individual element degrees o f freedom. The synthesized system impedance model for these combined elements would have been a 473 x 473 term matrix, considering elimination of dupl icated interconnection points for the gear meshes and bearings. Reduction of many extraneous degrees of freedom from this impedance model reduced it to a 48 x 48 matrix. Although further reduction could have been achieved, this matrix size was compatible with the available computer capability, and no additional reduction was performed. It is interesting to note, however, that the most commonly performed response calculation, prediction of case surface response to a single gear clash force, could have been accomplished with a 1 4 x 14 impedance matrix, which represents a 34/1 reduction in size from the unreduced system impedance matrix.

Transmission Case Modeling First, Modeling of the transmission housing is accomplished in four steps.

id mass representation of the housing is an approximate, physically Val intuitively derived. Next, the actual housing is shake tested with modal responses measured at all physical coordinates considered in the intuitive mass model. The measured modal data are then used to adjust the approximate mass representation, consistent with the restraints imposed b y the orthogo- nal ity relationships. Finally, the adjusted or "identified" mass representa-

tion and modal data are used to derive the housing impedance, stiffness, and

damping matrices. The method is described in reference 3.

In the present program, modal responses of the SH-2D main transmission housing

were measured at 44 locations. The transmission case shake test set up is shown in figure 5. Prior to testing, an approximate mass matrix was estab- lished, consisting of 44 diagonal and 206 off-diagonal terms. Both real and imaginary case mobilities were measured, with the imaginary mobilities used as an approximation of the normal mode responses. Use of these data with the

orthogonality relationships produced only small changes in the approximate

mass matrix, typically less than five percent, indicating a high degree of physical as weil as mathematical model validity.

Comparisons of measured and calculated real case mobility were made in order to verify the accuracy of the resulting case dynamic model. This represents an independent check of model validity since real data were not used in the

derivation. These comparisons showed good correlation, as indicated in

figure 6.

Shaft and Bearing Modeling As for the case modeli'ng, the objective of gearshaft modeling was to obtain valid mechanical impedance representations of each shaft, which could then be joined with the case impedance matrix using component synthesis methods, to

form a dynamic response model of the total transmission system. In contrast

to the case modeling approach, which is based on physical test data, gearshaft An extension of the modeling was accomplished b y purely analytical methods.

Holzer-Myklested technique for dynamic modeling of slender shafts was used with local non-slender shaft elements, such as the gears themselves, treated as lumped masses and' inertias (ref. 4 ) . Both shaft flexure (bending) and torsion were considered with a typical shaft consisting of 100 degrees of

freedom. All shaft support bearings were modeled as nonl inear orthogonal

springs using the method of reference 5.

Excitation Mechanism

The basic mechanism for gear mesh excitation has been incorporated in an

analytical calculation technique which permits the determination of local tooth deflections, including fundamental and harmonic components, based on known tooth geometry and loading conditions. This technique was developed through previous Army research efforts and is described in detail in reference 6.

Within the present effort, this method has been improved to the extent of incorporating an equivalent spur gear approximation technique for representing he1 ical and spiral bevel gearing. This improvement permits calculation of helical and spiral bevel gear mesh excitations directly from gear data available on gear design drawings. The equivalent spur gear approximation is that described in Appendix I11 of reference 7.

Acoustic Source Representation

In the present analysis, the transmission case is assumed to consist of a

relatively small number of simple, baffled, hemispherical acoustic sources.

These sources, which are distributed over the case surface, are assumed t o act independently, with the sum of their acoustic outputs equal to the total transmission radiated noise. The output from each source is computed

directly in terms of sound power level. Use of this source representation

requires only knowledge of case surface motions, amplitude and frequency, and an estimation of the individual source sizes. The total housing radiated sound power is then calculated as the sum of the contribution from each individual source.

TRANSMISSION TESTING Testing was performed to determine the actual vibration and noise character- istics of an operating helicopter transmission. These data were needed, for comparison with analytically calculated transmission noise and vibrati'on The test article characteristics, to Val idate the analytical methods used.

used in this effort was the SH-2D helicopter main transmission shown in

figure 3. Gearbox identities are given in table I. This gearbox is rated at

1695 newton meters (15000 pound inches) of torque (continuous), at an output (main rotor) speed of 287 rpm. Speed reduction through the transmission is 21.311. This test article was subjected to simulated operational testing using a regenerative test stand. Measurements were made of all significant dynamic response characteristics i ncl udi ng : o Shaft bending strain o Shaft torsional strain o Lateral shaft displacement o Housing surface acceleration o Radiated sound pressure level -esting consisted of recording data signals corresponding t o each o f the lynamic parameters a t discrete points over a range of transmission torque md rprn settings. All t e s t data were recorded on analog tape and reduced )ff-line using a real time frequency analyzer.

METHOD CORRELATION halytical predictions of the dynamic responses of the SH-ED main transmission iere made for comparison w i t h the measured t e s t data. While predictions )f a l l relevant transmission responses were made, only case acceleration ind radiated noise proved t o be of value in correlating the analytical method.

iince the accuracy of the vibration and noise radiation predictions i s very iuch dependent on the accuracy of the shaft response predictions, good agree- lent between measured and predicted acceleration and noise radiation charac- leristics provides t a c i t correlation of the s h a f t response prediction method.

Case Acceleration 'redictions of case surface acceleration a t fourteen locations were made and :ompared t o accelerations measured a t these same points. Comparisons were lade a t each gear mesh related frequency of interest, including: o Planetary system fundamental and second harmonic o S p u r gear mesh fundamental and second harmonic o Spiral bevel gear mesh fundamental since two transmission speeds were considered in b o t h the analytical and t e s t ?fforts, a total of ten discrete frequency acceleration components were ivailable for comparison, covering the frequency range of 348 Hz t o 3060 Hz.

Examples o f comparisons of measured and calculated case accelerations are shown in figures 7 and 8. The data of figure 7 show the responses t o the ilanetary system fundamental gear mesh frequency a t 80% transmission speed For 60% torque conditions. Figure 8 illustrates similar data for the second harmonic of the spur gear mesh frequency a t 100% rprn a t 80% torque.

The h i g h degree of correlation indicated by these data i s similar t o that ibtained a t the other excitation frequencies considered.

Radiated Noise lnalytical pred ctions were made for the housing radiated sound power levels issociated with each of the gear mesh excitation frequencies considered for 3 0 t h conditions of torque. Sound power levels were also calculated from the neasured sound pressure levels. Sound power is not a directly measurable Darameter b u t must be calculated from sound pressure. Comparisons of measured and predicted sound power levels are shown i n figure 9 a t the 80% rpm t e s t Excellent correlation i s shown with the average deviation condition.

between measured and predicted sound power levels less t h a n 2 dB.

METHOD APPLICATION The transmission dynamic modeling technique developed in the present program permits the rapid and economical evaluation of transmission design changes.

Once the individual mechanical element models have been derived, they can be manipulated i n various ways without the need for rederivation. This is accom- plished through the use of a computer routine, which i s an inherent part of the system modeling method and which can be used t o perform the following functions : o Add (or subtract) structural damping t o any element or any part of an element o Add vibration absorbers a t any location of an element o Add (or delete) lumped masses a t any location o Add spring/damper systems between any two elements or from an element t o ground o Change system geometry The performance of system design studies i s further promoted by the fact t h a t changes in individual elements may be made separately. For example, i f a change in shaft stiffness or mass distribution i s desired, only the shaft The remaining shaft and housing models model in question need be changed.

i s synthesized using the new shaft are l e f t alone, and a new system model model with these unchanged models.

The purpose An applications study was performed using the analytical method.

of this study was t o demonstrate the range of transmission design changes which may be Ainvestigated with the method. Design changes which were con- sidered in this study effort are given in table 11. While a considerable range of design changes was investigated, none of these individual changes were studied in sufficient depth t o establ ish their ultimate practical value or noise reduction potential. The study results do, however, serve a s an indication of the relative sensitivity of transmission response t o the various design changes which were considered, a t least with regard t o the particular transmission studied.

Shaft S t i f f ness D i s tr i b u t i on

Stiffness distributions of the i n p u t , spur/bevel, and o u t p u t shafts were analytically simulated by changing the stiffness cross section o f the respec- In each case, shaft tive shaft models over a limited segment of each shaft.

stiffness cross section was increased by approximately 10% over one-third of the shaft length. Only the central section of the shaft was stiffened, and no mass was added t o the shaft.

The effect of increasing i n p u t shaft stiffness is shown in figure 10, i n .ems of changes i n radiated sound power level f o r each mesh excitation fre- luency. The changes g i v e n are r e l a t i v e to sound power levels calculated f o r .he baseline transmission. A s indicated, increasing the i n p u t shaft s t i f f - less caused significant changes i n radiated sound power level a t several mesh 'requencies and not merely a t the spiral bevel gear mesh frequencies of 2448 Iz and 3060 Hz which a r e most d i r e c t l y associated w i t h the i n p u t shaft.

~lthoughthe greatest change, an 11 dB reduction, d i d occur a t the 100% rpm !piral bevel gear mesh frequency of 3060 Hz, a comparable magnitude change i n t h i s case a 10 dB increase) i s shown f o r the 80% rpm, spur gear mesh #econd harmonic frequency, a t 2396 Hz. Furthermore, no change i n sound power eve1 was obtained a t the 80% rpm spiral bevel gear mesh frequency of 2448 Hz.

'he data of figure 10 provide a graphical illustration of the f a c t t h a t the nalytical model considers the transmission as a coupled dynamic system w i t h 'esponses determined by a l l the mechanical elements acting a s a u n i t . T h i s 'act must always be considered in applying this method, particularly when 'it Such changes, s used to evaluate potentially beneficial design changes.

.lthough usually predicated on the basis of reducing the response t o only me gear mesh excitation, will normally have an e f f e c t on a l l mesh induced 'esponses, and furthermore, these e f f e c t s will be a function of transmission ,peed. While a given design change may produce a reduction i n response It the principal mesh frequency of i n t e r e s t , this same change may very well 'aise the responses a t other mesh frequencies, t h u s curing one problem and :reating others. In addition, a reduction obtained a t one transmission speed lay not prevail a t another speed, even i f these two speeds a r e reasonably :lose. Because of these considerations, transmission design changes should ilways be evaluated w i t h regard t o t h e i r e f f e c t on a l l gear mesh induced 'esponses and f o r pertinent transmission speeds. Although this approach loes require extensive evaluation of each design change, the analytical iethod has been s e t u p t o perform the required analyses i n an economical, !fficient manner requiring a minimum e f f o r t on the part of the analyst.

P1 anetary System Carrier Stiffness iince, i n many cases, helicopter transmissions exhibit t h e i r highest responses lue t o planetary system excitations, an attempt was made t o develop and !valuate a method f o r changing these responses. These e f f o r t s concentrated in the e f f e c t s of planetary system c a r r i e r s t i f f n e s s modificatlon, and an ?xample of the r e s u l t s of these investigations is shown i n figure 11.

the e f f e c t s of reducing the radial s t i f f n e s s -he data of figure 11 i l l u s t r a t e T h i s change was considered practical because if the planet c a r r i e r by 50%.

mly radial s t i f f n e s s was changed w i t h torsional s t i f f n e s s held constant.

iince system torque i s reacted by the c a r r i e r i n torsion w i t h l i t t l e o r no ; t a t i c load reacted i n the radial direction, the c a r r i e r radial s t i f f n e s s is l o t a primary s t a t i c design factor and can be changed based on dynamic %equirements. A s shown, reducing planet c a r r i e r s t i f f n e s s causes si\gnificant :hanges i n the planetary system responses a t 348 Hz, 435 Hz, and 696 Hz.

:urther, the e f f e c t s are isolated to the planetary system excitations w i t h i t t l e or no response change shown f o r the remaining gear mesh excitations.

While further analytical work i s required, i t i s f e l t that the beneficial effects of this concept could be readily applied in future helicopter trans- mission designs.

Transmission Housing Modification One of the major advantages of the present analytical approach i s the ability t o model the transmission housing.

While a prototype housing i s required t o develop this model, changes in the housing can be simulated by purely analytical means. In this way, changes i n mass and stiffness d i s t r i - b u t i o n s and housing damping can be considered.

The addition of external damping treatments t o transmission housings has often been suggested as a means t o reduce housing response and radiated noise.

W i t h the present program, this approach has been evaluated analytically by simulating surface damping t h r o u g h increasing the housing structural damping coefficient. Three levels of damping increase were considered, with structural damping coefficient (9) of 0.05, 0.1, and 0.2. The structural damping of the housing itself was determined t o be very low, w i t h modal damping coefficients ranging from 0.0015 t o 0.03, Increasing damping t o the degree considered, then, represents a substantial increase, b u t one which can readily be obtained with commercial materials.

The effects of increased housing damping are indicated in figure 1 2 . As shown, appreciable sound power level reductions were obtained a t several gear mesh excitation frequencies, b u t the reductions were by no means universal. This i s t o be expected since the effects of damping are dependent upon the proximity of excitation frequencies and system natural response frequencies. For exci- tations close t o natural frequencies, damping can be effective; while, i f excitations are substantially removed from the natural frequencies, damping will have no effect. As shown, damping can also produce an adverse effect since added damping may increase response t o excitations which are close t o system antiresonant frequencies.

Given the data of figure 1 2 i t i s apparent t h a t housing damping i s a sensitive parameter which can be adjusted t o reduce transmission response. Proper application of this approach, however, requires knowledge of system dynamic response characteristics, most importantly the proximity of gear mesh exci ta- tion frequencies and system resonant and antiresonant frequencies.

FUTURE USE The noise modeling method can be applied in various ways depending upon the development status of the subject transmission. During preliminary design,

rough estimates of transmission noise can be made using a simplified noise

The model which has been derived from the more complex method (fig. 13).

simplified noise prediction method i s given in equation form i n reference 8.

The accuracy of the simplified noise prediction method has been subjected t o 1 imited evaluation through comparison of predicted and measured U H - 1 internal noise levels, obtained from reference 6. As illustrated in figure 14 the :alculated UH-1 internal noise spectrum agrees well w i t h the measured data.

'he noise prediction method i s presented i n equation form i n the appendix.

/hen detail design data become available, the complete system modeling method :an be applied u s i n g a housing model based on f i n i t e element methods. A t this itage, significant efficiencies can be achieved u s i n g a substructure approach ri t h component synthesis. Finally, d u r i n g hardware development, improved !lementa1 models may be obtained through the use o f mobility test data.

CONCLUSION AND R E C O M M E N D A T I O N S 'he analytical methods devel oped i n this study represent a significant advance- lent i n the s t a t e of the a r t of helicopter internal noise prediction. These iethods a r e limited, however, t o the prediction of the airborne component If transmission noise, although the approach used i s compatible w i t h the ncorporation of a structure borne noise prediction capability. Extension If the methodology t o include structure borne noise prediction capability is :onsidered both feasible and appropriate.

APPENDIX H E L I COPTER TRANSM I SS ION N O I S E P R E D I C T I O N METHO'D MICHAEL A. BOWES I t 1s often desirable to have reasonable, though A parametric study was performed, using the approximate, estimates of transmission noise available SH-2D transmission analytical model , characteristics. T o answer this need, a simpli- considering three types of gear meshes, a l l of fied transmission noise prediction technique Gear which were represented in the SH-2D model.

has been developed using parametric trending data mesh types considered were: spur gear, spiral generated with the SH-2D transmission analytical bevel gear and planetary system. Based on this model. The validity of this simplified method i s study, three equations of the form of Equation predicated on the assumption that the S H - E D trans- (1) were derived for the three gear mesh types mission has dynamic response and noise radiation considered. These are: characteristics which are representative o f hellcopter transmissions. This assumption i s believed to be appropriate, since the SH-2D transmission i s similar in design t o most exist- ing he1 icopter transmissions and i t s operating torque and rpm conditions are near median values for current and planned vehicles.

The simp1 ified transmission noise prediction method i s based on a simple parametric relation- ship between the physical variables of a given where: PWLsG = sound power level of spur gear mesh and the sound power level of the dis- gear mesh crete frequency component due to t h a t mesh. The general form of this relationship is: sound power level o f spiral PWLsBG = bevel gear mesh PWLG = A Loglo(') + B Loglo(f) + C + D (1 1 PWLps = sound power level of planet system where: PWLG = sound power level - dB re.

watts A = a constant indicative of the relationship between torque and sound power level

- -

T transmitted torque (in-lb) B = a constant indicative of the relationship between gear clash frequency and sound power level

f = - Hz

gear clash frequency c = a constant indicative o f the type of gear mesh Harmonic No. D = a constant indicative of the gear clash harmonic number 2 3 Gear Clash Type -5 -22 Spur Gear +7 +6 Spiral Bevel Gear -10.5 -23 Planet Sys tem 1 1 I I REFERENCES

. Bowes, M. A., et al, Helicopter Transmission Vibration and Noise

Reduction Program, Eustis Directorate, USAAMRDL TR 77-14, February 1977.

!. Berman, A . and Giansante, N., CHIANTI - Computer Programs for Parametric

Variations in Dynamic Substructure Analysis, Presented at the Shock and Vibration Symposium, A1 buquerque, N.M., October 1975.

Berman, A., System Identification of a Complex Structure, AIAA Paper No. 5 .

75-809, Presented at AIAA/ASME/SAE 16th SDM Conference, Denver, CO.

May 1975.

I.. Pan, C. H. T., Vibration Analysis for Shafting o f Power Transmissions, Report No. SRC 76-TR-20, Shaker Research Corporation, Ballston Lake, N.Y. , November 1976.

5 . Jones, A. B., A General Theory for Elastically Constrained Ball and

Radial Rol ler Bearings Under Arbitrary Load and Speed Conditions , Journal of Basic Engineering, ASME Transactions, June 1960.

5. Laskin, I., Orcutt, F. K. and Shipley, E. E., Analysis of Noise Generated

He1 icopter Transmission, USAAVLABS TR 68-41 , Mechanical Techno1 -

b y UH-1 ogy, Inc. , June 1968.

Badgley, R. H . and Laskin, I . , Program for Helicopter Gearbox Noise Pre- 7 .

diction and Reduction, USAAVLABS TR 70-12, Mechanical Technology, Inc. , March 1970.

3 . Bowes, M. A,, Development and Evaluation o f a Method for Predicting the

Vi bration and Noise Characteristics of He1 icopter Transmissions, AHS

Preprint No. 77.36-76, Presented at the 33rd Annual National Forum of the American Helicopter Society, May 1977.

TABLE I. SH-2 MAIN GEARBOX I D E N T I T I E S E x c i t a t i o n

No. o f S p e e d - Frequency -

P a r t Teeth H e r t z r P m

- -

Input S h a f t 61 20 S p i r a l Bevel P i n i o n 30 6120 3060 S p i r a l Bevel G e a r 47 3906 S p u r G e a r P i n i o n 23 3906 S p u r G e a r 87 1033

-

S u n G e a r 35 1033

-

P1 anet G e a r (6) 28 R i n g G e a r

-

P l a n e t C a r r i e r 287 O u t p u t S h a f t 287 TABLE 11. SUMMARY OF DESIGN CHANGES ANALYZED REDUCED BEARING STIFFNESS SUN GEAR ISOLATION A L L SHAFTS INPUT SHAFT ONLY PLANET CARRIER ISOLATION INCREASED SHAFT MASS INCREASED SHAFT STIFFNESS INPUT SHAFT INPUT SHAFT SPUR/ B EV E L SHAFT OUTPUT SHAFT SPUR/BEVEL SHAFT SPUR/SUN SHAFT INCREASED CASE MASS INCREASED CASE DAMPING BEARING RELOCATION PREDICTED FOR ARMY HEAVY-LIFT H E L I C O P T E R 1 15 MINUTE EXPOSURE L I M I T PER M I L - S - 8 8 0 6 B

-

OCTAVE-BAND SOUND-PRESSURE LEVEL d B @e 0.0002 dyne/cm2) l o o t I I I 1 1 I I

-

' coo 20 50 100 200 500 1000 2000 5000 10

CENTER FREQUENCY - HERTZ

Figure 1 . - Allowable and predicted sound pressure levels at personnel locations for crew members wearing SPH-4 headgear.

(1 500)

200 4 MAIN GEAR BOXJ I

I WEIGHT I NEWTONS I (POUNDS) I I I I I I I I I I I I I 1 1 1 1 1 1 1 1 l 1 ~ ' 1963 1968 1973 1978 1983 1385 YEAR Figure 2 . - Weight of gearbox and soundproofing as a function of time.

ROTOR SHAFT PLANET CARRIER R I N G GEAR PLANET GEAR

\ ? SUN GEAR

SPUR GEAR SPUR P I N I O N SPIRAL BEVEL GEAR ‘INPUT SHAFT

v

SPIRAL BEVEL P I N I O N FWD Figure 3.- SH-2 main Cransmission.

HOUSING MODEL O R TEST DATA F I N I T E ELEMENT SHAFT MODELS HOLZER/MY KLESTED BENDING/TORSION SYSTEM MODEL SOUND HOUSING MOTIONS -POWER FULLY COUPLED LEVEL MESH EXCITATIONS REDUCED S I Z E MTI (GEARO)

’ \

S P I R A L BEVEL AND HELICAL APPROX.

BEARING MODELS

\ INTERFACE MOTIONS

B. JONES ORTHOGONAL SPRINGS Figure 4 . - Transmission noise modeling approach.

LOW RATE SUSPENSION TEST TRANSMISSION CASE ACCELEROMETER MOUNTING IMPEDANCE HEAD BLOCK (44) ACCELEROMETER ( 4 4 ) SHAKER \\\\\\\\\\\\\\\\\\\\\I\\ \ \ \ \ Figure 5.- Transmission case shake-test setup.

REAL INERTANCE -10 -20

- 30

-40 -50 1000 1500 2000 2500 3000

FREQUENCY - HERTZ

Figure 6.- Measured and predicted transmission case inertance.

67 3 MEASURED ----e PREDICTED

2 * o I-

80% SPEED 60% TORQUE I 1 I I I I I I I I I I I 1 3 4 5 11 12 13 14 15 16 17 18 19 20 STATION NUMBER Figure 7.- Measured and p r e d i c t e d case a c c e l e r a t i o n for p l a n e t system e x c i t a t i o n ( 3 4 8 Hz).

PEAK - G

1 3 4 5 11 12 13 14 15 16 17 18 19 20 STATION NUMBER Figure 8.- Measured and p r e d i c t e d case a c c e l e r a t i o n f o r spur g e a r e x c i t a t i o n (2994 Hz).

6 74 120- MEASURED

4 PREDICTED

-

L E V E L - POWER

SOUND WATTS) 10-l2 d B (re

-

348 696 1198 2396 2448

FREQUENCY - HERTZ

80 per- Figure 9 . - Measured and predicted sound power levels.

cent RPM, 80 percent torqug.

SPUR GEAR

MESH - SECOND

CHANGE I N SOUND POWER

LEVEL - d B

-10 S P I R A L BEVEL MESH

- 20

I I I I I I I I I I 1 348 435 696 870 1198 1497 2396 2448 2994 3060

EXCITATION/RESPONSE FREQUENCY - HERTZ

Figure 10.- Effect of increased input shaft s t i f f n e s s .

CHANGE I N SOUND POWER

LEVEL - dB

-10

- 20

I I I I I I I I I I 1 34% 435 696 870 119% 1497 2396 244% 2994 3060

EXCITAT ION/RES PONSE FREQUENCY - HERTZ

F i g u r e 11.- E f f e c t of reduced p l a n e t carrier r a d i a l s t i f f n e s s .

% CHANGE I N SOUND C POWER LEVEL - d B -1 c I I I i I I I I I I I

- 2c

1198 1497 2396 2448 2994 3060

EXCITATION/RESPONSE FREQUENCY - HERTZ

F i g u r e 12.- E f f e c t of case damping.

INPUT TORQUE - POUND-INCHES

20 000 50 000 100 000 200 000 40C 00 I I

- SPUR GEAR

-- - - - SPIRAL BEVEL GEAR

---

PLANET SYSTEM A 0’ Y O -10 -20 2000 5000 10 000 20 000 40 000

INPUT TORQUE - NEWTON-METERS

Figure 13.- Simplified transmission noise prediction model for fundamental mesh frequency.

- THIRD-OCTAVE BAND LEVEL I N d B (re 0.0002 M I CRO BATJ) I I I I

60 ’

100 1000 10 000

- - - - -

CALCULATED SPECTRUM MEASURED DATA, SPREAD OF THREE MEASUREMENTS 0 GEAR CLASH FREQUENCY Figure 14.- Measured and calculated UH-1 (B-205) internal noise using simplified transmission noise prediction model.

THE INFLUENCE OF THE NOISE ENVIRONMENT ON CREW COMMUNICATIONS John W. Leverton Westland Helicopters Limited SUMMARY A general review is presented on the influence of the noise environment on c r e w communications i n helicopters. The signal-to-noise (S/N) r a t i o a t the microphone and the e f f e c t of t h e attenuation provided by the helmet i s dis- cussed. This shows t h a t the most important aspect is t h e S/N r a t i o a t the microphone, p a r t i c u l a r l y when helmets with improved attenuation c h a r a c t e r i s t i c s are considered. Evidence i s presented which shows t h a t i n high noise environ- ments, the system S/N r a t i o is w e l l below that required and hence there is an urgent need t o reduce t h e cabin noise l e v e l s and improve the microphone rejec- t i o n properties.

I n t h i s paper t h e emphasis is placed on environmental/acoustic considera- tions and no reference is made t o t h e e l e c t r i c a l aspects such as d i s t o r t i o n e f f e c t s o r s i g n a l "clipping" INTRODUCTION The noise l e v e l s inside many helicopters are s u f f i c i e n t l y high t o give rise t o severe communication problems as w e l l as causing c r e w f a t i g u e and general annoyance. The noise l e v e l s at t h e ear are e s s e n t i a l l y a function of the real- a t i v e l e v e l s of the speech a t the microphone, the l e v e l s of the cockpit/cabin noise and the amount of attenuation provided by t h e helmet. This r e s u l t s i n a poor s i g n a l (speech) t o noise r a t i o (S/N r a t i o ) which cannot be improved by the communications system.

Data are presented i n t h i s paper t o i l l u s t r a t e these aspects and although t h e values r e f e r s p e c i f i c a l l y t o h e l i c c p t e r s , the general trends and i m p l i - (fixed wing) case.

cations are equally applicable t o t h e m i l i t a r y a i r c r a f t Care must b e taken, however, i n comparing t h e results, because of t h e higher speech l e v e l s and higher helmet attenuation values, r e l a t i v e t o those f o r t h e helmetlboom (or throat) microphone combinations used i n helicopters, associated with the i n t e g r a l helmet/mask worn by a i r c r a f t c r e w .

SIGNAL-TO-NOISE AT MICROPHONE The range of noise l e v e l s e x i s t i n g i n t y p i c a l helicopter cockpits a r e indicated i n f i g u r e 1 together with the corresponding long term rms speech level.

The speech levels quoted are those measured 1 cm from t h e l i p s and hence are a p p r o p r i a t e t o t h e level experienced by a 'boom microphone' of t h e type commonly used i n h e l i c o p t e r s . Also i n d i c a t e d on t h e f i g u r e are t h e levels Fo appropriate t o t h e Sea King h e l i c o p t e r and a pre-production Lynx h e l i c o p t e r .

convenience, t h e upper l i m i t of t h e band of n o i s e levels which r e p r e s e n t s t h e m a x i m u m l e v e l s measured i n c u r r e n t h e l i c o p t e r s - has been termed "noisy" cock- p i t and t h e lower boundary as ''qu5et'' cockpit. I n p r a c t i c e , of course, a h e l i - copter spectrum is 'peaky' i n n a t u r e and even i f some of t h e octave band levels are n e a r o r on t h e upper l i m i t shown, o t h e r octave band levels could b e n e a r the lower l i m i t indicated.

The 'boom' microphone used i n h e l i c o p t e r s has n o i s e c a n c e l l a t i o n p r o p e r t i e s which reject c e r t a i n regions of t h e ambient n o i s e relative t o t h e speech. The appropriate c o r r e c t i o n s f o r a t y p i c a l boom microphone have been applied t o t h e are shown i n f i g u r e 2. This shows t h a t t h e S / N h e l i c o p t e r d a t a and t h e r e s u l t s r a t i o i n t h e important 1 kHz/2 kHz regions i s extremely poor on t h e "noisy" h e l i copter. The h e l i c o p t e r spectrum, however, contains d i s c r e t e frequency compon- e n t s and even i f t h e cockpit is r e l a t i v e l y q u i e t , one o r t w o octave bands w i l l be of a high level - t h i s is i l l u s t r a t e d f o r t h e case of t h e Lynx and Sea King on f i g u r e 1.

Thus a t one o r more octave bands t h e signal-to-noise r a t i o on most h e l i c o p t e r s is very l i k e l y t o be near t h e value shown f o r t h e "noisy" cockpit.

Throat microphones are o f t e n used i n UK h e l i c o p t e r s i n place of 'boom' microphones. These have b e t t e r noise-cancelling p r o p e r t i e s b u t t h e speech i s g e n e r a l l y of i n f e r i o r q u a l i t y . Precise f i g u r e s are n o t r e a d i l y a v a i l a b l e b u t it has be-en suggested ( r e f . 1 ) t h a t t h i s reduction i n voice q u a l i t y is o f f s e t by t h e improved n o i s e r e j e c t i o n . Thus, it seems reasonable t o apply t h e boom microphone c o r r e c t i o n s and assume t h a t they are equally a p p l i c a b l e t o t h e use of t h r o a t microphones; t h i s approach has been adopted i n t h e b r i e f review pre- sented i n t h i s paper. Some measurements have been made under operating and laboratory conditions and these suggest t h a t above 1 / 2 k;Hz t h e n o i s e r e j e c t i o n p r o p e r t i e s are considerably enhanced by use of a t h r o a t microphone. This apparent advantage appears, however, t o b e o f f s e t by t h e lower speech l e v e l s and thus, f o r a l l p r a c t i c a l purposes, t h e boom and t h r o a t microphones can b e assumed t o give similar r e s u l t s .

HELMET ATTENUATION The a t t e n u a t i o n values f o r helmets normally quoted by manufacturers and r e f e r r e d t o i n t h e g e n e r a l l i t e r a t u r e are based on r e s u l t s from ' R e a l E a r at Threshold' (REAT) tests. These tend t o g i v e an o v e r o p t i m i s t i c impression of t h e n o i s e p r o t e c t i o n provided by a helmet and from a p r a c t i c a l p o i n t of view, i t is t h e a c t u a l Transmission Loss (TL) r e s u l t s which g i v e a t r u e i n d i c a t i o n of t h e a t t e n u a t l o n p r o p e r t i e s of a helmet i n a real environment. I n t h e REAT method, t h e a t t e n u a t i o n f i g u r e s are obtained from t h e d i f f e r e n c e of hearing threshold measurements with and without t h e helmet ( r e f s . 2 and 3 ) , while t h e TL v a l u e s are obtained by determining t h e d i f f e r e n c e between t h e n o i s e l e v e l at t h e ear (measured by a microphone i n s e r t e d i n s i d e t h e ear muff) and t h e noise e x t e r n a l t o t h e helmet. I n t h i s l a t t e r test it is usus1 t o make t h e measurements w i t h i n a real h e l i c o p t e r o r i n a test chamber i n which the h e l i c o p t e r environment i s simulated. It is w e l l known t h a t t h e s e methods give very d i f f e r e n t r e s u l t s , although t h e r e appears t o be a g e n e r a l confusion i n t k e use of t h e two forms A t y p i c a l set of r e s u l t s for t h e SPH-4 helmet, which is manufact- of r e s u l t s , ured by t h e Gentex Corporation of Carbondale, Pennsylvania, and used e x t e n s i v e l y i n theU.S.Army,is shown i n f i g u r e 3. The R E A T r e s u l t s are those quoted f o r t h e helmet by t h e U.S. A m y l and t h e TL values have been obtained by West- A s can be seen t h e real attenuz.tion o r transmiss- land Helicopters Ltd, (WHZ).

i o n l o s s v a l u e s are considerably lower than those.obtained by using t h e REAT method. The d i f f e r e n c e s at t h e l o w frequencies (250 Hz and below) are of t h e order expected and of p a r t i c u l a r importance i n t h e case of a h e l i c o p t e r because of t h e high l e v e l s of low frequency n o i s e present i n t h e cockpit/cabin. The d i f f e r e n c e between t h e two methods i n t h e low/mid frequency (500 Hz) range are l a r g e r than a n t i c i p a t e d and those which occur at high frequency ( 4 kHz) were n o t expected. It w i l l a l s o be noted ths.t i n t h e midlhigh frequency range (2 kJ3z) t h e two methods g i v e , f o r a l l p r a c t i c a l purposes, i d e n t i c a l r e s u l t s .

It has a l s o been found t h a t s l i g h t l y d i f f e r e n t r e s u l t s are obtained w i t h d i f f e r e n t types of n o i s e sources; t h i s is, however, of secondary importance when compared t o t h e v a r i a t i o n from 'rest-to-test'.

It is a l s o clear from t h e r e s u l t s presented i n Figure 3, and o t h e r r e s u l t s , t h a t i f REAT a t t e n u a t i o n values are used t o e v a l u a t e t h e p r o t e c t i o n o f f e r e d t o I n a p i l o t l c r e w member by a helmet, then misleading r e s u l t s can be obtained.

t h e a u t h o r ' s experience, it is n o t p o s s i b l e t o c a l c u l a t e t h e d i f f e r e n c e between TL and REAT test r e s u l t s and thus a t r u e evaluation can only be made i f TL tests are conducted.

It a l s o follows from such a n a l y s i s t h a t many of t h e claims made r e c e n t l y about t h e dramatic i n c r e a s e i n p r o t e c t i o n prcvided by t h e new generation of helmets are inc.orrect 'since t h e comparisons have i n t h e main been made between t h e known TL v a l u e s f o r t h e e x i s t i n g helmets and REAT r e s u l t s f o r t h e new hel- m e t s . This i s i l l u s t r a t e d on Figure 3 which shows t h e TL values f o r the Mk.3 1).

helmet t r a d i t i o n a l l y used by t h e UK h e l i c o p t e r p i l o t s / c r e w menbers ( r e f .

A s can b e seen, although t h e new helmet o f f e r s considerable improvement, p a r t i - are f a r c u l a r l y a t t h e higher frequencies, t h e gain a t t h e lowlmid frequencies less than those suggested by i n c o r r e c t l y comparing t h e REAT r e s u l t s f o r t h e SPH-4 and t h e TL v a l u e s f o r t h e Mk.3 helmet.

A new helmet, t h e Mk.4, i s c u r r e n t l y being introduced i n t o s e r v i c e i n t h e This helmet has according t o a preliminary evaluation s i m i l a r o r s l i g h t l y UK.

s u p e r i o r a t t e n u a t i o n c h a r a c t e r i s t i c s t o t h e SPH-4. Thus, t h e observations made helmet, on which a f a i r l y d e t a i l e d in- i n t h i s paper i n r e l a t i o n t o t h e SPH-4 v e s t i g a t i o n has been conducted, are, i n general, equally a p p l i c a b l e t o t h e Mk.4 helmet.

I Comunication from Department of Army, U.S. Aeromedical Research Laboratory, Fort Rucker, Alabama, Oct. 1974.

INFLUENCE O F HELMET ATTENUATION The i n f l u e n c e of t h e helmet a t t e n u a t i o n on t h e S/N at t h e ear (with intercom o f f ) can be assessed from t h e TL d a t a , Consider f i r s t l y t h e standard Mk.3 helmet which is used by h e l i c o p t e r crews i n t h e UK Forces. This-provides a t t e n u a t i o n which i n c r e a s e s from p r a c t i c a l l y zero at low frequency (125 Hz) t o over 30 dB a t 4 kHz. These a t t e n u a t i o n values have been applfed t o t h e d a t a t o g i v e t h e corresponding levels i n s i d e t h e helmet and t h e s e are i l l u s t r a t e d i n f i g u r e 4 .

It is g e n e r a l l y accepted t h a t t h e long term speech o v e r a l l rms l e v e l should n o t exceed 105 dB s i n c e , at l e v e l s above t h i s , i n t e l l i g i b i l i t y is decreased.

However, i f hearing damage is taken i n t o account, a lower level would seem appropriate. This is, however, a complex s u b j e c t s i n c e f a c t o r s such as exposure rest periods must b e taken i n t o account. Within t h e UK d u r a t i o n , frequency, and t h e g e n e r a l consensus i s t h a t an a r p r o p r i a t e acceptable l e v e l would be 90 dB (A) This is i n l i n e with t h e g e n e r a l approach being adopted i n a number of f i e l d s (including t h e p r o t e c t i o n of t h e i n d u s t r i a l worker). It is d i f f i c u l t at the present time t o f i n a l i z e t h e most d e s i r a b l e l i m i t and f o r t h i s reason both "speech a t t h e ear" criteria have been added t o f i g u r e 4. Considering f i r s t l y t h e "105 dB l i m i t " then it w i l l be observed t h a t t h e S/N on a noisy h e l i c o p t e r is r e l a t i v e l y poor, If t h e "90 dB(A) values'' are assumed to apply, then even rise t o a problem i n t h e two lower octave bands t h e q u i e t h e l i c o p t e r gives considered. I f helmets w i t h improved a t t e n u a t i o n p r o p e r t i e s are used, then t h e o v e r a l l p o s i t i o n is improved. Figure 5 shows t h e r e s u l t s , corresponding t o those presented i n f i g u r e 4 , which would be applicable i f a SPH-4 helmet w a s used. There is t y p i c a l l y a 7 dB improvement ( r e l a t i v e t o t h e Mk.3 helmet) i n a t t e n u a t i o n over t h e complete frequency range (including t h e low frequency end) and thus t h e e f f e c t i v e S/N r a t i o s are considerably increased.

RATIO SYSTEM SIGNAL-TO-NOISE Speech s i g n a l s cover a dynamic range of 30f40 dB with t h e peaks being typ- For speech t o b e completely i n t e l l - i c a l l y 12 dB above t h e long term rms value.

i g i b l e , it is g e n e r a l l y accepted t h a t t h e r a t i o of t h e long term m s t o long term rms "noise" l e v e l a t t h e ear should be a t least 20 dB. Thus, t h e system should be capable of handling peak levels 32 dB above t h e b a s i c n o i s e level.

According t o reference 1, sentences used by aircrew can generally b e understood from t h e i r context, providing t h e ear is n o t overloaded; a long,term S/N r a t i o 9 dB is j u s t considered acceptable. A review within W H L has suggested, hm- of ever, t h a t with a more f l e x i b l e vocabulary, a S/N r a t i o i n t h e order of 15 dB would b e more appropriate. ~ The communications system e s s e n t i a l l y covers t h e frequency range from 250 t o 3000 Hz and i n d e r i v i n g t h e f i g u r e s quoted above, it is assumed t h a t t h e r e I f such r e d u c t i o r s are no major bandwidth l i m i t a t i o n s on t h e speech t r a n s f e r .

i n bandwfdth occur, then an i n c r e a s e i n t h e signal-to-noise r s t i o is required t o maintain i n t e l l i g i b i l i t y .

From t h e r e s u l t s produced i n f i g u r e s 4 and 5, t h e e f f e c t i v e system signal- to-noise r a t i o s can be derived. These have been determined f o r t h e Mk.3 and SPH-4 helmets, r e s p e c t i v e l y , and f o r t h e "noisy" and "quiet" cockpit conf igura- t i o n s considered. The r e s u l t s are shown i n f i g u r e s 6 ( a ) and 6(b) f o r t h e Mk.3 helmet/quiet h e l i c o p t e r and Mk,3 helmet/noisy he'icopter, r e s p e c t i v e l y . Figure 7 shows t h e Corresponding result f o r t h e SPH-4 helmet, b u t i n t h i s case, t h e I t noisy'' h e l i c o p t e r r e s u l t s only have been shown s i n c e t h e syste.m signal-to- n o i s e r a t i o is l a r g e l y c o n t r c l l e d by t h e microphone c a n c e l l a t i o n p r o p e r t i e s .

The summation e f f e c t of t h e two gndividual n o i s e s i g n a l s a r r i v i n g via t h e m i c -

rophone - and through t h e helmet has been talcen i n t o account av.d t h e shaded area

r e p r e s e n t s t h e system S/N r a t i o . A s can b e seen, t h e ''noisy cockpit/Mk.3 hel- m e t " r e s u l t s i n an unacceptable S/N r a t i o ( f i g u r e 6(b)) and even when t h e im- proved helmet is used ( f i g u r e 7) t h e S/N r a t i o is poor. It w i l l a l s o be observed t h a t t h e S/N r a t i o is n o t uniform across t h e communication band (250 Hz - 3000 Hz). I n a d d i t i o n , t h e h e l i c o p t e r spectrum l a r g e l y c o n s i s t s of d i s c r e t e frequencies and thus masking e f f e c t s and p o s s i b l e d i s t o r t i o n i n t h e system has t o be taken i n t o account. It is clear, however, from t h e s e r e s u l t s t h a t although t h e improved helmet is required, t h e ambient (cabin noise) l e v e l s must be lowered and/or t h e microphone c a n c e l l a t i o n p r o p e r t i e s improved.

DAMAGE RISK CRITERIA I n t h e preceding discussion, t h e problem r e l a t i n g t o Damage Risk has been ignored and t h e assessment w a s simply based on t h e signal-to-noise r a t i o a t t h e microphone and t h e "speech" l e v e l requirement at t h e ear. The d a t a concerning hearing damage are confusing and o f t e n contradictory. It is, however, g e n e r a l l y accepted t h a t f o r an 8 hour/day - 5 days/week exposure, an upper l i m i t of 90 dB(A) is acceptable. The s i t u a t i o n i n t h e case of r a t i n g h e l i c o p t e r n o i s e is f u r t h e r complicated by t h e f a c t t h a t t h e Damage Risk Criteria ccmmonly quoted r e f e r e s s e n t i a l l y only t o broadband noise. The audio spectrum on a h e l i c o p t e r is, however, dominated by a series of d i s c r e t e frequencies a r i s i n g from t h e gearbox.

It i s g e n e r a l l y accepted t h a t an allowance f o r such tones can be made by re- There i s a l s o a general f e e l i n g t h a t ducing t h e allowable l e v e l s by 5 dB(A).

t h e suggested criteria should b e applied t o aircrew even though they a r e not exposed f o r t h e f u l l 40 hours per week. Thus, i t seems reasonable t o assume This l i m i t c r i t e r i a should be applied i n t h e h e l i c o p t e r case.

t h a t t h e 85 dB(A) "at t h e ear" ( i n terms of octave band l e v e l s ) has been superimposed on t h e l e v e l s f o r t h e Mk.3 helmet and SPH-4 helmet, r e s p e c t i v e l y , as shown i n f i g u r e s 8 and 9. For reference, t h e octave band l e v e l s corresponding t o an upper l i m i t of A s can b e seen t h e noisy h e l i c o p t e r exceeds t h e recom- 90 dB(A) are a l s o shown.

mended values i n s e v e r a l octave bands when t h e Mk.3 helmet is used and even t h e q u i e t h e l i c o p t e r l e v e l s are very c l o s e t o t h e 85 dB(A) criteria values i n t h e t h e SPH-4 helmet would improve t h e s i t u a t i o n 125 and 250 Hz octave bands. Use of are below as i l l u s t r a t e d i n f i g u r e 9 and i n t h i s case t h e noisy h e l i c o p t e r values t h e 90 dB(A) l i m i t . Thus, t h e use of t h e SPH-4 helmet ( o r equivalent) would seem e s s e n t i a l .

68 3 HELICOPTER TESTS a modffied Mk.3 helmet, which has a miniature Knowles microphone By using mounted i n t h e e a r p i e c e t o measure t h e l e v e l t n s i d e and a microphone attached a series of measurements have t o measure t h e ambient n o i s e o u t s i d e t h e helmet, been made on a range of pre-production and Yn-service" a l r c r a f t . I n a d d i t i o n t o t h e n o i s e measurements, t h e electrical s i g n a l on t h e "tel l i n e s " t o t h e ear- piece w e r e measured, These tests have given r e s u l t s which confirm t h e general trends o u t l i n e d previously and h i g h l i g h t e d a number of p o i n t s .

I n one case t h e l e v e l s at t h e ear i n s i d e t h e helmet w e r e of t h e same order as t h e ambient levels o u t s i d e t h e helmet. The r e s u l t s obtained are i l l u s t r a t e d i n f i g u r e 10 which show t h a t i n t h e 1 ~ H Z and 2 kHz bands, t h e levels are t o a f i r s t order i d e n t i c a l i n s i d e and o u t s i d e t h e helmet. The aircrew concerned w e r e questioned, b u t could n o t give any s a t i s f a c t o r y explanation why t h e a m p l i f i e r volume c o n t r o l w a s set s o high. Thus, t h e r e is no real explanation f o r t h e s e r e s u l t s and so it would appear t h a t they r e s u l t e d from t h e c r e w attempting t o raise t h e i r speech above t h e level of t h e n o i s e i n t h e communication system and/or t h e annoying high l e v e l i n t h e low frequency (125/250 Hz) octave bands.

This r e s u l t e d i n high l e v e l s i n s i d e t h e helmet without, of course, any real improvement i n speech q u a l i t y .

a r e p e a t I n an attempt t o c l a r i f y t h e p o s i t i o n r e l a t i n g t o t h e s e r e s u l t s , t h i s has t o be c a r r i e d o u t on a d i f f e r e n t test w a s planned b u t unfortunately, h e l i c o p t e r . The same intercom system w a s , however, used and i n t h e s e tests t h e system volume c o n t r o l w a s adjusted t o t h e minimum considered acceptable by t h e crew, This r e s u l t e d i n t h e l e v e l s measured i n s i d e t h e helmet being consider- ably lower, as i l l u s t r a t e d i n f i g u r e 11 and although no s p e c i f i c s u b j e c t i v e tests w e r e performed, t h e c r e w tended t o agree t h a t t h e o v e r a l l communication w a s equally as good - o r r a t h e r e q u a l l y as bad - as on t h e previous tests.

These observations w e r e a l s o confirmed by a s u b j e c t i v e evaluation of t h e rec- ording taken with "speech", The i n c r e a s e i n n o i s e i n t h e 125 Hz octave band is, i n c i d e n t l y , n o t depen- dent on t h e intercom system and appears t o be due t o a resonance w i t h i n t h e Mk.3 helmet. Thus, t h e published a t t e n u a t i o n value at 125 Hz f o r t h e Mk.3 helme used i n d e r i v i n g t h e l e v e l s i n s i d e t h e helmet shown on f i g u r e 3 would appear t o an a t t e n u a t i o n of 1 dB, t h e r e appears to b e a 5 dB be i n e r r o r and r a t h e r than amplification.

It w i l l a l s o be observed on f i g u r e s 10 and 1 1 that t h e nus speech l e v e l s are only a few dB above t h e "noise" on t h e intercom system and only i n t h e 250 Hz and 1 kHz levels can a clear d i f f e r e n c e be seen. The corresponding "tel line'' recordings are i l l u s t r a t e d i n f i g u r e 12 and as can b e seen t h e S/N r a t i o i n t h e 500 Hz t o 2 kHz band is only 6 / 8 dB and hence inadequate f o r good communications.

REVIEW OF TEST RESULTS One-.third octave band a n a l y s i s h a s been performed on a number of conditions recorded i n t h e Lynx. P a r t i c u l a r i n t e r e s t w a s placed on t h e 'high l e v e l ' record- This ing and a t y p i c a l one-third octave band spectrum is shown i n f i g u r e 13.

shows t h e l e v e l s with t h e intercom disconnected (noise via helmet), l e v e l s when t h e intercom is switched on and t h e l e v e l s which occur during speech. The speech levels shown are t h e r e s u l t s of conventional rms "slow" a n a l y s i s and thus n e i t h e r A b r i e f review, however, r e p r e s e n t , t h e t r u e "peak" o r t h e long term rms value.

suggests t h a t t h e corresponding long term rms values are i n . t h e order of 6 dB below t h e maximum levels shown t h i s should be taken i n t o account when comparing The r e s u l t s i n f i g - t h e r e s u l t s w i t h t h e Tdealized values discussed previously, u r e 1 3 show c l e a r l y t h e impact of t h e combination of t h e high cockpit l e v e l s It w i l l b e noted and t h e poor t h r o a t microphone c a n c e l l a t i o n p r o p e r t i e s .

t h a t t h e l a r g e s t S/N r a t i o occurs i n t h e 250/800 Hz region. I f a lower system gain (amplification) is used, then t h e complete spectrum ( i . e . speech and noise) w i l l be lowered. I n t h e region above 800 Hz, t h e S/N r a t i o is l a r g e l y a function of t h e microphone p r o p e r t i e s and t h e speech-noise S/N r a t i o i n t h e ear piece w i l l remah f o r a l l p r a c t i c a l purposes unaltered. Between 200 Hz and 800 Hz, t h e level a t t h e ear is a function of t h e n o i s e transmitted through t h e h e l - as t h e g a i n of t h e system is decreased, t h e e f f e c t i v e S/N r a t i o m e t and hence, at t h e ear w i l l a l s o decrease. Thus, t h e o v e r a l l system S/N r a t i o w i l l decrease and t h e i n t e l l i g i b i l i t y degraded. It follows from t h i s t h a t a s u b j e c t i v e assess- ment of t h e a c o u s t i c a c c e p t a b i l i t y o r otherwise, which i s o f t e n used i n r a t i n g t h e cockpit-cabin n o i s e environments, can be very misleading s i n c e t h e apparent absolute level at t h e ear is simply a function of t h e gain s e t t i n g of t h e It is a l s o apparent by a comparison of t h e 1/1 octave communication system.

band d a t a i n f i g u r e 10 and t h e one-third octave band d a t a i n f i g u r e 13 - t h a t a d e t a i l e d evaluation cannot be r e a d i l y made from t h e conventional octave band a n a l y s i s .

t h e i n t e l l i g i b i l i t y i s f u r t h e r influenced by t h e I n a d d i t i o n t o t h e above, masking e f f e c t of t h e tones, ar?d t h e nonuniform e a r p i e c e c a v i t y response.

Masking e f f e c t s are d i f f i c u l t t o quantify, p a r t i c u l a r l y i n t h e case of helicop- ters where t h e l e v e l s are varying with t i m e by 10 dB and, i n some cases, 15 dB.

octave bands are used f o r a s s e s s i n g cabin n o i s e l e v e l s b u t l i m i t e d Currently, evidence suggests t h a t even i f allowances are made according t o a v a i l a b l e methods f o r d i s c r e t e frequencies, t h e s e methods of r a t i n g t h e n o i s e underestimate t h e I n a simple test conducted annoyance and influence on i n t e l l i g i b i l i t y .

using Lynx d a t a , it w a s ,found t h a t when t h e noise levels i n tl-e 1 kHz and 2 kHz octave bands were decreased by 10 dB from the l e v e l s i n d i c a t e d i n f i g u r e 10 it had no e f f e c t on t h e apprent c l a r i t y of speech o r t h e s u b j e c t i v e impression.

Preliminary evaluation suggested t h a t t h e signal-to-noise r a t i o i n t h e indi- v i d u a l bands, o r i n o t h e r words t h e d i s c r e t e frequency-to-broadband l e v e l s , had a marked e f f e c t on t h e s u b j e c t i v e impression and can influence t h e i n t e l l i g i - b i l i t y . It is a l s o apparent t h a t t h e nonlinear response of t h e ear c a v i t y -

of & 10 dB - t h e

measurements made by WHL suggest v a r i a t i o n s ( d i p s and peaks) spectrum i s far from ' f l a t ' as i l l u s t r a t e d i n f f g u r e 14.

CONCLUDING R E m S The S/N ratio at the e a r is controlled by the cancellation properties of When improved helmets are used, the the mkcrophone and t h e helmet attenuation, system S/N r a t i o w 5 1 l become more dependent on t h e microphone r e j e c t i o n pro- p e r t i e s i n most of t h e helicopters and w i l l remove the problems associated with hearing damage a r i s i n g from high levels at t h e ear.

It follows t h a t e i t h e r the noise levels i n t h e cockpit have t o be lowered o r a l t e r n a t i v e l y t h e noise attenuation properties of t h e microphone improved.

I n t h i s context it i s ' o f i n t e r e s t t o note t h a t the noise r e j e c t i o n characteris-

t i c s of t h e boom microphone - and, by implication of t h e WHL tests, t h e effec-

t3ve r e j e c t t o n of t h e throat microphone - decrease with frequency and approach

zero at 4 kHz. The mask/mask microphone provides, on t h e other hand, an eff- ective "shield" whose r e j e c t i o n increases at 1 kHz and above. Unfortunately, a t 1 kHz t h e value is only 5 dB b u t some general communication noise exclusion microphones provide even b e t t e r noise r e j e c t i o n with t h e values reaching t y p i c a l 20 dB a t 1 M z . Thus, it would seem desirable t o attempt t o incorporate the ad- A l t e r - vantages of both systems t o provide a wide frequency range rejection.

concepts of placing t h e microphone i n s i d e t h e helmet would seem w e l l n a t i v e l y , worth while, p a r t i c u l a r l y when helmets with bproved high attenuation at low frequencies a r e developed. Reduction of the noise at the source must, of course, be pursued with equal vigour but t h e r e i s a l i m i t , p a r t i c u l a r l y i n the Treatments can be readily applied t o the cabin area and although cockpit area.

these i n turn produce same reduction i n the cockpit area, it is unlikely t h a t can be made before r a d i c a l new fuselage design concepts current1 s i g n i f i c a n t gains being considered can be employed.

With the improved helmets, it i s worth considering placing more emphasis on the microphone r e j e c t i o n of t h e noise, since, if t h i s could be achieved, then higher ambient noise l e v e l s could be t o l e r a t e d without infringing Damage Risk Criteria. This solution could be applied t o a l l f o m of a i r c r a f t / h e l i c o p t e r s , whilst noise reduction techniques w i l l , i n general, have t o be r e l a t e d t o speci- f i c designs. The o v e r a l l cost of developing an acceptable microphone system i n the long term would, therefore, be most l i k e l y t o be less than t h e cost of in- dividual noise control schemes. Even s o , it does appear t h a t attempts at obtain- ing improvements i n both aspects must be considered i f the communications pro- blems are t o be overcome.

ReFERENCES RAE Technical 1. On t h e Specification of Maximum Noise Levels i n Aircraft.

Report 72089, June 1972.

2 , Method f o r t h e Measurement of t h e R e a l E a r Attenuation of E a r Protectors at Threshold. S3.19, American National Standards I n s t i t u t e , 1974.

3. Method of Neasurement of the Attenuation of Hearing Protectors a t Threshold.

BS5108, B r i t i s h Standards I n s t i t u t e , London, 1974.

e r E \ 4 ! J 110- a VI 60- LEVEL OF SPEECH 1 CM. FROM LIPS r compared with Figure 1.- Helicopter i n t e r n a l noise l e v e l s speech l e v e l s .

120 -

110 .

100 -

90 -

ao .

70 -

'.

.----ANTERNAL - LEVELS" \

'QUIET' COCKPIT 60 -

'BOOM MIC. CANCELLATION I I I I I I I 31.5 63 125 250 500 1000 2000 4000 8 0 0 0 18000 OCTAVE BAND CENTRE FREQUENCY (Hz) Figure 2.- E a r piece noise l e v e l s (boom microphone cancellation corrected) compared with speech l e v e l s .

CY---a 5 0

LOSS RESULTS 7-u

/ / X 4 M U . 3 HELMET- RAE: SPbC.

/ / 40- 'p v 2 30- Q

$ 20-

k

4 10- I I I I 1 I I I I 31.5 6 3 125 250 500 1000 2 0 0 0 4000 8000 1 6 0 0 0 OCTAVE BAND CENTRE FREOUENCY IHz) Figure 4.- Noise l e v e l s at ear - Mk.3 helmet.

100- 90- 80-.

\ 70- ' \ \ \ /OlSY' COCKPIT

M A X ALLOWABLE LEVEL '

r-----X OF SPEECH AT EAR \ \ '.

60- ' .

\ \ SAFE LEVEL AT EAR \ \ \

- - RAE PROPOSAL \

-a )--d NOISE LEVEL 3c a INSIDE HELMET UI \ 40- 'QUE$ COCKPIT , .

t 1 1 I I I I 1 I OCTAVE BAND CENTRE FREQUENCY (Hz) Figure 5.- Noise levels a t ear - SPH-4 helmet.

/ / _MICROPHONE CANCELLATION (a) k \ \ \ \ \ \ y QUIET COCKPIT M k . 3 HELMET ATTENUATION SYSTEM S/N RATIO / / MICROPHONE CANCELLATION M k . 3 HELMET NOISY COCKPIT ATTENUATION Figure 6 . - Signal t o n o i s e r a t i o s based on a speech l e v e l of 105 dB

-

helmet "quiet" and 'hoisy" c o c k p i t s .

*---x AlTENUATlON OF SPH-4 HELMET e - - . BOOM MICROPHONE p-. .

30 CANCELLATION EFFECT w / / /-

m T SYSTEM SIGNAVNOISE RATIO

/ /

-10 -

Figure 7.- Signal t o n o i s e r a t i o s based on a speech l e v e l of 105 dB

- SPH-4 helmet "noisy" cockpit

110 - "NOISY" COCKPIT

100 -

90- 90 dB(Aj 80- -0 85 dB(A) \ \ \ /'\ \ "QUIET" COCK PIT \ ' e . \ 70 - \e " \ 'e \ \ \ \ \ -0 DAMAGE \ 60- 'e \ \ -0 RISK CRITERIA \ e I I I I 1 I I I I 31.5 63 125 250 500 1000 2 0 0 0 4000 8000 1 6 0 0 0 Figure 8 . - Levels a t ear compared with damage r i s k c r i t e r i a - Mk.3 helmet.

100 1 \ ,/OI~Y' COCKPI \ \ \ \ \

60 -

@\ \ \ \\ \ \ -a \ 1 .

50 -

. \ 0 DAMAGE \ ' e RISK CRITERIA \

40 -

\ \ I I I I I I I I 1 Figure 9.- Levels a t ear c o m p a r e d w i t h d a m a g e r i s k criteria - SPH-4 h e l m e t .

I/C ON + SPEECH

100 -I

R l ABWJ AMBIENT X - X COCKPIT' AMBIENT HELMET NOISE I/C ON &---A HELMET NOISE I/C OFF cI-.--..-, HELMET NOISE OCTAVE BAND GEN h e l m e t .

N - E

ii 0

rn' A W P ai -1 W

e

-1 W P a VI P Q n 1c a VI I I I I I I I I I 1 31.5 63 125 250 500 1000 2 0 0 0 4000 8000 16000 OCTAVE BAND CENTRE FREQUENCY (Hz)

Figure 11.- Lynx intercom noise - levels at ear - effect of

reducing intercom system gain setting - Mk.3 helmet.

-10 1 \

- 5 0 i

\ \ \ \ \ \bee0**4N-- 1/C DISCONNECTED '\ / - - A 'N ,-do- I/C DISCONNECTED '5-9 31.5 03 126 250 5 0 0 1000 2 0 0 0 4000 8000 1 6 0 0 0 OCTAVE B A N 0 CENTRE FREQUEIYICY (Hz) Figure 12.- Lynx intercom noise - electrical signal on headset input.

A E

I

N W a

g

-1 W 2; 80 -I W a a VI VI W a IL.

n iL a u)

b

50-i

O N E THIRD OCTAVE BAND CENTRE FREQUENCE (Hz)

Figure 13.- Lynx intercom noise - comparison of speech and

background noise l e v e l s .

CAVITY RESPONSE 1 I 0 i 2 3 4 FREQUENCY (Hz) Figure 14.- Frequency response of Mk.3 helmet earpiece.

HELICOPTER INTERNAL NOISE REDUCTION RESEARCH AND DEVELOPMENT APPLICATION TO THE SA 360 AND SA 365 DAUPHIN H. J. Marze and F. d'Ambra S.N.I. Aerospatiale INTRODUCTION With the extension of the civil commercial market, the noise inside telicopter cabins is becoming one of the foremost problems for the comfort of iassengers.

As shown in the statistical study of figure 1, the helicopter is the most ioisy vehicle in comparison with other ground or flying vehicles.

QUALIFICATION AND IDENTIFICATION OF ANNOYANCE We have been concerned for several years,by the internal noise problem, tnd, as a first step, the sources of noise inside cabins have been investigated.

IS shown in figure 2, there are many possible sources os noise (main rotor, :ail rotor, engine(s), gear boxes, or accessories). As each of them has peculiar icoustic noise characteristics (rotational noise, principally), they can be .dentified by narrow-band analysis of the sound signal.

Analysis of noise recordings taken inside bare cabin helicopters in flight has shown that the most significant source of noise is the main gear )ox (fig. 3) which gives rise to a large number of pure tones emerging 1 0 to 3 0 dB from the broad-band spectrum.

Knowing the source of noise, we have quantified the annoyance by applying :he present conventional units (fig. 4 ) : 1) A and D weighted dB 2) Speech interference levels by evaluating the mean level in the voice frequencies 3 ) Perceived noise level from Noys curves and correction of this level for spectral irregularities Very soon, we saw from the passengers evaluation that the usual units zould be inconsistent with the subjective answers collected and that they qere accordingly only imperfectly representative of the annoyance actually Eelt in helicopter cabins. 5.) Such disagreement is bound to the (See fig.

ixistence of pure tones within the 500 to 5000 Hz range; these frequencies ixceed the 20 dB broad-band noise and even sometimes can exceed 30 dB.

In order to quantify the additional annoyance, we have carried out psychoacoustic studies dn the effect of the pure tone emergence. (See fig. 6 ) .

Wide band type noises actually measured on the helicopter were submitted to a large number of juries for comparison with the same noise to which a pure frequency of 1000, 2000 Hz or 1000 and 2000 Hz had been electrically added.

The results of this study have demonstrated that the conventional units,dBA, dBD, PNdB, and even TPNdB, underrated the effect of the emergence of pure tones within the 3 to 9 dB range for dBD and PNdB and within the 1 to 5 dB range for dBA and TPNdB (emergence of 20 dB and 30 dB in narrow-band analysis).

This nonexhaustive study of the emergence effect has shown that the significant parameters of the internal noise of helicopters are the pure tones at the meshing frequencies of the main gear box.

MESHING NOISE REDUCTION STUDY OF MAIN GEAR BOX The purpose of the study undertaken is the acquisition of knowledge on (fig. 7 ) :

( a ) mechanisms of vibration generation inside the main gear box - study

of meshing ( b ) mechanisms of transmission between source and cabin, through the study of dynamic behaviour of main gear box components (pin'ions, casings, etc.)

or those linking the main gear box to the cabin structure Noise Sources Gear meshing is a noise generator due to its design and realization; angular meshing errors are generating vibrations which will excite the structure.(See fig. 8 ) . Until the last few years, the compromise made at the design stage between the various gear parameters had for its main objective a minimum weight while ensuring a satisfactory service life. For that purpose, gear toothing was designed to work as closely as possible to the maximum permissible stresses and specific pressures but also to limit axial, radial, and tangential loads on bearings.

This choice is the contrary of the continuous meshing concept; as the tooth bending increases with the load, the low driving and overlap ratios achieved with low spiral angles and diametral pitch or high pressure angle induce sudden load variations during meshing and do not ensure the compensation of machining errors which would require the simultaneous meshing of several teeth.

In a first stage, we have measured the angular meshing error on a pair of pinions under no load by using a "GOULDER MIKRON" type checking machine 9 ) . The results, recorded on paper in analog form, clearly show the (fig.

existence of tooth profile errors superimposed on an offset error or distortion of the basic circle. The spectral analysis of these analog signals allows the separation of these phenomena and the quantification of the effect of idditional parameters such as backlash.

An example of toothing geometry modification is as follows. On the SA 3 6 5 main gear box, the input spiral bevel gear toothing has been redesigned, laking the acoustic aspect into account; the tooth bearing pattern has been Iptimized to ensure a better meshing continuity. The gain achieved over she original meshing is approximately 15 dB.

Dynamic Behaviour of Detail Parts If a meshing concept taking the acoustic aspect into.account is a iecessary condition to achieve a low noise level, it is not sufficient. In iact, in the transfer of vibration energy to the structure, the dynamic )ehaviour of each of the components constituting the transfer path (pinions, shafts, bearings, casing, and main gear box attachment fittings) has to be :onsidered.

Axisymmetric Part Modes (Pinions, Shafts) In a first stage, an experimental and theoretical mode determination i a s been made for the parts constituting the geat train.

Refer to figure 10 for the.results of a mode determination made by ising a laser holography method and a finite-element mathematical model ) n a SA.365 spiral bevel and planet gear assembly. The mathematical model tstablished allows the determination of the axisymmetric part modes under .oad and in rotation. The agreement between modes calculated and those ieasured in the laboratory using laser holography is excellent up to 7 to 8 kHz.

The search for agreement between the SA 365 main gear box natural and txcitation frequencies (fig. 11) shows that it is difficult to design a :omplete main gear box in which no component natural frequency would be in iccordance with a meshing frequency. This difficulty of mastering the full :ear train dynamic behaviour has been checked on an actual SA 360 main gear l o x in which the spiral bevel ring gear rigidity had been modified.

Figure 12 shows the changes in noise levels, measured on the acceptance .est bench, for one of the spiral bevel gear meshing frequencies versus -0tational speed and in two different configurations, initial ring gear and -einforced ring gear. According to the rotational speed, the modification l a y be beneficial or not, and for nonnegligible gains achieved at nominal -.p.m. at this frequency, there were appreciable losses at other meshing 'requencies .

The introduction of some damping in all the gear train seems to be a tseful line to follow in view of reducing the gear train dynamic responses.

calculation model of the forced response for damped axisymmetric parts hould be established to allow the design of such assemblies.

Casing Modes Knowing t h e main gear box casing dynamic behaviour i s a very important f a c t o r ; i n f a c t , - Due t o t h e v i b r a t i o n of i t s w a l l , t h e casing is a source of n o i s e .

- The v i b r a t i o n energy generated a t t h e source and t r a n s m i t t e d t o t h e casing through t h e bearings w i l l reach t h e s t r u c t u r e through t h e casing attachment p o i n t s (main g e a r box suspension b a r s and f l e x i b l e mounting p l a t e ) .

- Casing supports t h e s h a f t s and thus ensures proper p o s i t i o n i n g of

meshing g e a r s , hence t h e r i s k of coupling between t h e e x c i t a t i o n and casing response.

Modal determination i n 1aboratory.- A s f o r axisymmetric p a r t s , modal determination has been made i n t h e l a b o r a t o r y on SA 330-SA 365 main gear box casings using t h e laser holography method. Figure 12 shows two examples of mode determination on t h e SA 360 casing. On t h e prototype casing, it has been noted t h a t a n a t u r a l frequency of 1792 Hz w a s c l o s e t o t h e s p i r a l 1850 Hz. A s t r u c t u r a l change ( s t i f f e n i n g of bevel gear meshing frequency of casing through a r i b l o c a t e d a t midheight) has r e l o c a t e d t h e n a t u r a l frequency from 1792 Hz t o 1850 Hz and generated a new mode a t 1729 Hz. As t h e r e i s a s l i p p a g e of n a t u r a l frequency according t o t h e load (from 1792 Hz t o 1850 Hz) and as t h i s h a s been checked on t h e prototype casing ( f i g . 1 3 ) , t h e modified casing should not have any longer n a t u r a l fr&$uencies i n accordance with t h e s p i r a l bevel r i n g gear meshing frequency. I n f a c t , a gain of some dB's has been noted during t h e bench t e s t i n g of t h i s modified casing.

Forced response of complete main gear box on test bench and on a i r c r a f t . - To check t h e r e s u l t s obtained i n l a b o r a t o r y tests, a bench accelerometric measurement (see f i g . 14 f o r set-up) has shown t h e r e w a s r e a l l y a very l a r g e response of t h e main gear box casing a t 1770 Hz, and t h i s frequency w a s moving towards 1850 Hz when torque w a s g e t t i n g n e a r e r t h e nominal load.

A second check on t h e presence of resonance a t 1850 H z has been obtained i n c r u i s i n g f l i g h t by applying a damping product on t h e main gear box s u r f a c e .

The gain i n n o i s e level has been appreciable i n t h e 2 kHz octave (an a t t e n - u a t i o n of more.than 3 dB f o r 1 kilogram of damping product) although t h e a t t e n u a t i o n w a s n e g l i g i b l e a t t h e o t h e r frequencies.

As f o r t h e gear t r a i n d e t a i l p a r t s , t h e design of casings using materials t h a t o f f e r a l a r g e i n t e r n a l damping seems t o be necessary i n v i e w of l i m i t i n g coupling e f f e c t s between t h e g e a r t r a i n and casing modes, and a l s o t h e amplitude of responses a t t h e e x c i t a t i o n frequencies. This w i l l be t h e sub- ject of f u t u r e research t a s k s .

Main G e a r Box Suspension B a r Dynamic Behaviour The main purpose of t h e main gear box suspension b a r s i s t o ensure t h e ransfer of lift loads to the structure; the attachments on structure and Therefore, ain gear box upper section are made through metal hinge fittings.

he main gear box casing vibratory motions are transmitted to ithout possibility of energy dissipation.

For the SA 360 main gear box bars, the first bending modes, in free-free onfiguration, have been determined in the laboratory (exc tion through B nd K vibrating pot and accelerometric recording). This mo determination n the laboratory has allowed the validation of the mathematical model used o calculate the bending modes and the study of the effect on the bars of he hinges and weight (concentrated or distributed weights).

Figure 15 shows the results of the calculations made on a SA 360 main ear box bar. We can see the correspondence between the third bending mode requency (1850 Hz) and the spiral-bevel gear meshing frequency, together ith the inportant displacement of the resonant frequencies according to he type of weights added to the bars. The efficiency of these weights has een verified in flight as, with 1.3 kilogram of lead distributed on the our bars, the mean noise level dropped by 4.2 dB SIL (Speech Interference evel) .

IMPROVEMENT OF THE INTERNAL NOISE LEVELS BY OPTIMIZING THE CABIN ACOUSTIC TREATMENT Although some improvements have been made in the knowledge of means for oise reduction at the source, these improvements are not sufficient to nsure a satisfactory noise level in the cabin, and a sound-proofing treat- ent isolating the passenger has to be installed and optimized.

The treatments we have optimized associate the three following effects fig. 16): (a) An acoustic screen using the weight effect isolates the passenger rom the noise source. (Item 1 on fig. 16.)

( b ) A damping treatment limits the conversion of the vibratory energy nto acoustic energy. (Item 2 on fig. 16.)

( c ) An absorbing treatment achieved either through HELMHOLTZ resonators r through a glass wool blanket limits the propagation of accoustic waves nd the wave reflection effects in the cabin. (Item 3 of fig. 16.)

Figure 17, a section of the SA 360-365 cabin structure, shows the nstallation of the various elements.

Figures 18 and 19 show the efficiency of the various treatments and heir weight which is to be compared with the maximum weight of aircraft f about 3000 kg. It can be noted that the conventional sound-proofing treatments offer the minimum efficiency from the weight penalty aspect.

Weight Type of koustic Modification penalty action gain 2 kg (approx.

On the Modification of teeth 6.4 dB SIL source geometrical characteristics On the load (1) Treatment of a housing transfer using a damping material 5 dB SIL 1 kg (2) Treatment of attachments 4.2 dB SIL 4.2 kg by means of lead cloth On passenger's (1) Damping of cabin 6.8 dB S I L 26 kg isolation structure (2) Acoustic screen 3.3 dB S I L 50 kg CONCLUSIONS The combined application of treatments at the source in the transfer of vibratory energy level and in the optimization of the sound barriers allowed, at high cost (weight,'price), the obtainment of interesting results as shown by the narrow- and octave-band analysis of noise signals recorded in a SA 365 aircraft in flight, both with a prototype main gear box without cabin sound-proofing and with a modified main gear box and sound-proofed cabin. (See fig. 20 and fig. 21.)

(1850 Hz) which Areduction of about 53 dB was optained on a pure tone (fig. 19).

was at the origin of the main annoyance on the prototype aircraft The overall noise level expressed in dB S I L and dBA has been improved by about 30 dB (fig.21), which ensures a good comfort in this aircraft.

The comparison with MIL specifications shows that it was dangerous to fly without ear protection device in the prototype aircraft and that it is now possible in the treated aircraft to fly for more than 8 hours per day without ear protection.

The comparison with airliner specifications shows that a great deal remains to be done at the mean octave frequencies of 1 kHz, 2 kHz, and 4 kHz.

The possibilities of further improving the conventional acoustic treatments seem to be small.

Only an important echanisms generating a 3me additional gains.

an be obtained in heli assengers, and at a cost (weight, elicopter manufacture U

c

SEAT COMFORT

N O S E

TEMPERATURE

UP AND DOWN MOT\O1\1

I

PRESSURE CHANGES

S \ D E T O S\DE MOTION

WORK SPACE

I

L 1 G T I NG

f I I

SMOKE Figure 1 . - Interior noise considerations for various transportation vehicles.

--- Figure 6.- Ps-ychoacoustic study of the noisiness of pure tone internal noise of helicopters.

Figure 7 . - Gear box noise generation and propagation considerations.

1 Machining

Figure 8.- Gear tooth design factors i n the generation of gear meshing noise.

7 09 m -# W L 0 L Y f f E Y u-3 S a I N I f c- Figure 10.- Comparison between c a l c u l a t e d and measured n a t u r a l frequencies of a s p i r a l bevel pinion s h a f t on t h e SA 365 h e l i c o p t e r main gear box.

P

u

P

r)

f

J c)3 I: c w K TT .

Figure 14.- Effect of torque on the dynamic behaviour of the SA 360 helicopter main gear box housing.

\

I

\ I \

J I

\

$

I

\ a , m .I+ rl (d

E