/J- Jd-o_ __-_---xJ-88

Uniled Slates Environm_nlal Aguncy Noise Protraction Office of Noi_ AIlal_menl _nd Control EPA 5ro0/9-79"103 November 1979 W_$hln01on , DC _04fi0

_EPA

Annoyance, Loudness, and Measurement of Repetitive Type Impulsive Noise Sources

_xy. ,nEPA 55019-79-I03

ArlNOYANCE, LOUDNESS, ArID _IEASUREMENT OF REPETITIVE TYPE IMPULSIVE NOISE SOURCES

NOVEMBER1979

Prepared

by:

L, C. Sutherland R. E. Burke WYLE RESEARCH El Segundo, California

90245

Prepared

for:

U.S. EIIVIRONMENTAL PROTECTION AGENCY Office of Noise Abatement and Control Washington, D.C. 20460

This report has been approved for general availability. The contents of this report reflect the views of the contractor, who is responsible for the facts and the accuracy of the data presented herein, and do not necessarily reflect the official vimls or policyof the EPA. This report does not constitute a standard, specification, or regulation.

PREFACE

The United States Environmental Protection Agency (EPA) was charged by Congress in the Noise Control Act of 1972, as amended by the Quiet Communities Act of IgTB, to conduct or finance research to investigate "...the psychological and physiological effects of noise on humans and the effects of noise on domestic animals, wildlife, and property,and the determination of dose/response relationships suitable for use in de'cislonmaking..." (Section 14(b)(1)). Pursuant to and as part of this mandate, EPA has undertaken investigations to determine and quantify subjective reactions of individuals and communities to different noise environments and sources of noise. specific series of studies has been initiated to determine the bes: methods for evaluating subjective magnitude and aversiveness to noise on the basis of spectral and temporal properties, and to ascertain the importance of and means for including nonacoustical factors in the evaluation of general aversion to noise, The overall purpose of this line of research A

is to derive a more solid basis for assessing the aversiveness of noise and the benefits of noise control. The aim of the investigation described in this report was to perform a detailed analysis of data pertaining to potential annoyance responses that may be attributed to repetitive type impulsive noise. Specifically, a

program was undertaken (1) to review and evaluate the literature on human subjective response to repetitive impulsive noise, and (2) to assess the

need

for and relative order of magnitude of a subjective

impulse

adjustment factor that would better define effective level in terms of annoyance reactions.

The report provides much useful information on the annoyance and loudness of repetitive impulsive noise. Moreover, it is expected that

the results of the investigation will form the basis of future experimental psychoacoustic work to derive, if appropriate, more precise correc-

tions factors or noise prediction methods to effectively account for the inherent annoyance associated with impulsive noise. EPA believes that

further research and evaluation of data on the subjective effects of noise will foster the development of techniques to demonstrate additional

benefits of noise control beyond that exhibited by currently used procedures, series, Fulfillment of this objective awaits further study within this

The results published in this report, however, do provide an of the phenomena of

important step toward a more complete understanding human subjective response to noise.

The conclusions

reached in this report regarding moderate

level impulsive

noise are the authors' and do not necessarily reflect the opinions of the individuals listed above. Moreover, the U. S. Environmental Protection Agency

does not endorse the findings of this investigation factor" applicable

for use as a "correction

to impulsive type noise, nor have similar correction

factors been used by the Agency in past or current noise impact analyses.

OFFICE OF THE SCIENTIFIC ASSISTANT TO THE DEPUTY ASSISTANT ADMINISTRATOR OFFICE OF NOISE ABATEMENT AND CONTROL U. S, ENVIRONMENTAL PROTECTION AGENCY

ii

ABSTP_CT

This study was undertaken to evaluate subiecfive levels oF no_se from impulsive sources. rlsk or annoyance from building

and objective

aspects of maderate af hearing damage

The study excluded evaluation

vibration by high level impulsive noise, which were Research Council, Committee on

covered by recent recommendations of the National Hearing Bioacoust_cs and B_omechanics, Working arlginal investigations _nta some of the objective

Group 69.

While the study included

aspects of impulsive noise, o detaTled Based an thls available impulse noises

review of the llteroture on the subjective aspects was emphasized.

Hterature, the annayance and loudness from a wlde variety of repetitive were evaluated These results were applied

to the evaluat_an of _mpulslve noise from it _sten-

a number of specific noise sources. tofively

Based on the mast pertrnent literature,

cancluded that a subiectlve impulse correction

factor of 7 dB applied to the impulsive noise sources No addlt_onal

A-werghted

equivalent sound levels of these types of repetitive

would better define their effective correction is identlfled

level _n terms of annoyance reactions.

at this time for crest level or repetrfion rote.

Research an sub-

jectlve correction factors for hellcopter

blade slap is also reviewed and potential 0 to 6 dB) for annayance

reasons for the smaller subjective correction factors (i.e., response to this type of sound are discussed. subiectlve

It _s recommended that refinements to this methods

correcfian factor be based an the use of standard loudness calculation mcdffled to include

{Stevens Mark VII or Zwlcker)

provision for o shorter time constant

to reflect subjective response to short duration impulsive sounds. The study also included a brief experimental _mpulsive-type evaluation of the measurement of o signals vorylng in duty cycle, repeti-

wlde variety of simulated repetitive

tion rate, pulse frequency_ and ratio of peak _mpulse signat level to continuous background noise level. When repetitive impulses are measured using maximum values of A-weighted no ob[ectlve correction is necessary in

(slaw) readings on an Impulse Sound Level Meter,

order to measure, with an accuracy of _ 1.5 dB, the equivalent sound level (Leq) of the wide variety of impulsive signals investigated.

iii

ACKNOI#LEDG_ENTS

A program to develop techniques for evaluating no_se sources was inltloted U.S.

the noise from selected impulsive

in December 1975 for the Office of Noise Abatement and Control, Mr. Jeff Goldsteln served as technical monitor.

Envlronmental Protection Agency.

The problems addressed in this study encompc,ss areas in psychoacousHcs and acoustical measurement technology many years. which have received intensive study by many investigators for individuals who, during the conduct of

The authors wish to thank the following

this study, pfovlded much helpful have conducted. Dr. Robert Gales, Dr. William Mr.

information from research which they or their colleagues

Naval Undersea Center, San D_ego, California Bolt, Beranek and Newman, Canaga Park, California Virginia

J. Galloway,

B.W. Lawtan, NASA langley Research Center_ Hampton, Denmark

Dr. O. Juhl Pedersen, Technical Universlty, Dr. D.W. Roblnson, National

Physical Laboratory, England Unlvers_ty, Boston, Massachusetts Engineering Research Laboratory,

Dr. Bertram Scharf, Northeastern Dr. Paul Schomer, U.S. Champaign _ Illlnois Dr. Milton Whitcomb,

Army Construction

Natlonal

Academy of Soleneest CHABA, Washington, San Diego, California.

D.C.

Dr. Robert Youngt The valuable

Naval Undersea Center,

contributions

which were made to Appendix A by Dr. Mark Lee, presently Pasadena, California, and to Appendix B by Mark Montroll,

of the Jet Propulsion Laboratory, are also appreciatively

acknowledged.

iv

TABLEOF CONTENTS Page i.O 2.0 INTRODUCTION .......... SELECTION OF A BASELINEMETRIC 2, 1 2.2 3.0 DeFinition oF impulsiveNoise Baseline Noise Metric ..... I-I 2-I 2-1 2-10 3-I 3-2 3-2 3-4 3-8 3-17 3-21 3-27 3-33 3-34 3-39 3-47 3-50 4-I 4-1 4-I 4-2

SUBJECTIVERESPONSES TO iMPULSIVENOISE 3.1 Loudness or Noisinessof impulsive Sounds . 3.1.1 3.1.2 3.1.3 3.1,4 3.1.5 3.1.6 3.2 A Model Far the Hearing Process Experimental Data. Subjective Response to Impulsive Pure Tone Sounds Subjective Response to BurstsoF Noise Loudness VersusNoisinessoF ImpulsiveNoise . Subjective Response taComplex impulsive Sounds

Annoyance and Other Subjective Response to impulsive Noise. 3.2. I 3.2,2 3.2.3 3,2.4 AnnoyanceResponse to Impulsive Noise Helicopter Blade Slap Noise Loudness Versus Annoyanceof Impulsive Sounds Other Subjective EFfects el: Impulsive Noise

4.0

CONCLUSIONs SUBJECTIVECORRECTION FACTORS FOR EVALUATION OF IMPULSIVE NOISE 4.1 Subjective Correction Factor As 4.1.1 Subjective Correction FactorsBasedon Loudness Response Data for Toneand Noise Bursts 4.1.2 Subjective Correction Factors Basedon Measured Loudness of Real Impulsive Noise SOurces .

TABLEOF CONTENTS (Continued) Po_e 4. I. 3 4.1.4 Subjective Correction FactorsBasedon Annoyance . Summary of MethodsForComputing the Subjective Correction Factor _
S

4-3 4-8 A-1 B-1 C-1

APPENDIX A APPENDIX B APPENDIX C REFERENCES Part A Part B Part C Port D Port E Part F Part G-

OBJECTWEMEASUREMENT OF IMPULSIVE NOISE ISO ROUND ROBIN TESTS FREQUENCYSPECTRA OF REPEATED TONE BURSTS

Annoyanceof Impulsiveor Fluctuating Sounds Noisiness and Loudness oFImpulsiveor Fluctuating Sounds Detection or Perceptionof ImpulsiveSounds Speech Interference Sleep Interference Hearing Damago Moosurernen oF t ImpulsiveSounds

R-I R-5 R-IO R-11 R-12 R-12 R-14

kJ,.

'

Vi

LISTOF TABLES

Table No. ! 2 Typical Physical Parameters of Four Real Sources of Impulsive Noise Index of ExperimentalStudiesan Loudness/Nolsiness of Impulsiveor Fluctuating Sounds Indicating ExperimentalVarlables Investigated Description of Natura fly Occurring ImpulsiveSounds asComparison Signals in Evaluation Experimentby Fidell and Pearsans A Summary of Literature on AnnoyanceResponse to s Impulsive Noise (Excluding Studiesfor Hefleopter Blade Slap) Summaryof RecentStudiesof Helicopter BladeSlap Noise Including Sumrr_ryoESub iective Corraction Factor for Impulsiveness Comparlson of Several PredictedSubjectlve Correctlon Factors for AnnoyanceApplied to the Four Impulsive Noise Sources Sumrnawof Subjective Correction Factor (As)EsHmated from Existing Methodsor Data, dB

Page No. 2-9

3-6 3-29

3-35

3-40

4-7

4-B

vii

LIST OF FIGURES Fi_lure 1 Examplesof Time History Envelopesof Nonimpulsive and Impulsive Sounds Examplesof Time Histories of the Instantaneous'Pressure from Impulsive Sources Physical Parameters of a Typical ImpulsiveSound Conceptual Illustration of Auditory Process to ShowCharacteristic Response Times(I') in VariousElementsWhich Govern the Dynamic Response of the Ear to TransientSounds. Measuredand Normalized Values ForChange in Signal-to-Noise Ratio of Single Tone Burstas a Function of BurstDuration Rangeof MeasuredLouclness Level - Duration TradeoffReportedfrom Various Studies to Indicate Possible Rangeof Uncertainty in Predicted Loudness of 20 ms Pulse Subjective Correction L_ s for R,:peated1000 Hz ToneBursts Pa_Ie 2-2

2-4 2-8

3 4

3-3

B-9

3-12 B-13

7 B

Subjective Correction ,', for RepeatedPure Tone Bursts_ PulseDuration 20 ms, Repetition Rate--s25/Second as a Functionof (a) Reference Intensityt L (Ref) at 1000Hz and (b) Frequencyfor the80 clBReference Level eq 3-14 Time History and FourierSpectrumof a Typical Impulsive Signal SubJectiveCorrection Factor for Loudneso s r NoisinessResponse to Short Bursts of Noise Bands Relative to a Reference Noise Difference Betweenthe Leqof a Repetitive Noise Burst Superimposed on a Steady BackgroundNoise and the Level of a Continuous Noise which Sounds Equally as Loudasa Function of the Ratio of the Burst Level to the Background Level Time-Amplitude Sequence Diagramof the StimulusPresentation Resultsof Experiment] Comparisonof Loudness and NoisinessversusRepetition Rate for a BurstTime Fraction of 0.063 viii 3-16 3-18

9 10

11

3-19 3"_22 3-23

12 13 13a

3-25

LIST OF FIGURES(Continued) F.i_ure 14 15 16 Preliminary Validation of"Assessment Methods Comparisonof Loudness Celculotlon Methodsfor Trlongutar Transients Comparisonof Time-lntegrated Measures of the Impulsive Noiseswith the SameMeasureof the Equally Noisy Reference Sound CorrelaHon of Judged Degree of Helicopter Blade Slap Versus Crest Level illustration fromTwo Groupsof Helicopter Blade Slap Data "l_at Rank Order of AnnoyanceDoes Not Correlate with Judged ]mpulslveness Page 3-26 3-28

3-30 3-43

17

18

3-43

19 20 21

Comparison of Noise Levelsfor EqualAnnoyance VersusEqual Loudness 3-48 Typical Transmission Responso e f the Outer and Middle Ear "l'he 1968CHABADamage-Risk Criterion for Impulsive Noise Exposure and a Proposed Modification Fora Nominal Exposure of 100impulses Per Dayat Normal Incidence Comparison of Measured and Estimated Values of the Sub iective Correction Factor _ as a Functionof Crest Level
S

3-51

3-54 4-10

22

iX

1.0

IN 11_ODUCTION Under the mandate of the Noise Con_ol Act of 1972, the Environmental

Protection Agency is charged with taking steps to abate sources of noise potentially, detrimental to the public health and welfare, implicit in this is the need to establish

the meansfor evaluating and monitoring the noise from impulsive noise sources. Thisreport excludes cons_deratlon of humanresponseto'and measuremeno tf hlgh level impulsivesounds such as sonic booms, weaponsfire, or quarry blasts. The latter topic hasbeen the subject of recent recommendations to the Federal Government by Working Group 69 of the National Research Council, Committeeon Hearing, fiioacousticsand Biomechanics(CHABA). With this/imitation in mind, a research study was carried out to develop an interim methodfor the evaluation of moderate levels of impulsive noisebelow hearingdamage risk levels. The methodwasto be compatible with the existing methodology currently in useby the Environmental ProtectionAgency (EPA) for evaluating community noise impact. The investigation was divided into three basic elements: 1. Selection of a baseline metric for evaluating impulsivenoise to whichsubjective and objective correction factors* could be app/ied as necessary. 2. Review and evaluation of the literature on subjective effects of impulsive noisewith emphasis on date relating to annoyance, noisiness, or loudness of repetitive typesof impulsivenoise.

*'l_roughoutthis report, the term "subjective correction factor" is usedas a convenient label for the difference between Ihe subjectively effective and objectively measured value of Ioudness_ noisiness or annoyanceas defined in the text. It is not intended to imply that the velues cited Forthese "correction" foclorscan be used without careful consideration of their validity and applicability Forpractical evaluation of real Impulsivesounds.

1-I

il

3.

Basedon this review, the developmentof a suitable method to account Forsubjective (annoyance) effects of impulsive noise utilizing suitable measuremenm t ethods and curren!ly available instrumentation.

This report presentsthe resultsof this investigationin the following sequence: Section 2 discusses the selection of the baseline noise metric used throughoutthe study. Section 3, the heart of the report, reviews the literature in detail on loudness, naislness,and annoyanceresponset s o impulsivesounds. Other subjective effects are alsobriefly covered. Section 4 summarizes the overall findings in terms of the differential subjective response betweenimpulsive and nonimpulslvesounds. Threeappendicesare also included, covering: Appendix A - Objective factorsinvolved in the measurement ef impulsive noise. This includespresentationof resultsof a laboratory test of variousnoisemetricsobtained from a precision impulsive sound level meter when applied toa wide range of artiflcially-generated impulsivesounds. e Appendix B - Summery of the resultsof on international Round Robin teston response to and measurement of impulsive sounds recently cenducted by the International Standards Organization. Appendix C - Frequency spectraof repeated time bursts. This appendixbriefly illustratesthe spectral content of various ideal repetitive tonebursts which roughly approximatesomeimpulsive seunds.

1-2

2.0 2.1

SELECTION OF A BASELINEMETRIC Definition of Impulsive Noise Soundscan be defined as impulsive when they exhibit someform of rapid and

substantial variation _n the envelope of the time history of the instantaneous peak pressures. Th_s envelope can be visualized as a llne connecting the instantaneous peaks of a noise signal as measured on a h_gh-speedoscillograph. Examples of envelopesof impulsive and nonlmpulsive sounds,illustrallng this qualitative definition, are shownin Figure 1. Figure la shows the envelope of peak pressures for Fairly steady soundsfrom a stationary no_sesource suchas an electric motor runningat constant speed. Figure lb shows a noise with a noticeable fluctuation of the envelope. This may simply be called an unsteady or fluctuating noise such as froma streamof highly variable traffic. The first step in defining a baselinemetric for the impulsivesounds considered in this report was to classifyall typesof impulsive-like sounds into categories. As illustrated in the figure, mosttypesof impulsivesounds fit into two basic categories. Figureslc and ld show envelopesof the time history for sounds in these two categories that are clearly impulsive - Figure lc illustratesa single impulse suchas froma quarry blastand Figure ld shows a repetitive impulsivenoisesourcesuchas from an unmuffled rock _,r[ll or drop hammer._' Thereare clearly other exampleswhich fall somewhere in between the time history charaoteristicsshownhere. For example, the enveloperepresentingthe time historyof an aircraft may look quite slmilar to that of the single impulsivesound except that the tlme scale is stretchedout to many secondsinsteadof hundredths of a second. However, in order to take advantage of any usefulresearchthat could be related to impulsive noise, investigations on subjective reactionsto all of the last threeexamples illustrated in Figure 1 were groupedinto threecategoriesaccording to the typeof soundas Follows: *The lati_"r is a wheeled vehicle equippedwith a hydraulically operateddrophammer and |s used for demolition of road surfaces. 2-I

t
a) Steady Sound

r_ea.--t

I sec..._

b) Unsteady or Fluctuating Sound

I_.,0ms-I
c) Sfngle ImpulsiveSound

d) Repet_tlve Impulsive Sound Figure I. Examples of Time History Envelopes oF Nonimpulsivo(seea, b) and fmpulslva(see c, d) Sounds 2-2

I i! I!! -

Repetitive Impulsive Sounds Single ImpulsiveSounds UnsteadySounds

This review of impulsive noisesis necessarilybroadand potentially applicob/e to a wide rangeof moderateto low level impulsivesounds. To illustrate the concepts presented in this report pertainingto loudness and annoyanceof repetitive _mpulsive noises, four particular sources were selectedas typical of impulsive community noise. These are: 1 Truck-MountedGarbage Compactors Drop Hammers Two-Cycle Motorcycles Rock Drills

Clearly, someof thesesources can generate impulsive noiselevels which may representa hearing damagerisk to the equipmentoperator or an immediatelyadjacent bystander. However, hearing damageas_cls of impulsivenoise are not considered in any detail in this review. Undercertain operating conditionsor with suitablenoise control features, thesenoisesources may not emit what would be called impulsive noiseaccording to our qualitative definition (i.e., rapid and substantialvariation

in the envelopeof the peak pressure time history). However, according to our three categoriesabove, all fourof thesesources,when9eneratln_ impulsivesound,will fall into Category !, i.e. s sources of repetitive impulsivesounds. Typical time historiesof the instantaneous signals for each of the above sources are illustrated in Figure 2.* For garbagecompactors,ignoring the steadynoise of the power sourceusedfor its operation, the impulsTve nature of compactornoisewill consistof randomor irregular impactsof metal against metal so that the term "repetitive" must, in this case, be interpretedas including suchan aperiodic or randomrepetition. For the other three sources,however, one can expect that under any given operating condition, the repetition rate will be fairly constantso that the envelopewill exhibit a definite periodicity. It shouldhe pointed out that repetition ratesof concernin

this report will fall below the auditory range, that is, belowabout 20 Hz.

The time histories shownin Figure2 were obtained from a small samplewithin eaoh source category. Theyare not necessarily representative of all equipment t_at fall within those categories. 2-3

-_

200 ms I"

.t_ m_
a) CommercialGarbage Truckwith Compaclor

Figure2.

Examples of Time Historiesof the Instantaneous pressure fromimpulsiveSources

2,.4

b) Drop Hammer

Figure 2 (Continued)

5-5

='J 2.5 ms 14 c) Two-Stroke Motorcycle

F_gure2 (Continued)

2-6

I__l

ilrll

f ,0ms r"
d) Rock Drill Figure 2 (Concluded)

2.-7

A typical train of impulsive sounds is illustrated in Figure 3. The flve physical parametersimportantFordescribingimpulsivesoundare defined Forpurposes of this report as Follows: Crest Level - The difference in sound pressure level between the peak and rmslevel c_fthe noise. For a backgroundnoisewith a normal (Gaussian)dislrlbut_onof instantaneous pressure,the peak pressure may be considered as the value at about three standarddeviatlons above the rmsvalue. Thispeak, which ideally is exceeded only 1

percent of the time ForGaussiannoiset will be about 10 dB higher than the rmsvalue. Thus, the crest level shouldnormallyexceed

about 10 dBbeforea noise is considered impulsive. Duration- Theamount of tlme that the envelope of the instantaneous pressure exceedsthe rmsvalue. I Period (if repetitive) - The time durationbetween two successive impulsesin a train of impulses. Spectrum - The Frequencydistrlbutlonof acoustic energy in the impu]so. Rise Time - The time required for the impulseto rise From the backgroundnoiseto the peQk.

-7-

RiseTime -Crest Level_ _ P __

D: Duration P: Period

Figure 3. PhysicalParameters of a Typical ]mpu]slve Sound 2-8

Representativevalues for these impulsive noiseparameters for the two-stroke motorcycle, the drop hammer, rock drilJ, and truck-mounted garbage compactor are listed in Table 1.* For these sourcesof impulslve noise, the crest level lles between 13 and 30 dB, the duration varies from several m_lllsecondsto half o second, and the period varies from 10 milliseconds to 1-1/2 seconds. A frequency range of 200 Hz to 2 kHz covers most of the acoustic energy of the impufslve noise. This table provides a general indication of"the magnitude of the parameterswhich define the general physical characteristics of the impulslve noise sources considered in this report. However, this range of parameters, in fact_ _ncludes many other impulsiveno_ses so that research_nto subiective response to all of these can be applied, in part, to the evaluation of subiectlve response to the four particular sourcesidentified _nTable I. Table ] Typical PhysicalParameters o Four Real Sources of Impulsive Noise Peaks _n impulsive Noise Source Two-Stroke Motorcycle Drop Hammer RockDrill Truck-Mauntec Garbage Compactor 30 300 1500 0.25 - 1 10 Crest Level dB 13 Pulse Duration ms 2 - 20 Repetltian Period ms 30 - 100 Frequency Spectrum k Hz 0.30 - 2 Typical Rise Times ms 2

19 19

10 500

50 5000

0.040 - 0.400 0.200 - 1

2 50

*The values listed in Table 1 were measuredfrom a smallsample within each source category. Although there is no reasonto suspectthat the values listed are atypical, the reader should apply caution in generalizlng the conclusionsof th_ss'tudy as necessarilyrepresentativeof all equipment that fall wHhln each sourcecategory. **Although selected as a repetitive impulslvenoisesourcefor purposes of thls analysis, recent in_'ormatlon as presentedin EPAReport No. 550/9-79-257, RegulatoryAnalysis of the Noise Emission Regulatlonfor Truck-MountedSolid Waste Compactors, indlc:etes that th_sfeature may not be necessarily characteristic of the majority of truck-mounted solid wastecompaotionunits. 2-9

2.2

BaselineNoise Metric Some sort of baseline noise metric is necessaryfar evaluating these various

impulsivesounds. Thisbaseline metric shouldbe: (1) reasonablyunambiguous( , 2) measurable with precisionlaboratory equipment, (3) measurable with standard sound level metersin the field with suitable correction factors, (4) compatiblewith the day-night soundlevel (kdn) or the equivalent (energy average) sound level (Leq) metric, and (5) able to provide a foundationfor application of subjective impulsive noisecorrectionsto allow comparison of the subjective response to impulsive and nonlmpulsive sounds. The baseline metricsapplicable to the Category ! impulsive sounds could takeone of the following alternate Forms. SoundExposureLevel - The tlme-lntegrated measure of the A-welghted soundlevel is identified by the symbolLS. e Equivalent Sound Level - The equivalent soundlevel is the energyaverage of the integrated A-weighted soundlevel over a specified observation time Tand is identified by the symbolkeq. Peak SoundLevel - The maximuminstantaneouA s -welghted sound pressure level duringa given observationtime is identified by the symbolLAp k. Peak SoundPressure Level - Themaximum instantaneous unwelghted (linear) soundpressurelevel duringa givenobservationtime is identified by the symbolLpk. All of thesemetricsare essentially unambiguouq s uantities measurablein the laboratoryand potentlally measurableby someof the advanced integratingsoundlevel meters. Measurement of the peak levels (LAp k or Lpk)with soundlevel meters equipped with a peak-hold posltion is straightforward,providing the rise-time of

2-10

the signal _sgreater than 50/_secs. Th_scorresponds to an upper Frequencylimit of 20,000 Hz for significant energy in the spectrum of the impulsive sound. ]ntenfionally excluded Fromthe candidate baseline metrics are the other quantities measurable on a sound level meter. Those which will be consideredlater for application to measurement of impulsivesounds include: SlowSoundLevel - The exponentlal-averagedA-weighted soundlevel measured with a nominaleffective (squaredpressure)tlme constantof 1 second, identified, for this report, by the symbol LAS. Sound Level or FastSoundLevel - The exponential-averaged A-weighted sound level measuredwith a nominal effective time constantof 125 ms, identified, Forthis report, by the symbol LAp" e Impulse Sound Level - Theexponentlal-averaged soundlevel measured with o nominal effective time constantof 35 ms,. identified by the symbolLAI. Other noise metrics could have been considered, such as measures of statistical distribution, L x , where x is the percent exceedencelevel, or noise pollution level (LNp)which attempts toaccount for subjective reaction to fluctuation of a noise. These were rejected as not being directly compatiblewith current EPAnoise metrics and are not readily measurableon standard soundlevel meters. Returningto the Fourcandidatebasellne metrics, the last two measures of peak level may be rejected at the outsetas unsuitable becausethey Fall to fit directly ; into EPA's time integratedmeasuresot'nolse, namely, day-night sound level Ldn a,3dequivalent level (Leq). In order to make a final choice, it is nece_ary to consider the general nature of the noisesignaturesthat may be Tnvolved. For example, the typical no_se exposureof an individual at any one place to garbage compactor noise might consistof several minutes of exposureto a relatively randomseries of impulses generatedby the clanking together of garbagematerials as they are compacted, superimposed over the risingand falling humof noise From the engine which drives the compactor.

2-11

The duration oFthe exposurecan only be roughly estimatedand will vary w_delyfrom one site to another and Fromone day to the next. The soundexposure level of such

a varying noise exposurewould also vary accordlngly, making it difficult to utilize for realistic noise evaluation or certification unlessone observation time were arbitrarily fixed. In this case, however, an equally useful measurewould simply be the equivalent (or energy average) sound level (Leq) during the measurementperiod. ]n contrast, duringa passby of a motorcycle, the only unambiguou= energyrelated measure of the no_se signaturereceived by a nearbyobserverwould be the sound exposurelevel (Ls). It would be possTh[e to normalize the soundexposure level by a standarddurationof, say 10 seconds to provide what would amountto the equivalent sound level over I0 seconds (i.e., Leq(10 sec)) with the sameenergyas the actuat event. On the other hand, if noisecertification testsof motorcycles were to be applied to stationaryvehicles, the equivalent soundlevel (Leq) during the observation period wouldbe a logical baselinemetric. For the drop hammeror rock drill, a typical nolse s_gnature could consist of a relatively long period of exposure, on the order of an hour or marewith many periodsof moreor lesscontinuous exposure to the repetitive impulsive sound. In this case, agaln, the equivalent soundlevel (Leq) during the observation period appears suitable as the baseline metric. Thus, with the one exception of noiseexposureto single events, which are conveniently defined by the soundexposure level, it appearsthat the equivalent sound level (L ) is the logical choice fora baseline metric for the impulsivesourcesconeq sideredin this study. The A-weighting inherently incorporated in this metric is expected to provide a moreaccurate or a moreconsistent correlation with humanresponse to low level impulsivesounds than wouldbe providedby a nonwelghted (linear) sound pressure level. As will be discussed later, this observationis also consistent with the observedloudness or noisiness of low level sonic boomsounds. Thesehave been shownto correlate best

2-t2

with Frequency-welghted measures (i.e.,

loudnessin phons) of the sonic boomenergy

spectrumwhich deemphaslzes the low frequenciesas doesA-welghtlng.*46' 551 60 It remainedonly to define theobservation time uponwhich the average sound level will be based. For the general case, the equivalent sound level (Leq) over an observationtime T will be defined as T ,j where PA(t) = instantaneous A-weighted soundpressure at time t, Pa Po = reference pressure (20 h_Pa), and T = observationperlod_ sec Far prediction of the day-nlght sound level (L.), the L for the impulsive dn eq soundis evaluated far the daytime (Ld) - 0700-2200_ and for nighttime (kdn) - 2200 to 0700 hours. 111e normal 10 dB penalty factor would be imposed on /dn Forthe baseline metric, but the possibility of increasing this For the potentially even greater annoyance at night of impulsive soundscan be left as an option to be defined upon the basis of examining the available information on sleep interference Fromimpulsive sounds. For application to defining the k of"repeated single events, the some techo eq n|que employed Farspecifying aircraft soundexposurewill be used in the form Leq LS + 10 log N - 10 log [ T/t'l dB (2) Leq = 10 IOgl0 ( PA

*Note that for _impulslv.e. sounds,suchas fromquarry blastsor artilleryt C-weighted levels appear to predict communityresponse quite well.21t 147

2-1J

where LS = sound exposurelevel of one event, dB re 20p.Pa . sec N = numberof eventsduring the time T T = observationperiodin seconds t = referencetime of I second The observationtimeT toapply in the measurement of the equivalentsound level will dependon the application, ranging froma minimumof 1 second (correspondingto the duration of reference sounds often usedin laboratory evaluation of impulsivesounds), to 1 hourfor an hourly equlva!ent soundlevel (Leqlh)), to 15 hours for the day soundlevel (Ld) - the energyaverage during the hours0700 to 2200. In summary,then, the baseline metric used in this study for evaluationof. impulsive noise will be the A-weightod equivalent soundlevel (Leq) measured over a time to be specifiedas appropriatefor each source. This provides a baselinenoise metric that is compatiblewith the existing methodsdevelopedby EPA for evaluation of noise impact.* By providingadjustmentfactors(nominally identified hereinas correctionfactors) to the Lq toaccount for any subjectiveeffects and measurement errors for impulsive noise, it will be possibleto properlyinclude impulsivenoisesin EPA'sevaluation of environmentalimpact of impulsive noisesources. This metric is also considered appropriatefor application to each of the three categoriesof sounds defined earlier= (a) Category ] - Repetitive impulsive Sounds,(b) Category|] - Single ImpulsiveSounds,and (c)Category Ill - Unsteadyor Fluctuating Nonlmpulsi_,e Sounds.

*U.S. EnvironmentalProtection Agency, "informationon Levels of Environmental Noise Requisite to ProtectPublicHealth and Welfare with an Adequate Margin of Safely." EPAReportNo. 550/9-74-004, Mo_ch 1974.

2-14

'_

3.0

SUBJECTIVERESPONSES TO IMPULSIVE NOISE Subjective responses of people to noise can be conveniently groupedinto three

general (and overlapping) categories: Health-Critical Responses Hearing damage Long-term medical or psychologicaleffects other thanhearing damage Activity or Behavioral-lnfluence Responds Speech interference Sleep interference Task interference

Attitudinal or Judgment-lnfluence Responses Annoyance responses Loudness (or noisiness) judgments

The primary concern for subjective responses in this report is in the last category (i.e., attitudinal or [udgmeni responses)e and therefore that category is the only catggo_ that hasbeen reviewed in depth. An extensive bibliographyhas been compiled, however, on most of the above categoriesand is included in the Referencesectionat the end of this report. For convenience, the bibliography is arrangedchronologicallywithin each of five general subjects: Part At Annoyance of Impulsivo or Fluctuating Soundl; Part B_ Loudness or Noisinessof Impulsiveor Fluctuating Sounds;Part Ct Dal_ction Qr Perceptionof Impulsive Sounds;Part D, Speech Interference; PartE, Sleep InterFerence; and Part Ft Hearing Damage. An additional subdivisionPart G, for the referenceson measurement of impulsivesoundoisalso included in this bibliography. While all of the sources are listed in the bibliogrophyt for convanience_only the principal onesof concern for this report are cited as references in the main bodyof the text.

3-1

3.1

Loudness or Noisinessof ImpulsiveSounds

Aswill be shownlater in Section3.2t a correction to L to account for the eq annoyanceof impulsivesounds can range From approximately5 to 15 dBt dependingon the correctionmethod. Clearly_ sucha wlde range of correctlon,Factorsisof little value so that a moreprecisemethodfor selectinga subjective correction factor is desired. The extensive literature on loudness or noisiness of impulsivesounds was therefore reviewed emphasizingexperimentalresultsas a more reliable basis,at this point, for assistingin the selectionof a subjective correction factor. In additlona these basic experimentalresultson response to transient sounds are expectedto assist in defining optimumwaysto monitor impulsivenoise. Following the review in this sectionof the avai]eble experimentalresuIts on loudness end noisiness of impulsive sounds_information related to the annoyanceof suchsounds and compa risenof annoyanceand loudness or noisiness is consideredin the next section. First_ howevert it is helpful to considera simplified modelfor the auditory process as a frameworkfor examiningthe data relative to impulsivenoiseresponse. 3.1.1 A Model Forthe HearingProcess A s_mplifiedconceptualdiagramof the audltory systemis illustrated in Figure 4 to assistin defining the principal features significant in thisstudy. As indicated in the figure, characteristicresponse times for the "acoustic" partsof the auditory chaln (i.e._ up to the point in the inner ear wherespectrumanalysisoccurs)are muchless than the "RC" tlme-constant inside the lastbox where the overall detection, integration_ . 146 and recognition of sound signalsis assumed to occur. Even consldering the lowest reported value for this tlme-constent, it is still more than me orders of magnitude greater than for the earlier partsof theauditory chain which mustbe able to respond to instantaneous pressure changesat rotesup to 20t 000 timesper second(_"= 50 _sec). The "RC" tlme-constant, on the other hand, only limits the abi/ily to track the envelope el _ a sound. Thus,experimentalstudieson response of humansto transient roundshave focusedmoreattention on this part of the hearingprocessand have utlllzed the RC smoothing filter concept illustratedas one of the ways to emplrlca[ly model the results. We will considerthe implicationsof the model illush'ated in

3-2

[n_r!enl Sound Beflaled Flora Nearb_ _oundoeies / ''4 / /

p ......................................... I I Analysis of Signal 1 Into (Ci*ial1 Dondl

Brain

D;f flaction By Hood Tnclder*t D;_ecrlv Sound f_o,e T=5OZ_s T'3Su=

( Spoclr ur'_ Analyzer) t , D_l_ctlon and [ J "X

Prace_lnU and Pattern

Smoorhlng Filter L........ r=RC 13IOOO ms -I

Recognition

Figure 4,

Concep_'ual Ir[ustrat'ion of Audltory Processto Show CharacferlsHc Response Times ('r) TnVarious Elements Which Govern the DynarnTcResponseof the Ear to Transient Sounds. i',Jote: the Range for the Value of. "RC" TimeConstant Reflects the Exl"reme Rangeof"@bserved Values. (After Bruel, Reference 146)

Figure 4 again later, but first let us examine the experimentaldata on /oudness and noisiness. 3.1.2 ExperimentalDqta The independent and dependentvarlables involved in the noisiness of impulsive sounds maybe categorized as follows: independent Variables (The Stimulus) Signal Format Repetition, Single or Multiple [rnpuTses Signal Spectrum Tone, Narrow BandNoise, Complexor Wide Band [mpu[slveNoise (the compleximpulseincludes the type of real impulsivesounds of concernin this report) SignalCharacteristicsVaried Pulse Duration PulseFrequency Pulse Repetition Rate (for Repeatedimpulse) Spectrum of Total Signal Riseand Decay Time af Pulse PhaseoESignal Components Ratio oF PulseSignal Level to any Background Noise Duration of Total Exposure Method of Signal Presentution(includingsoundField charocterlstios for loudspeaker presentation)

3.-4

DependentVariables(The Response) Thresholds AbsoluteDetection (absence of noise) MaskedDetection (in presence of noise) Flutter or FluctuationDetection

Magnitude Loudness Noisiness

Althoughnoisiness and loudness are listed as separatedependentvariables for subjective response to impulsivesaunds_ it will be shownthat they may be takenas essentially identical. However_as shown later in Section3.2.3_ the annoyanceresponse to impulsive sounds mayt in somecases,be significantly different from a loudness or noisiness magnituderesponse. Utilizing Iheabove frameworkof independent and dependentvariables on loudnessor noisiness of impulsive(or fluctuating) soundsa , n index of the pertinent available literature is presentedin Table 2 which covers mostof the major experimental studiesof subjective response to impulsiveor fluctuating sounds. It will be convenient to briefly review the pertinent findings of these experimentalstudiesby three general groupsaccordingto the type of stimulus. Pure Tones

Bursts of Noise e ComplexSounds (real or simulatedimpulsivenoise including heHcopierblade slap)

3--5

Table 2 Index of Experimental Studies on Loudness/Noisiness of Impulsiveor FluctuaHng Sounds Indicating Experimental Variables Investigated
F.efereaces No 25 26 27 28 30 31 32 33 34 35 Author Hu_het Garr_er & Millt Mumon Garner Miller Garner Garn*r Nietl Green HomIIton Yw_r (1946 (t947) (1947) (1947) ((947) (1948) (19,_) (1949) (1956) (1957) (1957) {1958) (1959) (1960) ((960) _1962) (]965) 0964) (1965) (1965) 0965) (1965) 0966) (1966) (1966) (1967) (1967) (1967) (1967) 11967) O 5|,L LL,A LL O LL D D O i LL LL LL AC, HBS L AM, R N N FT D D 55 D,R D,R O D,L O,F D D,L D,L D MT LL LL LL LL MT R LL LL_MT LL LL O_L D,F D D LL LL MT MT D LL 'D,L AT, LI. D,F,R,L LL Tones 0_F D,F,L DtF, L Narrow 6and ar Nolsa Sinqle ImT_flset(I) W_c_a Land C_mplax N_e M_asur_d (2) AT MT LL D,F D*F e MT,AT AT,MT Ton_l MulHoI_IRe-pentPdl ImpL,ht.t (I) Narrow [_and W_de _aad _f NQI_n Com_le_ Noise Measured (2

29 Garner

36 i Pollock 37 _omp 3S M-Fodor 39 41 42 43 44 45 46 47 48 50 $1 _2 $a 54 Nits* Sr_31[ Port + Sheilly + Carter Zw[cklr Zipler Garrllt Bcman Stewlnt + Zwlcker P*_ns Peon_

Du_ov_kll + dobson

_16 II_lulr

Table 2 (Concluded)
S[n_F_ Imp_he_
FPt_'r_n_ _'_O'rxgw

No. Aulhor 57 Horbert + 58 _9 Shepherd + Rothc;uJen *

year (1968) (19681 (1966) I]969) (]970) (1970) (]970) (]971) (1971) (t97t} (1972)

To,_el

Gand' of No_sl

Complex 5B

Withe 6and No_le

Measured LL,A

Multlple Re.PeQled[ lmpulJe_ [ Notto_Band ' Wide fiance Tone1 _ ef Nolte Complex No_ll R,_ _,RT

Measured FT

R 56 D D O D LL LL LL N,A D D_,F D_R D_R L,AC D D. D D LL J.k D D dD D H_

LL,A

60 Johnson * 61 RelckQrdt + 62 6364 65 Rdticha{dt Fid_ll+ Ollefhfod

kl. N_A LL LL LL LL,A

66 ShlplQn " 67 Thom._o_ 70 Leverton

72 Carter 73 St epher4 74 Carter 75 Boone 76 I Leverton 77 Pederlon 70 I Gustm'_on 79 Tethatdt I]0 Fuller +

(1972) (19731 (_9;_) (19731 (]974) (1974) (1974) (19741 (1975) D D LL

R_ET

LL

R,_'T D D HB$ D D,F,R, L LL LL AM, A D 121 DependentvxrlabJ_ m|a_ured JdentMed by obl:cevlatad ca,de. D,F,R,L

AT LL LL

LL FT IA

( ) Jndllpendlnt varlabhl idenllfied by abb_evtotedodl unde columnhead_ng wM_h define type oF_lgnol_ p*arlt_l, narrowband _f no_le, comple__lgnol, I_depe_dentVo_abl_s DF L X RTDuration _lseF_equency Signal /,evll _epetlt]on _tl 9Jll(_ndOlcoy}Time SB H6$ AM AC

Dependent Vcl]_blel A AT FT LL _T. Annoyance Ab_luteThrelkold Flmler Th,'el_Id Loudness Level M_kadT_'_lhold

. Sonic: Bo_mSignal . Hel_copfer61uda$1op . Amplhud# Modul_tIon . Alrcroft._und

(*+ +col)

3.1.3

Subiective Response to Impulsive Pure ToneSounds

Following pioneeringwork by Bekesyin 1929 on the effects of duration on 25 26 27 28, 29, 31,32 loudness oF tanest Hughes, Garner and Miller, Munson, and Garner lald the groundwork for subsequens t tudieson loudness of slngle or multiple tonebursts. Typical results of thisearly workare representedby the data of Garner and Millert 26 shownin Figure5. Figure 5a showsthe measuredslgnal-to-nolse ratio at detection

thresholdfor a single toneburst of varying duration presentedin thepresenceoFa wlde band maskingnoise. Figure 5b showsthese same resultsnormallzed acco'rdlng to a simple empirical modelfor the auditory detection process correspondingto theoutput of a resistance-capacitance(RC)circuit. The latter is driven by a signal (El) which

is assumed to representthe detected envelope of the tone burst. If we assumethat the tone is just detected when the peak output of the RC network reaches some fixed thresh' old detection level (Eo), then it can be shownthat for burst durations(T), much less than the time constants "r = RC (analogousto the ear's time constant), the required signal level increases inverselyas the pulseduratlon decreases, or (RequiredSignal Level, El) :" (r/T) (Detection Threshold,E O) (3)

Thus, the product of the signal magnitude E1 and the pulse duration T isa constant, as given by E1 " T_'E o 1"= constant (4)

Since the product of the slgnal magnitudeand pulseduration is a measure of the "energy" in the signal, this relatlonshlp is simply another way to define the so-called "constant energy" law normally invoked to explain why_ for pure tone burstswith a short duration relative to the ear's time-constant, the requlred slgnol level for detection increases 3 dB for every halving of the burstduration. This very sameresult wasalso obtained by 27 Munson when a tone burstwas adjusted in level to equal the loudness of a fixed duration reference tone longer thanabout 50 reset. However, as suggested by Munson 27 and manyothers subsequently,a simple "RC" circuit model Forthe ear'sresponseto trans'ent sounds has o I re'ted appl*cohon.

3-8

40
a) Measumd Dato

' i

' i
O

b II
400 Hz

35 -_ A

_ D 0

I000 67 19 00 0 inTelm*

Noise _eesufod

_ 30 2-

of Sp*cr_u L.v m ol o

o 25

For

The practical implication of this model for impulsiveno_seis that it could offer a way to select an optimumprocedurefor measurement of impulsivesounds by duplicating, electronlcally, the ear's internal tlme-constant. Sucha rationale is the basis for the 35 mstlme-constant selected for impulse precisionsoundlevel meters(see Appendix A). Unfortunately, there are several co'replications in this simplisticmodel which

are broughtout by the experimental data. Reichardt and N_ese,6] employingo subject panelof 50 people, found that the loudness matchingof a toneburst of variable duration againsta fixed durationreference tone, usuallyof the orderof 1 second long, was a very difficult experimentaltask for the averagesubject when the twoburst durationswere substantially different and led to o great deal of data scatter not indicated by the smaller subject panels (four to six) ia most other studies. By usingreference tone durationsnear the middle of the range evaluated for the test tone, they foundmuch lessscatter in the loudness balances. On the basisof their refined technique, therefore, they measured o time-constantof 30 milliseconds. 61 Theserefinements also included a careful selectionof the temporal spacingand durationof the test and reference signalsto avoid possible maskingor memory errors in comparinga testand reference tone or to avoid what they termed"roughness" which was observed whena rhythmically repeated pattern was usedfor the test or referencesignal, particularly at pulse repetitionrates on the order of 3 to 50 Hz (Reichard162). Thi._.s qualitative measure, "roughness," may be importantin the evaluatlon of impuls;ve aolse. Other factorswhich can causevariation in the observedtrade-off between signal level and durationare: (I) the "energy law" fails either when Ihe signalduration T is so short that a substantialportion of Hsfrequencyspectrum falls outside thecritical band centered on the pulse frequency (Garner29), or the signaldurar;on is muchlonger than the ear's time-constantr (2) the apparenttime-'constant increasesas the signal 26 75 level approachesthe threshold of bearing (Garner and Miller, Boone ), and (3) the time*constantapparently varles with frequency, as ;mplled by the data shownin

3-10

Figure 5. It hasgenerally been accepted practlce, however_ to assume that the timaconstant does not vary with frequency. The lack of agreement between investigatorson the time-constant still continues. A recent study by Boone75 an loudnessof repeated short tone burstsin noise, using 20 subjects, produceda value For the tlme-constant of about 110ms. Terhard7 t9 has suggestedan RCtlme-constant of 13 msecto fit his unique measurements of the detection of periodic slnusoldal modulation (which he calls roughness) of pure tones. In summary, there Ts substantial evidence to supportvalues for the time-constant ranging from 13 msto over 200 ms (see Figure 6). Becausethere is no apparent way to resolve this issueunequivocally for this report, the only practical choice appears to be to work with the ex[stlng recommendations or practice for the choice of timeconstraints in impulse precision soundlevel meters. This lack of agreement on the auditory t[me-constent is most unfortunate for it [mplies Ihe potential for conflicting evidence about a subjectlve correctio_ factor for impulsive sounds. This point is illustrated in Figure 6 which showsthe potential range of the tlme-constant basedan the range of experimental data relating perceived loudnessof an impulsivesound versusits duration. Thus, for a given duration of an impulsive sound, Ihe potential increase in the signal level to achieve a loudnessequal to that of a reference (nonlmpulslve)tone can he substantial. Clearly, any correction factor for impulsive noisemust be basedas muchas possibleon experimental date for subjective response to real impulslve sounds.

3-11

3_

20

15 Possible Range of Loudness for 20 ms Pulse

Measured

Renge of
tlon

"5_ cu o. _ "5 _ 10 "_ u .u_

"13 L._

,I 0.1

I 2

I 1

J 2

f 5

"t 10

100

5 1000

Pulse DuraHon, ms Figure 6. Range of'Measured Sound Level - Duration TradeofF (where the Pulse is Judged to be Equally Loud to the Reference Tone) Reported from Various Studies to Indlcate Possible Range of Uncertainty in Predicted Loudnessof 20 ms Pulse. (Note that the Time Constant Specified by I EC Sound Level Meier Specifications for "Fast"Would Tend to Fall Near the Middle of the Rangeof Measured Data -Adapted from P. Bruel, Reference 146.)

Pu!seRepetition Rate Another maj or varlable studied in loudness tests of repeated tone bursts is the repetition rate. A very definiHve study in this area was naported by Garner. 31 From tone bursts, a subjecHve correction factor for real

these results on repetitive

impulslve noises can be inferred

His exper;rnental procedure consisted of

presenHng, through monaural earphones to six subjeats, a oonHnualty repeated pattern of a steady 1 second reference tone, 1/4 second silence, a I second cycle of repeated tone bursts, a 1/4 second silence, ] second reference tone and so on. The repetition

rate of the repeated tone burst group was varied from 5 to 100 pubes per second, the pulse duration varied from 1 to 50 ms, the pulse frequency varied From 125 to 8000 Hz and the intensity level of the pulse varied from 20 to I00 dB, The subject varied

3-12

the intensity of the tone burstsuntil he obtained equal loudnessto the referencetone. In mostcases, the energy in each tone burst groupwas lessthan that of the equally Joud steady reference tone. The subjective correction A for thesetone burstsis simply s for equal loudness the positive difference between the sound exposurelevels of the reference end test signals. Since the tone burstgroupand the reference tone each lastsfor 1 second, the difference in sound exposurelevels is also the difference in equivalent sound levels (L ). Thlssubjective eq correction factor is shown_nFigure 7 for I000 Hz tone bursts and coversthe repetition rate and pulseduration range indicated. The intensity of the reference level was80 ,-lB. The typical variation of A S with reference level and frequency is shownin Figure 8.

16

'

'

'

Pulse Duration, ms

12

1 _.80

dB -.0

_" ,
a..,oJ IP

10 8

20

,
i.

,
5

I
10

,
20

,
50

I
100 200

PulseRepetition Rate, pulsesper second Figure7. Sub iectlve Correction A for Repeated1000 Hz ToneBursts (from Garner 31 s

3-13

a) 1000 Hz 6

"D

-2

[ 20

I 40

I 60

I 80

100

L (Ref), clB eq

10
I l I I I

'<

0 100

r 2

I J I 5 1000 2 Frequency, Hz

I 5

10,000

Figure 8.

SubjectiveCorrection _S for RepeatedPureTone Bursts, PulseDuration 20 ms, RepetitionRate = 25/Secondas a Functionof a) ReferenceIntensity, Le (Ref)at 1000 Hz and b) Frequencyfor the 80 dBReferenceLevel (From Garner31) 3-14

As exhlblted in Figures7 and 8, the subjective correction factor behaves in a complex Fashion,even for simple tone bursts. TheseFigures indicate the potential difficulty of developing any simple, general method for predlctlng a subjective correction factor Formore complex impulslve noiseswhichhave different spectra, rates of attack or decay, or amplitudes. Nevertheless, Garner was aerie to readily predict hls experimental resultsr suchas thoseillustrated in Figures7 and 8, on the basisof two basic factors:31 !. The spreading out of the Frequencyspeclrum of repeated tone bursts. As a result, the side band frequency componen_ of the repetitive tone burst can fall into critical bandsoutside the one centered on the honeburst carrier Frequency. 2. The shape of the. loudnessgrowth functlon. Due to thls unique shape,

the loudness of the s_de band componentsin each of the several critical bandsinvolved in this broader spectrum of repeated tone Burstscan add up to o greater loudnessthan the sum of their energies because, at noise levels well above threshold, the relative loudnessof e sound changes much moreslowly than the relative intensity (he., 2-to-1 change in loudnessfor

a 1g-to-1 (10 dB) change in intenslty). That Zs,the loudnessof soundsis roughly proportlanal to the sumof the loudnessin critical bands and the sum of these loudnessvalues in the slde band components will not decreaseas rapldly at frequencies removed fromthe tone frequency as the physical energles in these side band frequency componentsof repeated tone bursts. There is really nothing new here, of course; it is simply the basic concept of loudness summationof complex sounds whlch has beendeveloped into a fine art by Stevens,93, Zwicker, 89 and Niese. 86 However, appllcation of these well*developed concepts for loudness of sounds hashad only limited applicaHon to impulslve sounds. It is important to recognize that the conceptof "startle" _snot involved in a predlction that a weaker impulsive sound can sound louder than a strongersteady-state sound. This simply results from the accepted concepts for simulating the loudness

3-15

.....

.+ L

.....

.......

....

+.

perception of sounds. It remains to be shownthat there may _ndeedbe on additional effect that makesimpulsive soundsmore annoying than indicated by their loudness. Frecluenc _ Spectra of Repetitive Tone Bursls A brief consideration of the Frequencyspectra of repeated tone burstsis in order here since this plays such a primary role in the concept just outlined. As Figure 9 shows,

repeated tone bursts produce a spectrum centered at the frequency of the tone with s_de bandsabove and below this frequency.

RepeatedTone gurst

.-_1"J

(T!I
=1

_'z-.,_-_._

1" -_

Figure9.

Time History and Fourier SpectrumOf a Typical ImpulsiveSignal

In Appendix (3, it is shownthat the inverseof the duty cycle of the pulse('r/T) provideso qualitative indication of the number(N) of side band harmonies within the nominal "1/2 power" spectral bandwidth. The more the repetition rate increase=,the _ouder the pulsetrain will be, if the total energy staysconstant.31 With a very long duration, and a very slow repetition rate, all the energy of the signal is concentrated in a narrowrange of frequencies. IF this range falls within a critical bandwidth, then the loudness variesas the signal energy within this band. On the other hand_ if this signal spectrum bandwidthis muchgreater than the crltlcal bandwidth, then loudness of the signal will add approximatelyas the loudness of energy in each critical bandbut with emphasis on the loudestband. Thebroadening of the spectrumof tone burst=beyond the frequencyof the tone itself introducesan inherentcomplication in evaluatingsubjective response to intermittent sounds. Thiscomplication is overcome, in e senseFby usingo testsignal broadbandrandomnoise, which already has e broad spectrum. Thus, the spectrumof

3-16

repeated bursts of wide band randomnoised_Ffers from the spectrumof the uninterrupted noise only at frequencies, which are generally infrasonic, corresponding to the burst repetition rate. Spectra of s_ngle burstsof broadbandnoiseare nat significantly different from the spectrumof the steadynoise itself. 3.1.4 Subjective Response to Burslsof Noise Although Garner28 and Miller 30 carried out initial studieson responseto bursts of broadbandnoise, Pollack36 presented thefirst extensive studyutilizing pulses of wide band noise. For our purposes,his resultsmaybe summarized as showingthat the difference between the L of a continuousnonlnterruptedbroadband reference noise eq and the L of equally loud pulsesof the samenoise increasedfrom 0 to +10 dB as the eq duty cycle of theburstsdecreasedfrom I Io 0. f. Forduly cycles below O. 1, the difference remainedapproximatelyconstantat +10 dB. Thus, the subjective correction factor AS wouldbe +10 dBfor duty cycles less than0. I. Small, et al, 41 useda moreconventionalprocedureof balancing loudness of repeated bursts of noiseof various durationsagainst interspersed 1/2 second burstsof a constantlevel reference noise. They found that whenthe sensationlevel of the reference noise burstwas60 dB (a typical listening level), the level of an equally loudvariable duration test burstwasconstant for durationsdown to 15msecand then in=teasedby 12.5 dB for each 10-to-1 decrease in duration Forshortertestbursts. For our purposes,this is equivalent to the subjectivecorrection As increasinglinearly, at a rate of +3 dB per halving of test burstduration, from a value of zero for 1/2 secondnoise bursts to a maximum of 15.2 dB for a 15 msnoise burst, and thendecreasinglinearly at a rate of -0.75 dB per halving of test burstduration for shorterbursts. "1his assumes that I/2 secondis the time base for computingthe I eq of thevariable durationnoiseburst.

Garratt47 hasrepeated the testsof Pollack 36 and Small, et al,41 with very similar resultsas shownin Figure 10 where the measured valuesof ,', versus ratio of test signal
S

duration to referencesignal duration is plotted.

Similardata on loudness of a short

burstof 2 to 4 kHz noise FromBaue5 r 6 is also includedalong with data on relative

3-17

noisinessof short bursts of noisebandsfrom Fidell ond Pearsons6 . 4 The latter showlittle agreementwith the other dato but this may be due to the uniqueexperirnental technique employed(flee-field presentation, interaction with a computar for signal presentatlon), and the "noisiness response"insteadof loudness. Thevalues for _s _n this case are actually differences in measured A-weighted noTselevels of the reference and test signals. The effect of A-weighting on the short noise burst levels is not clear.

20 I

I m_

_3 Parlack 6

'Ref. D.fallon * _" 0.S _ t.0 S_ 0.9 So_ 1.0 See

NQ'lei W}l_re Whito V,l_ite 2-4 kHz Oclav0 & Wido _orDd

S_all, _r ol4t T5 "_ -_' tO -._1_J I _ morn O Garrott47 Bo_er, et ol56 Fidorl P_arso,_ 64 s*

"Equol No;sleetsc 1 Annot, ancd (F_eo-Fiald)

.jSl <1

0.002

O.01

O,I

T.0

Dural'ionof TestSignal/Duration of ReferenceSignal Figure 10. Subjective Correction Factor for Loudness or NoisinessResponse to Short Bursts of"Noise Bands Relative to a Refei'en_:e Noise. (Solid and Dashed Lines[dentlfy Mean LinesThrough ExperimentalData which Vary _ 42 dB a_oout Mean). Finally, returning to Pollack,36 one particular set of his data provides a good model for examiningloudness of more realistic impulsive sounds. Thesedata were obtained on the loudness of partially interruptednoise. This consisted of a continuous background noise with a superimposed periodic increase in noiseby amounts varying from

3-18

0 to 45 dB. Figure 11 shows the resulting data obtained under one condition oF a repetition rate oF I pulse per second (pps) and a burst duration of 1 ms. The ordinate defines the loudness level oF the composite signal relative to the loudnesslevel Ear continuous noise at the same intensity as the noise peak. The dashed llne on the figure shows the computed

L For this noise signal to illustrate, again, that the equally laud impulsive noise has an eq L substantially less than the L oF a continuous signal with the same maximum level. eq eq The resulting subiectlve difference Factor '_s approaches a maximum value of about 10 dB For a ratio of noise burst to backgrounJ nolse greater than 30 dB. Other data by Pollack 36 and Garrett 47 on partlally 0 Pollack (Figure 1 oF Ref. 36) -5 I "_ l- ",_". I "_ o x Pollack Pollack Garrett Drawn (Figure 4 of Ref. 36) (Figure 5 oF ReF. 36) (Ref. 47) to F_t Data interrupted noise gave similar results as shown in Figure 11.

"_

-10

-15

_\__

,'<
',

/.-

Equal Loudno ess Level

- 20
-25

,
,_

o A S _ 10 5dB

-30

Computed L eq I 10 I 20 I 30

.............

-35

I 40

I 50

t 60

70

Burst to Background Ratio, dB Figure 11. Difference Between the L oF the Repetitive NoTse Burst Superimposed on a eq Steady Background Nolse and the Level oF a Contlnuous Noise which Sounds Equally as Loud as a Function of the Ratio of the Burst Level to the Background Level (Burst Duration : 1 ms, Repetition Rote : 1 pps)

3-19

In summary, with the exception of the resultsof Fidell and Pearsons 64 , the experimental data on loudness or noisiness of shortburstsof randomnoiseshawconslstont trendssimilar to the toneburstdata in termsof order of magnitude values for the subjective difference Factor. Limited resultson the'ear's time-constant from these studies are nat inconsistentwith the values observedFromthe pure tone testS. Additional support For the time constant valuesdiscussed in Section 3.1.3 is provided by data by Dubrovskil and Tamarklna 54 on"subjectiveperceptionof the relative loudness of amplltude-madu/oted noise. They hypothesizea tlme-constantfar the ear of 10 ms to explain their data a value slmlfar to the 13 mscited earlier for testson the modulation thresholdof pure tones. l_e appllcatlon of a "tlme-constant" modelagain appearsconvenient to explain experimental results. However, this device mayindeed be misleading based on the unique results and resultlng hypothesisposed by Miller 30 in his study of the delay in detectability of a low level noisesignal fallowing the interruption of a higher level masking no_se. Basedon his results, Miller suggests that: '_ .... the auditory system as a whole doesnot have a fixed rate of decay of sa many decibelsper second independent of intensity. _us the auditory systemcannotbe said to have a "time constant" in the sensethat this term is generall); used,and we have been careful to use the term "critical duration" instead. Thisis not to say that the mechanismof the ear hasno time constant, however. As in air mechanical systems there is a finite time required for the ossicular chain and the cochlear fluids to begin and to stop their motions. Themechanical time constants of this system_ however, are far too small toaccount for the 65 msecperiods of perceptual growthand decay. 'It hasoften been convenient to liken theauditory systemto an integrating circuit ..... The evidence seemsto showthat the ear is not sO much an integrating device as it is a delaying device ..... According to our

hypothesis, the growth of the perception of na_seis the integral of the

3 -20

.J

distribution of transm_sslon t_mesof the various pathways from the cochlea to the higher center, and not the integral of the sound intensity. " 30 3, 1.5 LoudnessVersusNoisiness of Impulsive Noise None of the preceding studies cited on noise burstsemployed standard loudnesscalculation proceduresto predict thek results. However1 bath Pollack36 and 4/ . .. Garrett usedd_fferent ernp_rJcalapproaches basedon we_ghtlng theFr no_seburst signals by a Function related to the observed level-duratlon trade-off. Garrett was

particularly successful _npred_cHngthe loudnessof 48 complex transient signals consisting of repeated decaying s_nusolds. Other examples of this approachare covered in Section 3.1.6 on response to complex impulsive sounds. However_before consldeHng more complex impulsive sounds, let us examine one final (and very recent) study on the loudnessand noisiness of no_sebursts. A new and unique approach to the predlctlon of human responseto _mpuls_ve noise is provided by the work of Izuml. 82 in order to examine the passible subjective

difference between loudness and noisiness, IzumJ conducted a set of two laboratory controlled psychophyslcol experiments. In the Firstexperimenta periodically intermltlent pink noise signal wasused to determine if there was indeedany difference between these two subiectlve parameters. Upon finding a significant difference, the secondexperimentwasconductedsothat an effective assessment methodcould be established. For the first experiment, consistingof two phases,subjects wereasked to ompare_ using the paired cornpaHson method, the test signal (intermittent pink noise)with a standardsignal (continuous pink noise at 70 dBA). "111 signa e lswere presented_na fade-in, Fade-out sequence as shownin Figure 12. In order to overcomeany passlble error due to sequenceblast the stimuli were presentedboth signal first and standard Iirst for an equal numberof times. The whole procedure was repeated for six different burst-tlme fracHons(Bl1:); the BTFbeing defined as the signal-on time divided by the on-plus-off time.

3-21

[
0

I
3

I
5

Ill
t31e,_[*

I
t?

JI
2S_G

I
30 (_ec)

Figure 12. Time-Amplitude SequenceDiagram of the Stimulus Presentation (from Izumi82 ) During PhaseT, the subjectscomparedthe palr of signals in termsof their relative loudness,i.e., how much louder (or softer) than the continuous signal is the intermittent signal? During Phasel] the samesignalswere replayed, but th_stime the subjectscomparedthemin termsof their relative noisiness. The resultsof bath phases were tabulatedand comparedwith each other (see Figure13). From theseresults, Izumi concludedthat " .... as far as periodically

intermittentsounds are concerned, laudness judgments and noTslness judgments are signiflcantly and systematicallydifferent. consideredas different attributes." Once he determinedthatthe two parameters are indeeddifferent, izumi .-at up hissecondexperiment in order to arrive at a mode[which wouldaccurately predict thenoisiness of an intermittent signal. in thisexperiment the subjectswere presentedwith signalswith 25 different BTFs. They were asEed after each trial to rate the relative noisineso s f the signals as in the Firstexperiment. Therefore, Ioudnessandnoislnessshal]ba

3-22

f
--

LEFT COL.:"LOUDNESS" RIGHT COL. :" NOISINESS"

lo--

_
' --

N 5-- ,,
"_m

--

16/250 31/500

6311000 63011000 250/1000 95011000 TIME PATTERN

Figure 13. Result_ of Experiment]. (_ki=a (LoudnessD ) ata and Yahomashisa (Noisiness)Data are Comparatively Plotted. Filled Circles RepresentMean Relative BurstLevels Judged by EachSubject. Averages and StandardDeviations are Shownby Cenh'ol Lines and Rectangles on BothSides (from lzumi 82) Model Fromthe resufh_ of those triafs, Izumi developedwhot he calls the "Perceived NoisinessModeJof Periodically Intermittent Sounds 75-A."

LRB=61OgloBTF+(IOIogIoRR+

10)(1 -e "15TfF)

, dB

(,5)

3 -23
p_ ,_ .... ......... ._ .......... .,, .... ........ . ...... .

where LRB= relative A-welghted noise level of burst in dB BTF= burst time fractlon_ i.e., on-tlme/on + off time RR= repetition rate per second Totf = off time in seconds In order to test this formufar he predicted the value of LRBfor the 25 intermfftent noisesused in Experiment I]. The LR_'swere colculoted using nine different methods: peak burst levels in terms of Loudness Level_ StevensLL(S); Loudness Levelt Zwlcker LL(Z); Perceived Noise Levelt PNL; and A-weighted noise level; A-weighted equivalent sound level; Pollock's36 47 method; Garrett s method noise rating number (NRN) as specified by ISO1 la and Model 75-A, proposedby izumi. 82 The predicted levels were thencomparedwith the experlmentoldata. The results are shownin Figure 14.* From these results, lzumi_s Model 75-A appearsto be the best predictor. The other methods always underestimatethe perceived noisiness of the intermittent sounds, Startle Effect The major reason, according to Izuml, for the difference betweenloudness and noisinessis the startle effect created by the intermittence of the sound. The startle effect is basedon three physical parameters of the signal: repetition rate, rise time and the burst-to-backgroundratio. In theseexperimentsthe rise time and the burst-to-background ratio wereheld constantand only the repetition rate was varied. The contributlon of repetition rate to the nolslness-loudness difference was q_ntified end this informotion_shown in Figure 13e_ was usedin the developmentof Model 75-A. Accordingto lzumi_ work is still necessaryif the contribution of the startle effect is to be understood.

*Figure 14 is a corrected versionoF the form publishedin Reference82 which wes kindly suppliedby Dr. lzuml.

3-24

It will be pointedout later that one study of subjective response to helicopter blade slap as o function of rate of slop23 hasalsoshowna trendoF increasingapparent noisiness with increasingrepetition rate although the range of "pulse rate" exploredwas well abovethat (i.e., 10 to 30 pps) explored by'IzumL

,,,

I , I

I I r

O" NOISINESS"

i=

I,, LOUDr4ESS"

.J

.J

j,_ 5
m

i,iJ pw

O O._lllfl

2 I

4 i

, _ I_ 8

REPETITIONRATE, PULSES PERSECOND

Figure 13a. Comparison of Loudness and Noisiness vers_, Papefition Pate for BurstTime Froctmnof 0.063 (from [zumP'_),

3-25

10

rn '13 c -J o D
"1

O_

-5--

t o _J -10 -I-Q t_ ee n. -15 -

-20

'

i
Figure 14. PreliminaryValidation of A=essment/v_tho_s. Errors of Prediction are Calculated for 25 Intarmlttant Noises in Experiment It. Moan Errors ore Shown by Central Linesand StandardDeviationsby Reatangleson BothSides(from lzumi82). In summary,although this is only one study, Izumi shows quite well that nots;heSS and loudness are nat the samesubjectivequantitieswhen dealingwith interrn|ttont sounds,and that the startle effect of the intarrn_ttant sound is a prime cause of this difference.

3-26

3.1.6

Subjective Response to Complex Impulsive Sounds Early work on subjective response to more complex impulsivesounds other than

tone or noiseburstsinvolved rneasuringloudness of short h'iangular transients suchas 44 72 repeated gunblasts. For example, Carter, Carter, and Carter and Dunlop 74 explored the loudness and thresholdlevels of this type of transient, which had a pulse duration of I ms, Far varying rise times (. 05 to 0.5 ms)and repetition rates (1 to 256 pps). Theeffect of repetition rate wasadequately coveredby a simple energy rule (+3 dB increase in _ntensltyto ma_ntaln loudness Foreach halving of duration). For the highest repetition rate, the ratio of on-time to off-time never exceeded 0.S and was typically much less. As with all the preceding _mpulsive noisestudiescited so Far (except Fidell and Pearsons),earphonepresentation was used. The loudness judgments were made by comparisonof a 3 second reference (white noise)signal with two impulses separatedby 1 secondfrom each otherand from the reference noise. For mostof the loudnesstests, the reference noise was Fixedat 15 dB above threshold (sensatZon level of 15 dB) Foreach subject. The prlnc|peJ result from Carter's work is i_e evaluation of alternate means of predicting loudness of triangular impulsivesounds. For each repetition rate and rise time, the loudness of the reference noiseand impulse_ at lf,e '*equally feud" intensity love/s, was catculated from the signal spectra. The spectrawere computedFromthe pressure time history for the irnpu/slvesounds and measureddirect/y for the reference noises. The four calculation methods analyzed were: Zwlcker87, 89 Stevens, Mark V193 Perceived Noise Level103 A-Welghtlng

The average difference between the calculated loudness (basedor, the computed spectrum)of the ;mpulslve sound, which was judged equally as loudas the reference sound, and the calculated _oudness of"the reference sound wasmeasured for all the

3-27

combinationsof rise timeand repetition rates(306 cases). The resultsare summarized in Figure 15 for thesefour methods [n termsof thisdifference as a function of pulse repetition rate.
,+1o i i i i _ i i ._10

.
Lt - LR,
d_ -I I ! I I,.--_L r I

t J
i lie ] f I/ -i

I I I

I(c) l

O 'i I 4I I

" o X
I IS

X I

X _r

o X
1 256 ._

;_ x x.-_ _ x_ x _ _ xo
I l T I 4 i I 16 I fS S ! I6 25

RepetitionRate, pulses per second Figure 15. Comparisonof Loudness Calculation Methodsfor TriangularTransients. Variation about Iheoverall mean(within each loudness method)of the mean difference (aver suhiects)between the calculated Iou:lness (Lt) of the triangularI msimpulse and the calculated loudness (LR)of the reference noisessubjectively judged to be equQIly loud. Deviation fromzero is a direct measure of the error in each loudness calculation method= (o) Zwicker, phons; (b) StevensMark VI, phons; (a) PerceivedNoise Level; (d)A-Weighted Level. The symbols denotevarying rise tlme (e, 0.5 ms;+, 0.25 ms;o, O. 1 ms;and x, 0.05 ms). (FromCarter72) Surprisingly, the loudnesco s mputedon the basisof the A-weighted levels exhibits the least deviation aboutan overall mean. The Zwicker methodwas next in accurecy. There is reasonto doubtthe general applicability of theseresults, however, as shall be seenwhen Iheseloudness calculation methods are applied to other typesof impulsive sounds. Fidell and Pearsons 63 investigatedthe influenceof phase of harmoniccomponents on the judged noisineso s f five different simple transientsounds corresponding to (1) an ideal N wave, (2) an N wave wlth 1 msrise and decay times, (3) a triangular waveform, (4)a squarewaveform,and (5)a doubletor podtlve and negative sharp

3-28

impulse.

Power spectra For each basic waveform were maintained essenHally constant No significant

while phase was adjusted by a computerized waveForm generater. influence of phase on subjectlve loudness was detected.

They also evaluated eight artificial speaker system. difference

the subjective loudness of 12 actual

impulsive soundsand

sounds presented, as were all The characteristics

their signals, over a high quality loudThe

of these 12 sounds are listed on Table 3. objective

between a tlme-lntegrated

measure of the sounds and the same

measure for the reference sound is shown in Figure 16. Table 3 Description of Naturally Occurring Impulslve SoundsEmployed as 64 Comparison Signals in Evaluation Experiment by Fidell and Pearsons
Duration (m_ec) 300 150 425 450 Approximate Spectral Charaaterlatlcl Peaks at 0.5 kHz Near flat spectrum to TO kHz Rhes and foils about 0.8 kHz Highly leptokurHc at 4 kHz

Impulse I 2 3 4

IdentTficaHon Aulornahlfa Door Slam Paper Tearing Hand CIo_ Two Bottles CIInkTn_ Together Chain Col/alvin _ on Itself

580

! Near flat spectrum to I kHz, foils slowly at higher frequencies CompJex spectrum peaked at 0.125 and 2.5 kHz peaks at 0.g, 1.6 and $ kHz

4B0

Noclurnai Animal Noise

180

Squeaky Release of Air Through a Valve Balloon Burstln_ Balloon Burllln_ Automobile Horn

8 9 tO

400 600 180

Peaks at 0.2 kHz Peal_ at 0.2 kHz Discrete frequency peaks cancentrafed between 0.3 and ] kHz Predominantly Jew frequency, falling i_eaply from O. 125 kHz Energ concentrated between 0.2 end 1.6 k_'lz Octavo Band from 0.6 to 1.2 kHz

II

1200

Simulated Sonic Boom

12

900

Basketball Bounce in Highly Baverberanl Envlronrnant While Noise, I Slo/_d

Standard

1000

3-29

:_,_._...... _

-S

I O

8a
0 O0 O0 0 D 0

aa
+5 -

Ooo
[] ---_D'O---

_
__jD_O__ 8

3 +_0

Oo a ___o_ 9__
8o_
o
0 0 0

o
0

8oa
oO
;
0 [] A

o
0

,
+15 +20

o n
-_-----,oo 0013

8
o_

Naturally Occurring impulses Synthetic Impulses Synthefic impulse with Low Frequency Energy Mean of Data

A +25 I Leq(C) I Leq(A) I Leq(N) I EPNL

Figure 16. Comparisonof Time-Integrated Measuresof the Impulsive Noises with the Same Measure of tho Equally Noisy ReferenceSound (1 Sec of Octave Band Noise, '600-1200 Hz). (From Fidell and Pearson64) s

The ordlnate specifies the difference between the overage soundlevel (computed From a mean square averageof the drgltTzed trine history of the signal)64 and the samemeasurefor the equally noisy impulsive sounds. For the A-welghted measure, thTsdifference is identlcal to our subjective correction factor _s and was equal to 12.5 dB with a standard deviation of 3.5 dB. The standarddeviations for the other measures were s)ightly greater, thus indicating the A-weighted average sound level wassl_ght)y more reliable as a predictor ofnoislness of these impulsive sounds. Note that these impulsive soundsvary substantlafly }n their characteristics; somemay not be very impulsive. However, they are all essenP[aJls yingle events and not repetitive. The average

value at" 6 s observed, ]n this case, has considerably more validlty than the values g_ven up to now Forthe fo)low]ng reasons: T. It wasmeasuredw_th a loudspeakerpresentation thus rnsurlngthat realistle head diffraction effects are included. 2. The objective measuremeno t f the average sound level should be very accurate - they wereperformed by dfgltar analysisof"a recording of the actual soundreproduction. 3. The soundscover a variety of actual impulsive noisesto which the subjectscan relate. 4. The instructions to the subjectsasked for a judged nols_nesb sut promptedon annoyanceresponse as well (i.e., the test instructions

defined a noisy soundasannoying, unacceptable, objectionable, and disturbing if heard in the home during the day and nlght). 64 Loudness measurements of decayingslnusoldal transients similar to those used by Gorrett were carried out by Gustafsso7 n8 but at soundlevels from 95 to 117 dB. While the resultstend to substantlatethoseg_ven earlier_ the high noiselevels used place these data outside the area of interest for this study. 3.1.6.1 ]SO RoundRobinTests The most complete set of data onloudness of impulsivenoisesis provided by the final resultsof an international cooperative RoundRobintest programorganized 3-31

'11 '

under the auspices of the International StandardsOrganization, ISO/TC 43/SC-1, StudyGroup B, "Loudnessof Impulsive Sounds." The f_nal rel?ort, prepared by Pedersen, et al_77 represents results from 22 laboratories and "close to 400 subjects. " Add_tionel detailed supporting data were reported by Shipton, Evans, and Robinson, from the National Physical Laboratory,66 on the specific results from their testswith the ISO Round Robin data tapes. Detailed informaHonan findlngs of the ISO RoundRobin Tests, drawn from these two sources, is presentedin Appendix B. Although the tests consistedof an evaluation of subjectiveand objective correction factors for the following three typesof impulsive sounds,results for only the subjective correction factors for the first group are considered here. Group I Nine quasi-steady impulsive noises recorded from ach_al sourcessuchas a teletype, pneumatic hammer, outboard motor. Group If Five single impulsenoises, suchas from a gunor mechanical ram. Group [I1 Six 1 kHz tone pulsesof 5 to 160 msduration.

The sounds were presented to the subjects vie budspeoker in repeatedA-B sequences and matched, in loudness,with reference slgnalspresentedat three soundlevels (55, 751 and 95 dBre 20 /s Pa). The overall grand average subjective correction factorl "s' for all reporting laboratories, nearly 400 subjects,and for the nine repetitive noisesin Group b is 12.5 dB. The standard deviation over the nine average values for each noise is 0.9 dB. Thisis a highly smoothedstatistical result since the variation between subjectsfor any onelevel and test soundcan be l0 to 15 dB. However, it is estimated that the final result is reliable within :kl.5 dB. No estlrnate could be madeof subjective correction Factorsfor the five :.ingle impulsesounds since the equivalent noise levels for these sounds were not available.

3-32
.... =.--

3.1.6.2

Loudness of SonicBooms The evaluation of the loudness of sonicbooms providesadditional information

pertinent to the subjective response to impulsivesounds. Zep/er and Hare|46 successfully predicted the relative loudness of sonic boomsounds by applying a loudness frequencyweighting to the Fourier energyspectrum of thesimulatedN waves. Johnson and Robfnson 55' 60 have extended this type of approachto successfully correlate the annoyance responsefrom explosiveblastsand sonic bandsas well as conventional aircraft saunds on the sameloudness scale. They utilized the S.S. Stevens, /V_rk V|, /oudnesc s alculation method 93 wlth a modification to extend fts low Frequency range to encompass the strong, very low frequencyenergy inherent in sonicbooms. This low frequencydeficiency in the loudness calcu/ation methods has been observed by others.64' 67 Howevert this may nor be a significant problemForthe type of impulsive sourcesof concern in this report. A key element in Johnson and Robinson's approachis the use of a specific 70 ms integrationtime for measuringthe signalspectrum, Thiswas intendedto duplicate the ear's integrationHme.60 Note that this is twice the value of"the rlme-constant specifiedfor the impulse precisionsoundlevel meter. This is obviouslya critical point that will require careful considerationin the selection of an optimumimpulsive noise monitoringtechnique. Johnson and Robinson applied a loudness calculation schemeto the prediction of annoyance for impulsivesources. Are these two formsof human response(loudness and annoyance)really synonymous? The answer, based on available data is that they are not nocessorHythe same. Thfspoint is fundamental to describing impulsivenolse and deserves the morecareful review taken up in the next section. 3.2 Annoyanceand Other Sub]ectlveResponses to |repulsive Noise Reviewof the existing literature dealing with annoyancedue to impulsive noise yields a wide range of approaches and result. Theseresultsfrom available studies,

3-33

excluding thoseon helicopter blade slap, are briefly summarizedin Table 4. Annoyance of helicopter blade slap is consideredlater. Annoyancedue to aircraft sonic booms were of primary concernin about haft of the studies cited in Table 4. While moststudies attempted to measure annoyance, the terms "unpleasantnes ''1 s and "unacceptabilily"12 were also used. |t wasassumed that thesetermsrepresenteda similar measureof subjective response. The qualitative descriptor"annoyance" is not well defined but maybe assumed to representan overall subjective reaction to an impulsivenoisestimulus. Thisreaction may very wen integrate nat only the loudness or noisiness sensation but also the response to other non-acoustlc factorssuchas startle, emotional contentor intrusive noise level relative to the ex_stingbackground ambient level. 3.2.1 AnnoyanceResponse to ImpulsiveNoise A division of the referenceson annoyanceresponses into the three categories of impulsivenoisestudiesdefined earlier helped in selecting only thoseapplicable to this effort. Category I, which is the principal concernof this report, covers the "repetitive impulses"produced by two-strokemotorcycles,rock drills, pavement breakers, helicopter blade slap, and other repetitive impulsive noise sources. Category |[, "single impulse," includessonicbooms and artillery blasts. Category I|l, "unsteadynoise," coverstraffic and subsonic aircraft noise and isactually mare concernedwith noise "events"rather than with "impulses." In one sense_however, the Firstand last categoriesore similar, differing basically in the time scaleof and between "events" and in the crest Factor or ratio of maximumpeak pressure to rms pressure In the studiescited, correctionFactors were developed to account Forannoyance on the basisof one or morefeaturesof the impulsivesounds: numberor Frequency of impulses,amplitude, fluctuation (rate of change in amplitude), and durotlon. Some investigatorsproposed correction factorswhich were applicable to impulsivenoisein general, independent of ffs characteristics.11, 1la, 1lb Thus, EldredI 1 and ISO R199611aproposea 5 dBcorrection shouldhe added to any communitynoisewhich is

3 -34

Table 4 A SummaryoF L_teratureon Annoyance Responseto s Impulslve Noise (Excluding Studiesfor Helicopter Blade Slap)

Nalse Saur _o NolSe Type NO, el Subject1 4

Param_eers

Mealurernenfs

Retult ! Irnp_Jfle b Ba_e Scare Ca root;on (

Authol PJ_lchlk L

Date 1957

lemt Tone I_ulst

ReFere_l_e tqc*r_

Vazled Freq_rlc_, Repetltloa Rare, dB

CoFis_anr D,Jrotlon

SubieCl]_e Ju*t NaDaeabl_* Urlpteotanrnel*

C)hje_l_Ve Atle_ua_axl _dfi

Relpont_ S_ule

Ropetiti_e Pulses

Kelghle_ .8

1970

1903

Live OfFice

No_

Acceptability

Peak Index (PIJ Avera0e d_.

Average L A L eq

! 3.S3 P[ ]/2 I d8 c

Acnplublllty

Anderson, 10 Bobinwn

1971

24

Recorded Rood Drift

None

Durofionond Nun'ber of Bollts

Bk0d L_vel E_p,_ure Duration

Adject6.e Pair ( I I Paint Compaei_oa)

L eq

14a, dB (d)

Annoyance

EJd red f9961h: ISO'R

]97t

Com unity Na_lm eS

None

Souece. Lecel Sire

Comp mmu nity Co Joln _D

teq

Leq

5dE

Comm nllyNa hv ee_ Sq _.Iwu lo _t Le

Fucfm 13

1972

100

Ra_o_d_d Hondcrnp

Tone Bursl

Hondclop SPL I ttan,_lap Tone B_,_sl Dula t;on DuraHon

Anr_,_y_nce

L A

' L A

Annoyance

If Single

Broadbent, 2 Robinson B_tlky 3

19L.4 I

79

Jab PropA,'C ;Jel _;o nla Beam LI_I _nJc Boom None

Level Source O vet peellure fpsl)

Durallon

Anr_yan_ Raring A nl:oyonce

LpN

LpN

"{e)

Anr_yanca

1_96s i 3_(} 0 I

O _'erplelsure ProDab UIty o1 l_f

Kry_ 7

1970

Sonic Boom

Sublo_ia Jet

U_aep_ab_l_t_

LpN

EPNL

2_ (I 7-_ (_ LPN)' dS

Unacce_hablJlty

Rahor_r I&

1973

i Arlilfer)'_ 5_,rla_e _rost Sonic Boom

)_o_0

Anr_yancn Complalnt_

CNR

EPNL

t0 Jog]0_,

dB

Community R#_1)o n_l

CHABA 2] WG _69

19_5

AililJery, Blasi_, _ni_ l_oom

A n,x_y'_nca

Led n

Lc

0 for C_wel_h_ed l_veJl

Community !lie" Sl3on_a& Structural Vibrm_on

Table 4 (Concluded)

No_ i_lo_le r_'pe IJl Um_dy NoiN No. oF Subjects

Saurce

Pa+ameterl

p_,etal.tom+ nh

Rnlults mpu .e b Bate Scnle Cmrect_on L eq 2.5_ a, dEI

Aulhor i_b[nlon 6, 9

Date J 1969

lntt [raffle. / Aitcrofr

Reference

Varied

Cnnl_nnl

S_Jbiecfive 1/_e;ory

Obincr_e

Regpanle Scale Ann_ ance

PartY,i 2 Pa.y

J972

A ffcmfr

13_eor y

PNL

0 (I_._mrio n)

Acce plab;lily

Fullet, 15 Jt_b[n_n

1973

24

Traffic

None

LA(rnaN)

Event Dumtlon, Fmquency

Annoyance

LNp

Lq

2.56

a. d8

Anr_y_nea

M Jr, t, 14 Mu all llcho Ztmmirman ) I (o) (b) () (d) (e) (1)

1973

352

A _m |t Li]l ve

Norm

J[ _Jrtlon Leve lO , _ and Frequency of [yen h

A ce ti' _i ly Tol m b lllty, ail!ur_:Qr_e, A nno_,ance

L, Leq , NNI, I'Np

Lq

j.(dLt _

)2 dt (g)

Gl. lml Sub ]ectlv_l Reaction IQ i_o_le

e, ated on c_nceptt by Rmenblll_ a,_d St_+e_ ,_ ,'ited In R+lfe_enoe I I. [_ept at noltd, I_lU_] to tub]ectlve P[, Ihe hmk ;n_l_, _ - Tllmpo_ _)n[c Correction /aclae _ in dJ), to Cmtect lot lu_Jct]w 15 und 20 _ _bo_ responle Io irnf_tslv_ rmiw. Leq ;n I mlnu/_ _r_ple.

II Ihe _umof nurdber o[ Impeller which are 5, 10,

Sl_ndgrd _eviel[on. I 16 dD ;n(_oon equa Jly annay[ng al )e t noi_ _td_'t with LpN _ 110 PNdfi.

b_em, wlfh LC -

_ LpN DIIferer_ce in LpN, propaetionol

In PNdB, ben_een Impul_ive sound ond bockgtound r_lle, rltpectlvely.

(g) Comlcllon

to mean _lU_re mt_ o; level Flucluolion,

deemed impulsive, while iSO R19'991 ]b recommends a Fixed10 dBcorr:tion be app)ied to assess the hearing damagerisk of impulsivenoise, in most cases,as noted in Table 4, the varioussubjective correction Factors developed Fromthe referenced studies are added to the measuredL of an impulsive eq noise in order to obtain the effectfve L of a nonimpulslvenoise that will produce eq the sameannoyance. Forone method, however, the correction is not to be addedto L but rather to the EFfectivePerceived Noise Level (EPNL). 7 It might be assumed eq that this correction factor proposedby Krytor could alsObe applied to the L scale. Another correction method(developed by CHABA Working Group o7)

,,,21

rove yes the predic-

e,q

tion of annoyance that may, in partt be due to building vibration induced by impulsivesounds. Since this response is potentially quite important for large amplitude impulsi_.,e sounds and since it is not treated in any of the other studies, it is also included in the table However, as clearly pointedout by the CHABAWorkingGroup, the concept, based on the useof"a C-weighted L without Furthercorrection, was designed to be applieq cabte on)y to single )_igh_ntensityimpu)sive soundswith a peak soundpressure level above 100 to ! 10 dg- well above the peeksoundlevel rangeof concern_n this report. In investigationsor repetitive impulses, Category I, much of theemphasis has been on how the level of the impulsesfluctuates over time, or equivalently, what is the probability density t'uncHonof the impulslva noise. Two correction factorst basedon fluctuationt identified in Table 4, are the peak index (PI), proposed by Ke]ghley 8 for ravingacceptability of office noise, and a measure proportional to the temporal standard deviation! o', proposedby Andersenand Robinson 10 for application to Impulsive noisessuperimposed on steady randombackground noise. The peak index (PI) measuresthe numberof impulses in the sarnpllngperiod at various peaklevels while the standard devlation measures only the rme variation about the mean- the rate of impulses is not accounted for. For invesHgationsof single impulses! Category 11, the numberof impul_ss in 146 7 the samplingperiod and the level of"impulses above the backgroundlevel are the

3"37

main aonsldereHons. A scale which incorporates both types of corrections into one number may be needed Forgeneral application ta Category II type impuJslvesounds. In investigations of unsteadynoise (i. e., Category If! noise), the primary concern appearsto be in the degreeand rate of. noise level fluctuation. Matschat, et

al, 14 use a tlme-averaged measureof the rate of change of level while Robinson_ et a16, 9, ;5 use the average standard tier;arian of" the level. Theseare not unrelated

since the formercan be considered an approximate measure of the mean frequency of the latter. Theduration of fluctuating or repeated s;ngle no;seevents (specifically

aircraft noise) was found to be inconsequentialby Parry and Parry12 contTary to the 7 conclusions of Kryter and others. Although theseconflicting conclusionsabout the effect of duration have never been fully resolved, the currently accepted practice is to assumethat an energy summationof. noke events shouldbe employed in rating their nolse impact. Thus_ForCategory 111 noises, this is equivalent to a rule that the effective noise level will increase directly as the duration (or more exactly as 10 log (duration)). For multiple events of fixed effective duraHon te, the equivalent duroHan correction is 10 log te + 10 lag N, where N is the number of events. To summarizeso far_ the previous studies on impulsive correction factors for a_noyance of other then helicopter blade slap lead to several choices for the Form and magnitude of subjective correction factor. The form varies froma constant value to a variable dependent upon, for example_ the relaHve magnitude, rate of.occurrence, or rate of fluctuaHan of the ;mpuls;vesound. The magnitudeof. the correction factor for annoyance will vary widely according to these concepts covering a range of as much as 30 dB. A more def;niHve evaluaHon of the magnitude of As for annoyance for the four impulsive sourcesof concern for this report is developed in section (4). It is shown that/ basedon several of the conceptssummarizedin Table 4_ the value of. _ varies from O to 28 dB; the latter _8 s value is based on the use of Keighley s peak index concept and is probably too high. More reasonable values of As are shownto fall in the range of 5 to 13 dB.

3-38

3.2.2

H.e!|copterBlade Step Noise Her|copterblade slap isa troublesome;repulsivenoisesource which has

received a great deal of attention as to causes and effects. 81 l_is attention has Focused,mostrecently, uponthe practical problemsassoc;atedwith noise certification of bet;copters, it is, to a large extent, this morerecent work which is briefly reviewed here. Major aspectsof 14studiesinvolving measurements of subjective response to helicopter noiseare summarized ;n Table 5. The first study, by Pearsons, 3a did not considerblade slap per se but only attempted to rate various momentary noisedescriptors as to their accuracy for predicting the relatlve noisiness of helicopter sounds, As indicated in the last columnof Table 5, the PerceivedNoise Level metric appeared to be superior over others. Leverton 1lc madej perhaps, the First attempt to quantify a blade slop correction factor Forhelicopters and found that theA-weighted noise level from nons/applng helicoptershad to be increased4 to 8 dBAabove the A-weighted noise level Fromhelicopters with bladeslap to achieve the sameannoyance in a simulated living room I;stening situation. Thiswould imply an average subjective correction ForannoyanceFrom blade slap of +6 dB. 17 Munch and King found a subjective correction factor to the sound exposure level to predict annoyanceof blade slap that increasedlinearly From +6 dB to +13 dB as the crest level of the recorded he/icopter noise signature;ncreasedfrom 14 to21 dB. Thecorrectionfactor did not increasebeyond 13 for higher crest Jevels. Berry, Rennieand Fuller18 evaluated methods of measuringrelative impulsivene, of blade slapand found the Followingtypical crest levels Forvarying degreesof impulsiveness. Slightly Impulsive Moderately Impulsive Very Impulsive Crest level = 5-10 dB Crest level = 10-15 dB Crest level =" 20 dB

3 -39

Table 5 Summary of RecentStudiesof Helicopter Brade Slap Noise Including Summary oF SubjectiveCorrectlon FactorFor ]mpulslveness

Pararr,eter_ Date h_elt;gaJor (Rel) No of Helicopter Noise, Source Ba_o Scale Poran_l_rl Varied Sul_iec_t

Mea_urnmonh Subjective Ohjec_ive Correction

I_e_ultb Hur_n Relponse

Pear'ant(a)

I Jar_ 1967 (3a)

21

RecordeJ Comperhon w/Jet Noise

RelBrer_e Noise

J_

LC, LA, LN' LPN' DUrl0' Dur_Ct L A

Noillnel_ Comp'_r;lonl

LC* L A, LN ' Lprq ' Durl0' O_rT0 i 4-fi _B (Sublet; t i,te)

Order alApptopl;at_r_ll_ LpN_ LN, LA, L C. Durali_n and pure'tone at_ecllanl _01 ienprave _edlction. d;d

Le_erton

h_r. 1972 ( I Ic )

Reac,lded

L A

Ar, no_ance CornpoHsom

Meat,jre_lb) Compor;lans

Impulsive helicap_erl_ subj0cth.elv 4-8 dB rno_e annovin _ I_n _nbang;ng heficopterl.

Munch _ing and Berry, Rem,Ie. Fullel

1974 (17) Oa_, 1975 (IB)

Recorded

I Ls

Ls Creb_ Level

Extent ol Blade _lap Degree of I rnpahi .'er_

I_ C_e_ L_vel LpN Imp_hivenest

6"13 _B Level) (Varie_ with Cee_t

No tellable

cotrectlan between

annoyance and impuhivene.. lmpuh;venes_ war hal overriding lacier in ]ud_* Lits impubi,_e t [gr_h frequeady ludged tougher, more Irregulae and le. pted;alivl.

20

Rec aedsd

LpN

I"in_ Constraints (_._ed in ]_te_rahl

ManAca,Jsl;at

July 1976 (18o)

12

Sim,_la_ed

LA , LI

LAla'_* I' Impu hK'e_s Level Impuhlver_e_l Ha.

Atlr_oyance Ral,_ A nnoyarc_t

L Colrected) EPNI.

_0

Pitat _tudy showe4 neg_tlve oer_ct[on I Not tta fi_tiaa fly Dgnlflc_nt ), Carr_taHoa increased with degree oF l*ub]ect;_,ely ]_dged ImpuJsi_,_ne., Thcee bate _coles undereltlm=fl annoyanc_ 2 dE.

24

S;mulated

LpN_ EPNL

I to 4 dB

barton

Dec, 76 (19 19)

40

Acoust;cal Simulation wfConllnuau_ and Impulte Noise

LC* LA _ LpN

al Sine Waves

Annorance

LC, LA ' LpN

Appeo_imotely 2 d8

per ngle m _se_ FreqSi uenc _"I a l pu Sine Wa,_e,_ Impulse RepatHian Fee" que,_y _ SPL Continual* and Cretl Factor

 oauoJojajp_,oi

o_..,; .pem!

o_Uald_mq_!q

. eldwmjoj_p_=l_tpe peole t];m t uo_ a_O_]Nd]_,a) u_!6 lue_e_beu V ._1_ Jpldo_!le_l leeJ F_o pwolnUJll ue_q d_ *_ mop e_ _d_ello I_QdeJ rlql 41_uoJIs doll _palq el_ o_ P_IPP t! (:lSg) .JeJ_D3 dofs aPOlg. V , t_aldO_fla,.I _ltlr_ulucu pup _lqndwl ue_eg ..j_ doI, *F_lq ON

(e) (p) (_} (q_ (a)

I $1;_auJ #_a:/

le_l '.lop .e._ ilu! ol._._ uo p_oq uol;oeJ_o _ gp ;._,_

_ _ 'V1 p:.uo_o_u V _u_,!ll.d_ul o^!_op_ 'V 1 _7

Joldo_!le H pop_o_o PWOln_ul$ S:_ (/p_) aunt" 'PI*JUJWS

,V 1 .

'Bu!_'_

qn t uJ teld_UaS _

_ Jo i_ue

I=_e'1 _t=J3

da_s _t_

uo!;Olr_._!S p.ne S

(w_)

_plwtl_

_ltdo_pH p_!pp) uoll_e_o_ i o Fe*_u OSt eqL (e)_P 8"Z _9 9] ,N_ 1 e_u_._uuv puuq_o_g _!_ N_t]

uoJl_lt_u_l$ pe'P_o_el 09 gL61 'm_y,p ':1

ea!,l_W!

_=W

'ssmul_!_ P'q_!P

I ({'{_11

,m| gp C

NBIH _a _fl

P*lU_"uo

pc=l:; Ioo_

(:_i

le'_e_ Is_3 u!mu _oq mj Fep_._u e_o_uofl=*_m 3 gP 9-0 _]Nd] e_u D__uv 'lNd] ]Nd_

uo_;ol'_,u!S Pw_Jo_el ;Jm:u!V _ gL61

e_ld 'uo_,_*_r

"_dS "P;,_-V

uml_

Oo!M P_!:l

"_ ._ _e_

:_l uof :_e_m_p uo 'gp _ o_

'e_o_ Bid_uo S 1Nd_ '9 lind e_Ac_t_V PUnS Pellnuqs punos LL61

nl_ tteu_*.l_d_u I

'lNd]

;_o _01 o_ _o_!p_

luelc_!_b_ ; o!lul_ Iqn_ _ s;gl _mo_od ']hJd] ZL6l

d=uo 3 'ozo W

s_;do3!l_ _ _oJ,%1_1

osuc_tn_ u=,_n H

uo!l_oJ_o 3

,_t

_lqo

_'*!l_[qn 5

p_!_n A _el_,n_o d

c*Jo_ S a_ofl

c*_rloS e_!o N

(jgH}

_0 OBI s_u I

t;_=[q_ S _o
F_

(P_pnl:'uo3)

_ elqoi

Thelr data relating crest level and judged relative degree of irnpulslveness are campared in Figure 17 wlth slmilar data from keverton76 and Munc h and King. 17 A consistent trend is apparent indlcatlng crest level is 9 reasonablygood predictor of 18 relative impulsiveness. However, Berry, et al, found quite a different story when they attempted to predict relative annoyance of blade slap nolsewith the sameob]ectlve measure. Figure 18shows the relative rank order ratlng of judged annoyance of two groups of helicopter noises with varying degreesof objectlvely and suhjectlvely observed'_bladeslap." Theexperimenters found that there wasnot a reliable cor-

relation between them; in fact, the subjectsseemedmore responsiveto the "roughness" quality of the soundthan to blade slap per seas a measureof its annoyance. As was polntod out earlier, Relahardtand Niese observeda slmilar problemrelative to subjective responseto repetitive impulsive sounds. 61 Thus, according to _is Iimlted set of data, crest level ma't not be a reliable loredicterfor rating annoyance of impulslve noises. Mabry, et el, lga measuredthe relative annoyanceoFsimulated and recorded real helicopter soundswith varylng degreesof blade slap in a laboratory setting and found that duration corrected noise level (using the Perceived Noise Level or A-'welgh ed noise metrics) correctly measured the annoyance response with little or no additional correction requlred for blade slap. However, simulated helicopter sounds with sub.iectlvely judged "light," "moderates " and "heavy" blade slap were about 1, 2, and 4 d8 more

annoying, respectively, in terms of EPNL values, than a ret'erencenonslapplng helicopter simulation. Lawton, 19 in an extensive laboratory investigation of continuous nolsesand s'mulated hehcopter seundssfound that Perceived Noise Level, A-welghted level and overall sound level measures oFsimulated blade slap noise all underestimatedthe levels that would produce the sameannoyance asa steady soundby about2 dB. Patterson, et el, 22 using real hellcopters, found that no correction factor wasrequired to correct for blade slap when helicopter soundswere measured in termsof time-lntegrated A-weighted level (SoundExposure Level, L) or comparablemetrlcssuch as I:PNL.

3-42

_ 0

I 0

Very Imput=i_

_rl

OD_,

[mpulllva

Mffd

I_--_

Slightly Impulltve

Morgina_

--_'_

0 _.

Munch ond King 17 Levmre_

TM

_'

_^lty t Renn e and Fullel l] *Peak W;thin 100-4_0 ttz Ba._d

N_m

Z_ Z_ z_d_o J l
I0

o J T
20

i dB

f
30

Crest Leve_ (Peak/rms Level),

F_gure 17. Correlation oF Judged Degree of Helicopter Blade Slop Versus Crest Level
I

-5

0 Hover(5 Sounds) Flyover (6 Sounds)

_'

_5 g4 "a

0 [>

-3

43 o
_.2 I
I

=>
-2 0
I I

0 I> Very

SIlghlly Moderately Degree of lmpu[J_veness

Figure 18. Illustration FromTwo Groups oFHelicopter Blade Slap Data That Rank Order of Annoyance Does Not Correlate with Judged Impulsiveness (From Berry, Rennle and Fuller)f8 3 -43

Galloway 23, 24a usingboth recordedand real helicopter sounds,evaluated tentative proposals by the British, Frenchand the U.S.A. to the International Civil

Aviation Authority (ICAO) and the International Standerdizatbn Organization (ISO) for blade slap penally FactorsForhelicopter noisecertification. Theseare intended

to account far subjectlve response to blade slap and were proposedto correct measured EFFectivePerceived Noise Levels of helicopter noise. Theobserved penalty factors For equal annoyance Foreight real helicopter sounds was 4 dB and 2 to 5 dB respectively for two different typesof simulated helicopter sounds. When each of the objective correction methodsproposed by ISO, which provideda measureof" sl,gnal impulsive23 nessonly, were adjusted by Galloway according to the rat._._ oee f blade slap impulses, they predicted the observedsubjective correction factors quite well, Galanter, et al, 24 Foundsubjective correction factors of 4 to 5 dB to equate annoyanceof helicopters with conventional iet aircraft when bothare measuredin termsof EPNL. in other words, the EPNL of thehelicopter sound would have to be about 4 to 5 below that of the CTOL aircraft Forequal annoyance. In more recent studies, Leverton_ et al, 2zlb haveexplored bothblade slap oorrectlon factorsand a potential additional subjectlve correction Factor to account for the pseudO-lmpulslve nature of tall rotor noise. For the former, a correction Factor varying linearly From 0 to 6 dB as "crest level" varies From 11 dB to 20 dB is recommendedto explain results of sub'jectlve testsfor blade slapannoyance. (Crest level, in this case, is measured by the difference behveen the peak level in the 250 Hz octave band and the A-weighted! Slow level. ) 24c An extensive seriesof Field tests by Powell using 90 subjects exposedindoors and outdoorsto two different real hellcoptersand a small fixed wing propeller aircraft demonstratedthat: 1. No significant improvementin nols_ness predictability of EPNL was provided by either an ISO-proposedcorrection factor or an A-welghted crest level correction for impulsiveness. 2. For equal EPNL, the mare impulsiva..helicopterwascons!.stentlyjudged !assnoisy than was the less impulsivqhelicopter (i.e., 6s was negative).

3-44

The latter anomalous result might be attributable to the fact that the subjects were askedto rate relative noisinessinsteadof relative annoyancealthough their instructionsimplied unwantedness, objectionabillty, etc. as measures of noisiness. Based on Powell's data, a blade slap penalty factor to be applied to EPNL would actually be negative. (The actual value was about -2 riB; however, the penalty, or

subjective correction, factor wasabout +2 dB when applied to maximum A-weighted Jevels.) in contrast, the study, by d_Arnbra and Damongeot2 , 4d carried out to validate the latest iSO proposal for computing a blade slap correction Factor, shows a small but finite subjective correction factor for annoyance of 2.8 dBbasedan the average result for 20 flights and 60 subjects. 24e In the studyby Klump and Schmidt, subjective responses to short recorded (17 sac) samples of helicopter sounds,presenled in o laboratory setting, were measured and consistent evidence was foundfor an impulse correction factor, whenapplied to A-welghted levels_ of about +2 dB. 24f For the last studyconsldered_by Sternfeld and Doyle, a subjective carreation factor could only be estimateddue to the uniqueexperimental (methodof adjustment)and data analysis techniques employedso the results are not included in the following summary'. However, the valuesof _ s estimated from their study do agree very well with the average of the otherstudies. Thefindings from thesehelicopter noise studies_which are pertinent'to this report, can be summarizedas follows: The mean observed b)ade slapcorrection or penalty factor (assumed roughly equivalent to lhe subjective correction factor As) was 3.3 dB 2.7 dB for the I1 studies which measuredthls quantity directly. However, three of these11 studiesfound essentiallyo zero or negative correction. The maximum correction for moderateblade Ilap (i.e., crest level of 10 to 15 dB) was about 6 dB. Tha maximum correction for _avereblade slap (i.e., crest level about 20 dB) was 13 dB,

comparable to the valuesmeasured for a variety of nonhellcopter sounds. 3-45

The methadsrecently proposedto abjectively compute a blade slap correction Factor do nat appear to agree consistently with the correction factors measured'subjectively to account for annoyance of blade slap. Galloway showsthat improvedresultsare obtained if same modification is made to account Far variations in the Frequencyof the blade slap. 23 He shows results Fromone series of testsindlcatlng _ can change From s about 2 dB Fora slap repetition rate of 10 Hz to 7 dB for a rate of 30 Hz. This effect may explain part of the wide range in measured correction Factors. This dependency on repetition rates in thls frequency range also suggeststhat the "correction factor" may, _n part, arise Frominherent errors in perceived nolse level computations for signals with significant energy below 50 Hz.

The proposed ob_ect|ve meansForpredicting a subjective correction factor depend on somemeansof measurlng the relative impulslveness. The proposedmethodsvary froma slmple measuremeno t f the crest level of the A-welghted noise level23' 24a to more complexprocedures Involving sampling the detected signal (e.g., instantaneous A-weighted level) at a high rate (,_5000 Hz) and computinga measureof meansquare fluctuation level fromthesesamples.

Finally_ it is desirableto attempt samedegreeof resolutionof the differences in blade slap correction factors that evolved from the various studiessummarir_ed in this section. Any attempt in this direction must first recognizethe substantiald_fferencesin experimentaltechniques involved in the studies. Perhapsmost importantoi_all wasthe variation in signal presentation. It varied from presentotlen to subiects in a laboratory setting of simulated or recordedreal helicopter sounds lasting FOr only a short period or [or a complete flyby t to exposingsubjectsin the

3 -46

Field to actual helicopter flyby nolse. A review of the variousresults seemsto indicate that any "impulse" correction factor may be partially maskedor substantially reduced in real field tests where subjects were exposedto the relatively long duration of the helicopter Flyby. Thus, larger duration corrections which are, _n fact, characteristic of helicopter no_se, may serve to partially maskout the potentially added annoyanceof blade slap. Thus, results of those studieson subiectlve response to helicopter blade slap probably cannot be useddirectly to accurately define the magnitude of a correction factor For impulsive noise alone. So Far, resultshave been presentedon measuredsubjective correction Footers to account for either the relatlveannoyance or loudnessof impulsivesounds. 111e next section attempts to showhow thesepotentially different responses maybe related. 3.2.3 Loudness Versus Annoyance of impulsive Sounds The limited data dealing with comparisonof annoyance versus loudness responsesto _mpuls_ve noisescame fromcontrolled laboratory tests, in an early laboratory stud),, Reese,Kryter, and Stevens B3 found someevTdence, shown in Figure 19, that hlgh frequencies, above 2000 Hz, were somewhatmore annoying than indlcated by their loudness. However, the data are I[mlted and exhibit considerable scatter. Parnell, etal, BB found no such indication in their studies of responseto bandsof" noise. 86 Niese also found no distlnct_on between loudnessand annoyance response Fora wide var[ely of steady-state soundsbut dTdfind a difference between loudness and annoyance when one-third octave bands of no;sowere presentedas impulses. Shepherdand Suthorland 5B found that judged loudness and annoyance responses to simulated sonic boomswere the sameForall casesexcept for the highest values of rise time investigated (i.e., ]0 ms). In this case, a slight decreasein annoyance was noted relative to the

loudness response. Thiscan be interpreted to supportReese, Kryter and Stevens _ data

3-47

"0

2o oo,

t_ -20
III

-30

i o0o Frequencyin Hertz

F;gure 19. Comparison of Noise Levels for EaualAnnoyance VersusEqual Loudness (FromReese, Kryter, and SPevens 83)

3-48

Iq I1."

_nd]caHng h_gh frequencies are more annoying at the same loudness. An increase _n sonic boom rise time would tend to reduce high frequency content and hence reduce annoyance more rapidly than loudness, Rothauser, eta), 4a investigated a panel to recorded typewrlter sounds. bath annoyance and )audness judgments of

They Found that for keystroke rates less than

10 per second, a typewriter noise that was adjusted to be equally loud as a reference wldeband noise with a similar spectrum, had to be decreased in leve_ about 2 dB to be judged equally annoying. for repetitive
second.

This wou)d ]ndlcaha a +2 dB correction

to loudness criteria

impulsive sounds I'ke typewriters at repetiHon rates less than 10 per

Fuch's, !3 in a brief study of response to single handclap sounds, observed that his subjects rated the claps about 5 to 6 dB more annoying than an equa}ly loud tOne burst of comparable duration, To summarize, laboratory data do not clear(y support a significant difference between loudness and annoyance of nonlmpulslve sound=t but there appears to be a consistent indication that there is a small positive difference between the annoyance and loudness of many lypioal impulsive sound=. An annoyance correction of +3 dB to a loudness-based subieotive correction factor appears reasonable Forrepetitive

impulses with o rate less than 10 pps with zero correction at higher repetition rates. Sumn_.ry So farr several possible approaches to the development of a subjective correction factor r,s to be added to Leq to account for annoyance effects have been suggosled: e Computation From previously developed impulsive noise - annoyance correction factors as outl;ned in this section.

3 -49

Estimation from data on loudness of _mpulslve noise in terms of the impulse slgnal parameters such as duty cycle and ratio of pulse amplitude to background noise (see Figure 10).

Application

of the 1SO Round Robinor Fidell and Pearsons _ data to and Appendix B). Stevensr

define As (see Section 3.1.6.1

Application of ex st ng loudness computation methods (i.e., Mark Vi or VIb or Zwlcker)_ possibly modified

for an annoyance

(startle) effect of impulsive noise to compute L_ s. Appllcatlon of the new approach suggested by Izuml (see Section 3.1.5).

To a large extent, the data for subjective correction factors for impulsive noise are based on artifical to certain limltations. I_stening sHuatlons in a laboratory and are thus subject loudness or noisiness

First, subjects who rated "annoyance,"

of impulsive sounds normally did so only while concentrating an the listening task and were not burdened with other stimuli or tasks. Secondly, no objective (e.g., physiological) measures of the subjects * responsewere made. information appears to be available for evaluation of impulsive noise. Nevertheless, sufflc_ent

to provide the basis for a subjective correction Before developing this, however, it is desirable

to briefly outline the other effects of impulsive noise which have not been discussed and which could conceivably influence the selectlon of a subjective correction. 3.2.4 3.2.4.1 Oth.er Subjective Effects of 1rnpuls;ve Noise ]mpulslve Noise and Models For the Hearln9 Process Returning briefly to our conceptual model For hearing illustrated earlier in

Figure 4_ there are other features to this model "related to audition of impulsive noise which have not been mentioned. The significant effect of head diffraction on

modifying the pressure-tlme history on an incident sound field that reaches the ear

3 -50

has been clearly reviewed by Shaw.75 Re/ated modelsfor acoustic resonances, in the external ear, analyzed by Teranlsha i nd Shaw,92 identify the major resonances which w_ll further modify the pressuresignatureh'ansmltted to themlddle ear. The combined transmission response of all of theseelements, including the middle ear, add up to a major Factorwhich shapes the spectrumof the pressuresignal processed in the inner ear. As shown in Figure 20, this influencewill be dominant in the high frequencyrange (above 1000 Hz) where many impulsive noisestend to have their dominantspectral content. Thus, subject-to-subject variation in theseelements of theauditory process will be more significant in consideringmeasurement and evaluation of impulsive noise than is the case for mast other meier no_se sources which tend to have their energy concentratedat low ffec_uencies.

o
Frequency, Hz

o
Frequency, Hz

Figure 20. Typical Transmission Res_0onof se the Outer and Middle Ear (Adapted from Bruel) 146

The simplistla elementsfor the detectlc_n nnd processingof auditory signais, illustrated earlier in Figure 4, do not really represent the moreadvanced approaches to this subject such asrepresentedby the moredetailed studies on auditory detection theory.84, 85, 90, 9'/, 94 Thiswork has _tentlal bearing on the selection of an optimum "time constant"model for application toopllmum methods for measuring impulsive noise. For example, the choice of the some "time constant" for beth buildup and decay of transientsoundsis not necessarilywell-founded by either theoryor observation(e.g., References 30, 56 and 91).

3-51

i..,

,_,_,_.l:_L_._:

` ,_._;_.._,_..

....

....

3.2.4.2

SpeechInterference From Impulsive Sounds Referencematerials for the interference of speech from impulsive soundsare

listed in References 96 to 106. An emplrlcal analysls of speech interference from intermlttent sounds,presented in the EPA "Criteria" document/06" indlcated that for steady and intermittent soundsof the sameEnergyEquivalent Level (keq) the speech interference of intermittent sound could be greater than that for steady sound under certain conditions. A more detailed analysls of speech interference of intermlttent soundsusing ANSI Standardmethods 100 indlcates that intermlttent soundsshould always exhiblt substantially lessspeech interference than a non-_ntermittent sound wlth the sameL . Thls is dues in part, to the eq fact that the ANSI Standard includes a posltlve noise on-tlme correction to the artlculotion index obtolned from a steady-state masking noise. Thus, speechinterference effects do not appear to be the basis for any po_itlve impulslve noise correction factor. 3.2.4.3 SleepInterference From Impulsive Sounds The effects of acoustlc stimulation on sleep depend on several factors10 : 6 1. 2. 3. "[henature of the stimulus. Thestage of sleep. ]nstruationsto the subject and his psychophyslological andmotivational state. 4. Individual differences, e.g., sex, age, physical condition, and psychopathology. Due to the complexnature of the effects of noiseon sleep, noattemptwill be made to elaboratean the sleep interferenae from impulslve sounds. Howevert pertinent materlal on thissubject can be found in References107 to 111. It should only be

3 -52

mentionedthat the current useof a 10 dB penalty for assessing noise exposureat nlght21' may not be enHrely adequate for evaluating nlghttlme exposureto impulsivenoisedue to the potential for greater disturbance to sleep. 3.2.4.4 Healing Loss Due to ImpulsiveSounds Boththe energy prlnclple and TTS 2 (temporary thresholdshift 2 minutes after cessation of noise exposure) have been utilized to derive damagerisk crlteria for impulsivenoiseexposure.121' 130 Any discussion on the divergenceof oplnlon on these two methodsis beyondthe scopeof this report. However, the topic is covered in References 112 to 130. The CHABA damagerisk criterion (1968)119 and its later modifiedverslon 130 are shownin Figure 21 to indlcate the general magnitudeof the acceptable pressure level asa function of impulseduration for a normalincidence condition at a normalized repetition rate. Therefore, the evaluationof hearing damage due to impulsivesounds involves the measurement of the peak soundpressure level and its time history. So Faras is known, no attempt has been madeto relate the type of predictive informationconcerning hearingdamagerisk of impulsive sounds,contained in Figure 21, to nontraumaticresponses such as annoyanceor loudness.

105

3-53

165 -

o o
IN

160 55 -_ABA(1968) _ ] 50 " _ _ "_ MODIFIED CHABA LIMIT _. "_4_._ _OC.' A-DURATION

"_ -,-"

. ,4s_
_" 140 -

.,

-._]_

".,._o4 '
A-DURATION

135 -

\
125 .02 B-Duratlon: Time Difference AD (+EF whena reflection is present)

I
I

m l
500 1000

.05

.1

.2

.5

I0

20

50

100 200

Duration, ms

F|gum 21. The 1968 CHABA Damage-RiskCriterion for Impulsive Noise Exposure (Solid Lines)and a ProposedModlflcaHon (Dashed Lines) br a Nominal Exposure of 100 ImpulsesPer D_y at Normal Incidence. PeakSound Pressure Level is Expressed as a Function of A- or BDuration in the Range25 Microseconds to I Second (Adopted from Reference130)

3-54

4.0

CONCLUSION: SUBJECTIVE CORRECTIONFACTORSFOR EVALUATION OF IMPULSIVE NOISE The approachestoward the development of subjective correction Factorsfor

evaluation of" impulsive noise are reviewed in this section and conclusions are drawn concerning a method to account Forthe difference in subjective responsebetween impulsive and nonlmpulslve sounds, The method is necessarily based on the type of psychoaaoustic responsedata available for impulsive soundsand does not necessarily include other aspects of the subjectlve responseclueto factors such as startle effects or emotional reaction to impulsive sounds. 4. i SubjectiveCorrection Factor As

Thevarious approaches consideredin SecHon3 for the development of" s were basedon: (1) computationfrompreviouslyproposed annoyancecorrection factors; (2) estimationfromdata on loudness response to impulsivenoise;(3) application of the ISO Round Robin 66' 67, 77, F. 64 or lzuml 82 data; or (4) application of idelf-Pearson someform of loudness computation method. Thesecandidate approaches are compared in thissection. 4.1. ! Subjective Correction Factors Basedon Loudness ResponsD e ata For Tone and Noise Bursts The laboratory data on toneand noiseburstscan be usedonly for roughestimatesof the subjective correction factor due to the large difference between the test signalsemployedand the real impulsivesounds of concern here. Nevertheless, based on the Iimlted informationon the impulsive noisesources in Section 2, the following roughestimate for/_s can be made. Theseestimates do not include any consideration of o possibleincrease in annoyanceresponse over loudness response.

4-]

Noise Source Motorcycles Drop Hammers All Sources All Sources

Basis for Correction Repeated ToneBursts (Figure7) RepeatedToneBursts(Figure 7) RepeatedNoise Bursts (Figure 10) RepeatedNoiseBu_ts (Figure 11) Mean of Range

As, dB 4-12 4-6 4-14 3-9 4-10 dB

Attempting to estimate values of _s from thesedata necessarilyinvolves conslderable uncertaintyand seemsto indicate lower valuesthan expected. However, it shouldbe recalled that in one case (Garner31), the observedvalues of loudness for repeated tone bursts were very well predicted by loudness aalaulatlons. 4.1.2 Subjective Correction FactorsBased on Measuredkoudness of Real Impulsive Noise Sources ISG Round RobinData The extensive ISO RoundRobindataon 6s summarized in AppendixB lead toan average value for As of 12.5 :_0.9 dB over all of the nine real impulsivesoundtests. Basedon selecting values of L_ s from the specific ISO sources that relate, appmxlmately, to the sources consideredin this study,the following estimatesare obtained. The values of As are roundodvaluesfrom the ISO data in Table B-2 of Appendix B. ISO ImpulsiveSource Outboard Motor Compressed Air Drill Cement Mill Mechanical Ram This Report Motorcycle RockDrill GarbageCompactor], Drop Hammer [ ......... Estimated As_dB 13 14 11-12

There is little justification Forthis attempt'to pelt-ofF the ISO and the Four spocific sourcesidentified in Section2, sincedifferencesin noise signaturemay be extensive. Thus,a single average number of 12.5 dB for As is considered a representative result from the ISO data applicable to thesources consideredfor this report.

4-2

FidelI-Pearsons

Data

From Figure 16, the average ^ s for the 12 impulslve noise sourceslisted in Table 3 was 12.5 dBwith a standard deviation of 3.5 dB. Most of the 12 sources studies differed substantlally from thoseof concern here so that the direct appllcability of this value to the source in this report is questionable. [zuml Data A_ discussed in Section 3.1.5, lzuml has proposed a method for predicting an effective burst level according to a noisiness response which seemsto agree well with sub]eatlve judgments(see Figure 14). Unfortunately, the parameters required by his predlctlve model defined in Eq.(5) were not available with sufficient accuracy to permit appliaatlon of the model for this report. However, as noted in Figure 14, his

data do showthat the overage difference (As)between the subjectively effective and measuredL for hls 25 intermittent noiseswas 13.5 dB. Thls numbermay be compared eq to the value of 12.5 dB from Fidell and Pearsons. 4.1.3 S.ubje.ctlve Correction FactorsBased on Annoyance Several methods to directly account for the annoyanceeffect of impulsive noisewere outlined in Section 3.2. Todetermine boththe applicability end the valldity

of the variouscorrectionschemes proposed for repetitive impulses- shown earlier in Table 4- corrections were calculated with someof theseprocedures for the four sources of repetitive impulsive noiseof concernfor this reportand for which data were available. Three correction schemestconsidered in Section3.2t were applied to the real impulsive noise sources: (1) Crest Factor (or Crest Levelwhen expressed in decibels); (2) Peak Index; and (3) StandardDeviation. Crest Level Method The first correction scheme, basedon the CrestLevel (C.L.)t has been previouslyproposed to predict a helicopterblade slap(subjective) correctionfactor by Munchand King 17and Galloway23, as discussed eorller in Section3.2.2. Munch and King, As was given by: 4-3 For

From Munchand_King,

As

C.L. - 8 dB for C .L. = 14 - 21 dB 13dBforC.L.>21dB l OdB for C.L. < 14dB 1

(6)

where C.L. =Crest Level = LA (peak)- LA (rms)l dB Galloway's23 results can be used to define two different predictive models for As . The first is basedon only the Crest Level (C.L.) for A-welghted levels. The secondis I_sed on the addition of a pulserepetltlon rate (_0) modifier. Bothof these "models" are basedon the psychoacoustlt c estsconductedby Galloway and on his regression analysis. Acknowledgingthe preliminary natureoF these resultsas polnted out by Galloway, they can be usedto predict values of ,x as Follows: S

(Crest Level only)

AS = -4 + 0.54 (C. L. }

, dB

(7)

(Crest Level + Repetition _ :-5.9 Rate) s where

+0.46 (C.L.) +0.19 (VO) , dB

(8)

C.L. -- Crest Level = LA (peak)- LA (rms), dB

and

u0 = pulse repetition rater Hz.

Basedon Galloway's rather limited data, which covered a range of 13.5 to 16 dBfor C.L. and about 11 to 25 Hz for v0_ hisfirst express_on could be replaced1 for all practical purposes1 bya simple linear equation_As _ C. L. - 11, dBt similar to that of Munch and King. Galloway also points out that the additive correction term for repetition rate is expected to reacha maximumvalue at 30 to 40 Hz and then decrease at higher repetition rates. The Crest Level Forthe impulsive sourcesconsideredin this report wasmeasured in the following manner. The tape-recorded test noisewas fed into an impulseprecision soundlevel meter (B&K2204/5) and the highest value of the A-weighted SLOW response

4-4

2I.

was readto define the maximumrmslevel (LA (rms))of the impulsivesound. The same signal wasmonitored on an oscilloscope and the maximum peak level LA (peak) determTned. The repetition rate wasestimatedfrom oscillographic recordsof the four noisesources(seeTable 1_ Section2.1). The precedingexpressions were then

usedto computevaluesof _ S . No attemptwas made to apply the other _mpuls_vn eoise 24 correctTon methodproposed by ZSO for helicopter noise_ which requiresthat the Aweightednoisesignal be sampled at a rate of 5000 samples per second. Peak index Method The Peak Index correction method, proposed by Keighley 8 for office machine noise, wasappTiedto each of the sampleimpulsive noisesources_ with the exception of the motorcycle. The value of the Peak index (PI) wasderived by examiningthe time historiesof the impulsivesounds on a graphic level recorderset to a writing speedof 125 dB/sacond. The numberof peaksin 1 mlnute_ which were at least 5_ 10, 15_ and 20 dBabave the averagegraphic level readlngl were tabulated and summed. The squareroot of this number1whenmultiplied by the constant' 3.52t givesthe value of As for this scheme. Motorcycle noisewasnot evaluatedwith this methodbecause the time variation of the noiselevel was suchthat true peakswere not registeredby the graphic level recorderat the pen-speed setting used. StandardDeviation/v_ethods The Standard Deviation correction, proposed by Anderson and Robinson i0 for general impulsivenaises_wasalso obtainedusing the graphic recorder_and again motorcycle noisewas excludedbecause of _tsrapid time variation. Using a sampling period lessthan the durationof typical impulses for the othersources, the average levels for up to 100 successive periodswere manually compiled fromthe graphic level recordingsand tabulated in a histogram. Fromthe hlstogram_ the StandardDeviation of the A-welghted level wasthen calculated and multiplied by 4 to give the value of A s as prescribed in Reference10.

4-5

ISO R 1996 Method The ISO R 19961 la correction of 5 dB is considered as a fourth method to be consideredfor predicting _s . Results Table 6 providesa comparison of the results of applying the precedingschemes for predicting As for annoyance. For each impulsive noisesourcet the various values of "s allow comparisonbetween the various methods_ even thoughtheir absolute validity remainsdependent on direct psychophyslcal experiments involving the noises themselves. It hasalready been pointedout in Section 3.2.2, that someof the studieson

helicopter blade slap demonstrated that crest level wasa reliable predictor of subjeo18 18, 24<: tlvely judged impulsiveness but on unreliable predictor of annoyance. Hence, values of As basedon this parameter alone maynot be reliable. However,

when repetition rate is included, Galloway's data showo substantial improvement in the abll fly to predict a value of "s in agreementwith the observedvalue. His correlation coefficient increasedfrom 0.42 (Ls _ predicted by crest level only) to 0.88 when the repetition rote correction wasoddedr thus indicating the potential significance of this parameter for subjectiveresponse to impulsive sounds. The fact that the Peak Index correction schemenecessitatesa 1 minute sample may make it inapplicable for manypassby or intermittent impulsivesounds. A similar correction basedon a ]0 ta 20 secondsample maybe mare practical in such cases. However, in the absenceaf any other supporting data, this methodfor predicting Ls _ is not consideredfurther in this report. For the remaining methods for predicting e, st basedon an annoyancerespanse_ the valuesof "s ranged from0 to 13 dB. The overage over all the four sources and the three remaining prediction methods (i.e._ Crest Level methodwith ar without repetition rate adjustments, the StandardDeviationMethodt and the ISO R 1996 method) Zs 7.2 dB.

4-6

Table 6 Comparisonof Several Predicted Subjective Correction Factors for Annoyance Applied to the Four Impulsive Noise Sources Impulslve Noise Source Correction Method Crest Level - Value, dg(d)' Repetition Rate-Value, Hz(d) Drop Hammer 30 0.7 Truck-Mounted Garbage Compactor 19 0.2 Rack Drills 19 ]

Motorcycle 13 10-40 (a)

Munch & King]7 (Eq. 6)

0 3 2-8

13 12 (b) 8

i1 6 3

11 6 3

A s, dB Galloway 23 (Eq. 7) I Galloway (Eq. 8i

Peak Index 8 Value _s' dB StandardDeviation 10 Value, dB 6s, dB 150 R !9961la As, dB

- (c) -

63 28

21 16

2.97 12

1.26 5

1.09 4

(a) Assumed maximum repetition rute of 10-40 Hz for purposes of estlmaHng maximumAS according to Eq. 8. (b) Computedvalue beyondrange of Crest Level ForGalloway's data. + (c) Data not available Forcomputing"s" (cl) EstimatedFrccn data in Table I, page 2-9.

4-7

4.1.4

Summary of MethodsForComputingthe SubjectiveCorrection Factor 6 s

The values far As derived Frommostof the preceding methodsare summarized in the Following table. It was feasible to break down the comparisonby the four noise sourcesonly Formethodsbasedon annoyance. Table 7 Summaryof SubjectiveCorrection Factor (As) Estimatedfrom Existing Methods or Data, dB

Annoyance(a) Impullive NolJe Source Motorcycle Drop Hammer Truck-Mounted Crelt Level(c) I Crelt(b) Plul Level Rel_tition RoII 0-3 12-13 6-11 2-8 8 3 12 5 Standard Deviation

NolIinIIII FidollPoorlons ]zuml I

_udnell Tonal Nolle lSO Round I_Oln

I
12.5 *3.5

13.5 5.5

7 :k2.5

]2.5 :1:0.9

R ack Drill GarbageCom pactor Average

6-11

3 6.7,4.1

4 13.0 7

I 12.5

(a) See Table 6. (b) Based on Munch and King (Eq 6) and Galloway (Eq 7). (c) Ba_ed on Galloway, Eq 8).

The average values ForAs summarized in Table 7 seem to Fall into two groups. Theaverage valuesof As, basedon the methods which involve direct measurement of noislness or loudness response with real impulsivenoise sources in a laboratory setting t are essentiallyiclentlcol(i.e., As = 12.5 to 13.5 dB). In conh'ast,the predicted

values of As Forreal impulsivesound basedon annoyance criteria or measuredvalues of tone or noiseburstsare lower (about 7 dB). The data in Table 7 do not provide the basisfor an unequivacalchoice Fora meansof predictinga subjective correction factor. Howevert the lowergroupof values observedfor the annoyanceresponse hasone basic point in their Favor-

4-8

the valuesare generally based on more realistic test data in termsof the test signals. For example, the results from the helicopter tests in many casesstem [ram real flyovers for which the observedvalue of a s was generally low or, in at least one case (Pawel124c)t actually negative. Thus, the actual lemporal setting of the impulsive noise signal may tend tcJ decrease the observedvalue of _s below thatobservedfor quasl-steady state sounds evaluated in a laboratory settin9. In an attempt to resolve the differences in theseoverage results, an effort was made to reexamine the use of Crest Level (Crest Factor in dB) to discriminate between various degreesof impulsivenessand hence,, presumably, annoyance. Thus, discounting, for the momenb the negative result by several investigatorsregarding a relationship between judged impulsiveness and judged annoyance, the ISO RoundRobin data werereviewed to see if such a relationship might be evident. By pooling the

information on the peak and rms values of the ISO impulsive noise samplesnumbers 1-9 from two specific ISO RoundRobin Tests(Shlptan, etal, 66 and Thompsone , tal), 67 it was possible to estimate the Crest Level for these sources. The subjective annoyance correction factor _ s was then plotted as a function of this Crest Level. The results are shown in Figure 22 along with the estimated value of L_ basedon the methodsproposed for helicopters by Munch and King 17 and Galloway 23 aSnd Kryter's method. For the latter, it wasassumed that his level, LpN(1) - LpN(b), defined as the perceived noise level of the impulse minus the perceived noise level of the background noise, would be roughly comparable, to a first approximation, to the Crest Level as defined in Figure 22. This comparisonseemsto again indicate that Crest Level alone is not

e valid basisFor predicting As. The potential improvementin a prediction model far as by including repetition rate is certainly an avenue to pursue. In any event, in the absenceof more definitive data, the following concluslonsare drawn concerning on interim method to estimate a subjective correction factor for impulsive noisesources. 1. For the type of impulsive sourcesof concern for this report (this excludes helicopters), a constant subjective correction (as) of +7 dBadded to the true A-welghted equivalent sound level for an impulsive

4-9

14 0o 'o 12 ,._ ,.o I0 .o 8 Munchand King17 t 'tf'_ .1 0 O0 0 0

EstimatedfromKrytor 7

_._"

._._ 4
_f

Galloway23 i_" O ISO RoundRobinData77

2 0

_.t

10

15

20

25

Crest Level, Lpk(C)- Loci(A), d8 Figure 22. Comparison of Measuredand Estimated Valuesof the Subjective Correction Factor A s As a Functionof Crest Level (basedon C-weighted peak level minus A-weighted rmslevel) (Note that Galloway defines Crest Level In totalsof the difference in Peakand m_sA-welghted levels)

4-10

noise sourcewould better define its effective L , that is the L eq eq of an equally annoying nonlmpols[ve reference sound. No additional correction _sidentified at thls time Forthe possible change in 6 s as a function of Crest Level or repetition rate, 1_is first approximation leaves much to be desired in developinga moredlscrlminating correction factor. Indeed, the strong evidence of the potential validity of improvedmethodsfar calculating loudness of impulsivesounds suggests just suchan approach. According to Relchardt,62 impravedaecuracy in predicting loudness of impulsive sounds would be providedby adding a secondary correctian to _ne Stevens or Zwicker loudnesslevel equalto 6[ = LA] - LAS where LA! and LASare the A-welghted "impulse" and "slow" readingstaken on an impulseprecisionsound_evel meter. 55 Alteroatlvely, Johnsonand Robinson have computedloudness directly, usingStevens Mark VI and a 70 ms integratingtime for acquisition of spectral content data Neither of"theseapproacheswere able to take advantageof the latest model(StevensMark VII) for loudness calculation The need to select an optimum loudness calculation method applicable to the type of impulsivesourcesconsidered here leads to the following racommandation. 2. A comparative evaluation shouldbe madeof alternate formsof existing loudness calculation methods based on either the StevensMark V]193 8789, or Zw_cker modelswhen apphed to existing or newdata on subjective responseto impulsive noises. Particular attention shouldbe given to the selection of an optimum time-constantor time-'averaged measures of level for the spectral analysis data required. Alternatively, the methodsproposed h .i. .87 may o_ , r an improvedprocedurefor _; , ,,;umt rre predicting As and shouldbe explored further. it is anticipated that values of _s computedwith suchimprovedmodelswill
show more discrimination as to the magnitude of "s versusoneor more signature charactorhtics of the impulsive sourcesuchas Crest Level or peak to background

noise levels. A valid data basefor computingthe subjective correction factor for any one category of impulsive noise sourcesis required. 11_is data base of ono-third octave spectramusfbe acquired for a sufficient numberof units to ensure a valid sampleof the total Population.

4-11

Finally, the evidence that repetition

rate is potentially significant in the

developmentof valid subjective correction factors leads 1othe Final recommendation: 3. Further research is needed to explore, in moredetail, the significance of repetition rah_on the subjective response to impulsive sounds. This researchshouldalso consider the potential need to extendor refine estimatesoF loudness or r,oisiness contours to lower Frequencies where spectral peaks due to repetition rate may be significant.

Other areasfor improvementin understanding subjective response to impulsive noisealso exist. Theseinclude such areas as developing a better understanding of hearing damagerisk to _mpulsive sound, correlating annoyanceversusloudness or noisiness responses and evaluafing sleep dlsturbanoe due h_ impulsive nolse. Th_s report hasattempted to provide an overview of most oF theseproblems and, hopefully, provide o basisFor practical stepsto be taken now Forevaluating the environmental. impact of impulsive noisesources. Problemsrelated to the objective measurement oF suchsounds are addressed in Appendix A to this report.

4-12

APPENDIX A OBJECTIVEMEASUREMENT OF IMPULSIVE NOISE ' A. I Introduc t_on A measuremen of t impulsivesounds whlch accurately represents their annoying quality hasnlearly presented a major challenge to acousticians. Thisdifficulty is mainly derived from the inability eta piece of"electronic hardwareto faithfully reflect the way the humaneardetects, processes, averagest interprets, storesand finally discardscomplexincoming acoustical s_gnals of w_delyvarying physicalparameters. An impulse precision soundlevel meter (_SLM) which attempts to approachthis ideal in one instrument hasseveralmeter settings:PEAKHOLD, iMPULSE, FASTt and SLOW which can be combinedwith the variousweighting networks:A, B_C, D, and LINEAR. Any one of thesesettingscan be applied to only a limited range of physical parameterl. For impulsive sounds,for instance, the readlng cannot be expected to be wlthln the accuracy limit of the instrumentunlessthe characteristic period of the impulsive sound is substantiallygreater than the overall response time-constantof the electronics for that particular setting. Thus_the taskof monitoring impulsivesoundsinvolves both the problem of finding a procedurewhich will accurately reflect the physical phenomena, as well as the even more difficult problemdiscussed in Section 3 of predicting the subjective response to impulsivesounds. Weshall be concernedin this Appendix with the first task - the measurement problemfor which References131-148 are pertinent. The goal Fore/aluatien of the objective correction Factorwas to define the difference between the true and measured L * for a variely of impulsivesounds. Based eq on laboratoryexperimentation, the A-welghted SLOW meter setting wasselected to most closely approximatethe Laqof impulsivesounds. An objective correction factor is then defined to add toreadingsfrom this metersetting to give the correctedbaseline metric,

*See page 2- 10 In Section2 for definition of Leq.

A-1

viz. A-weighted L . Variousphysical parameters of the impulsive sounds,such as eq the crest level, pulse duration, period, spectrumcontent, and rise time of impulsive sounds are presumablyparametric to the obiectlve correction factor. The difference between the calculated L and the A-SLOW meter readingwill be plottedagainst eq variousimportantphysical parametersof the signals. A.2 Current Stote-of-t.h.a-Artof Impulsive Noise Measurement With laboratorymeasuremen of t impulsivenoise, usingsophisticatedelectronic equipment, a ma._ority of the importantphysical parameters of art impulsivesignal can be studied in detail. "Ibisprovides a moreaccurate evaluationof impulsivenoises

than analysesmade in the Fieldwith simple equipment. However, if on-slta evaluation is required, the measuring instrument mustbe porlable_ compactand easy-ta'-opel'ate. The ImpulseSoundLevel Meter is suchan instrument. Although it is cons_ucted in conformancewith establishedstandards,it may give only a crude assessmeof nt the annoying quality of the noise. A.2.1 LaboratoryMethods Time History

One of the mostpowerfultools in a laborata_ Forinvestigation of transient signalsis the Cathode Ray Tube(CRT)Oscilloscope. With it the time hlstary of the instantaneoussoundpressure can be displayed vlsually anda photographtaken Fora permanentrecord. Such photographs were shownin the malnbody of thisreport (see Figure 2) FornoiseFroma two-strokemotorcycle,a pavementbreaker, a rock drill, onda commercialtruck-mounted garbage compactor. The rise time, amplitude, duration, and period of impulsive noiseare easily read usingthis method. If a significant pure tOne is presentin the nolse, its Frequency can also be estimated. IF the time historyof the detectedlevel is of interest, a high speed graphic level recordercan be employed w;th suitable writing and paperspeedsettingsto measurethe envelope of the rmsor peak value of the instantaneoul s evel. Figure A-1

A-2

:'--""---:'=--_ :L_' --- _ ................

_ ....

: ""- ......

_--'_'---" ZL_ _='="

._ .....

Figure A-1.

(a) Time History of a Two-Sh'okeMotorcycle

::._ :_=]-_-Z_ _- .................

Figure A-1. (b) Time Historyof a Drop Hammer

....-:].-.=

A-3

I,-V

Jo,looduJo3 _lM )lonJ.L qSm.L iol:d_wo_

D jo/_al=lH ew_ (p) * [-V _n_]..I

":--"_-'.-.-:---_ :-:i-:--: :.E-:_:---._-. ................................

.:-::.--_. --. _-"

,_.,::.._ ....

_.:-,

,:......

.:.::.... ,:

:._ .:_ _::,:::_-....... ::. ::_:.;::.:'_-_-_]F-O[::_-,___-._-_----__-.:___-_:_

=

::: - "," :-.

".:'.. : '" _ _ _:-" "_'-: :

" :"-"_----_ :" -, : :L_-_;'-'_-__-_';E,-"_'__ __I

: ,::_i-?:::- -::_:-_ ..................-: ....

: :: ,-::-

showsthe time histories oF the rmsmagnitudeof the four impulsive soundsources mentioned above. For very short duration lmpulslvesounds,an oscilloscape drlven by a log amplifier can also he usedto portray the time history of the signal envelope. SpectrumA nolye.is

Another powerful tool in the laboratory is the Rea_TimeAnalyzer (RTA)which can be used to determine the detailed spectralcontent of impulsivesounds over the audible frequencyrange. Figure A-2 showsthe frequencyspectrum from the four impulsive soundsourcesmentionedabove. However, the spectral ana)ysis measurements of short transient soundsis subiect to appreciableerror unlessdue considerationis given to the transientresponse of the filters and to the use ofan adequate integration time142
constant.

Dic.l[talAnalysis and Computation

A Fourier analyzer coupled to a high-speeddigitizer and an electronic computer provides the mostpowerfui_ state-of-the-art approachForanalysls of impulsivesounds. A.2.2 Field Method SoundLevel Meter (SLM)

Generally, the SLM is designedto conformwith one ar more internatlonallyrecognized standards. _erefore, the built-in specifications for any given SLM will not vary significantly from manufacturerto manufacturer. Thus,an important general observation may be maderegarding the four RC-_ntegrating and averaging tlme-constants in the so-called _lmpulseSoundLevel Meter (ISLM). ,,137 At the "PEAK HOLD" pos_tlon, the RC-_etworkhas a tlme-constantof"50 #s. At settingsof iMPULSE, FAST, and SLOW, the nominaleffective tlme-constantsare 35 ms, 125 ms, and I sac respective y, 136 137 Thesetime-constants, in general, donot include the magneto-mechanlcal inertia effect of the analog i,_dicating devlce which tendsto

A_5

::

T:::

n_::

22:.:22

:-:._n: _n-_zn:.:2:=i:.l:'-,'-

'_ : -:-::_":_-_ I _-/:-_-I ;_:_i'_:_;:_--:_-_-{_!l:t--i-_

i _=:-----,==_:li,tij 2
Cor,i_F_ue_CFI_ I_ ,_31_ 50 63 iO0 I_

='= ff.__2__O _)0 BOO Ik_x

--_*-,__--__-EE2E
li {I 3.1] 4 63 6

=r --"T-. r=l .... _----,_.

I_,._ III ]1._ W

ii=:L-_ E,E

gO0 _

FigureA-2. (a) One-ThlrdOctave Spectrum oFa Two-Strake Motorcycle

__.I_
'_ 2_

1-I_2_--__H--.2-nlnl-:l2" 2_T_ .... _J 2"::--

-=:------1--:In: i =::--.---'2_22 _ 2--

-:-:

-"=!: _-

" ,_-

_1_t-=1_t:t:t_1_-I

_I_-t_:-.I-_I-4:_-I.-._t._F__t__t__l:t

Figure A-2. (b) One-ThlrdOctave Spectrum of"a DropHammer

A-6

-_L

Z-V

pelunow__onJI io]oJ,l_uwoD o jo wnJ_:_e S cg,', l olo 0 pJjl.,ll-au0

(p) '_._ e.lnBj,:l

II!._C] :'l:_o_l o Jo unJ.IoQS d a^o.lo 0 pJ!H..[-_u 0

(o) "_-y eJn6!d

-I

.,

,,,_,

--Ic- F-' -W tq _---t:- ,--,

C_l:: r ,

-_--._-H_-+-_-,-,-, L

-_-,-F" if:_' LF,,

--r r-- r -i-i-....

i
............ )1 I :I

:l L I--I I
r2:r

--j :.ll

............ k--j-

increasethe overall end-to-end tlme-constantsof the ISLM.*

it [spossible to roughly

identi_ the characteristic period of repetitive impulsivesounds accordingto the degree of fluctuation which occurs in the read-out device as a resultof thesedifferent timeconstants. For instance, with a SLOW meter setting and an impulsivesignalof very short duration and period, the amountof signal chargingand discharging throughthe integrating capacitor on the output of the meter detectorwould be minlmal, resulting in a steadyreadingof the ISLM. Foran impulsive signalof very longdurationand perlod_ on the other hand_ the capacitor is fully chargedand discharged during each cycle and a large fluctuation in the meter reading would result. Therefore, from the degree of th-, fluctuation of the reading, a rough idea of the combined durationand period of the impulsecan be roughlyest_meted (see FigureA-3!. In addlt_on to the effect of internal time-constantsof the ISLM_another importantparameter which reduces the accuracy of the ISLM is the crestlevel of the input signal. By carefully adju._tlngthe petition of the ISLM input and c=utpua t ttenuators to ovoid saturating the ampllfler of the ISLM, the readingaccuracy can be Improved. However, the inherent uncertoin|y in the meterreading For maximumcrest level signals that can be handled by the ISLM is approximately 1 dB.141 The orientation of the ISLM with respect to Ihe impulsivesound source, the distancefrom it, and the general physicalenvironmentsurrounding thesound sourcewill also influence the reading obtained from the ISLM. Spectrum Analysis

An octaveband filter can be used;n conjunctionwith the iSLM to determine the approximatefrequency distributionof an impulsive noise; however, the accuracy is necessarily limited by the transientresponse characteristics of the Filter.

*A decoytime constantof 3 sac is provided for the ISLM to part;ally compensate for this meter slugglishness.

A-8

""

D = Pulse DuraHon 1"= RepeHtion Period ,_u r_ "3 0 , '10

D

D _ 100n_5

-
-_ . Small Meter Fluctuations I D< T> 'lOOms 1 second

'

T>

Isecond

.,- I.,.l_r=e Meter . J Fluctuations I

I
T< lsecond I Time Scale

I
I

Figure A-3.

Fluctuation or the SoundLevel Metar ReadingsVersusPu and Duration

P^rlod

A-9
_,_L;* =_'_ _,,m,,_.._,_,_,_._,.,_, _ ;_._.. ___ , _j _ _. . ., _ . ....... ......... ...... . .......... ,, _ ._%,..

A.3

Experimenta Pl rocedur.._ee in order to systematically investigate the output of an Zmpulse SoundLevel

Meter due to impulsivesignals, a wide range of synthesizedsignalswasusedto cover the three regions of meter fluctuation. Readlngs. fluctuated the least, of course, for Ihe SLOW meter setting. The moststeadyconditionswere obtained Forimpulsedurations lessthan 100 msand periods/essthan 1000 ms (see Figure A-3). Thus, the SLOW

meter setting wasusedto define an objective correction factor, Ao, for impulsive sounds as follows: A = Leq - LAS (A-I)

where LAS is taken to be the maximumreading of the fluctuaHng meter needle'and L is the equivalent soundlevel based on the duly cycle and soundlevels of the teneeq burstand backgroundnoise. The variation of"Ao with respect to variousphysical parameters of impulsive signalswas thenexamined. Since the A-welghted equivalent sound level (Leq) wasselected as the baselinemetric, the A-welghtlng network was chosen in conjunction with the SLOW meter setting to read impulsive sounds in order Ii0 m_nimize the variation of the objective correctionfacter as muchas passible. However, the objecHve correctionfactor wasexpected to be meaningfulonly For impulsive signalswhich produced o small meterfluctuation or produceda definitive trend of _.O based on the maximum meterreadingForsignalswith larger fluctuation (seeFigure

A-3).
The reasonForchoosingthe maximum meter reading for signalswith other than smell Fluctuationis basedon the fact thatany Fluctuation of more than 10 dB will be difficult to observewith the someattenuator settingsof the ISLM. Consequent/y,only either the maximum or the minimum reading can be read at any one time, and the maximum value was consideredmuchmoreinformative. The physical parametersof the _mpuJslvs eignals usedin this expprlmentare Iisled in TableA-1. 149 USASI. Three typesof backgroundnoise are used: none, pink and

A-]0

Table A-1 Range of Physical Parameters of the Synthesized Impulsive Signals Used in ThTs Study

Duration ..4-400ms

Period Frequency 2-4000ms 20Hz- 0kHz

Crest Level 15-35dB

Signal-to-Noise Ratiot dB 10-50

A block diagramof the instrumentation is shownin Figure A-4.

The measure-

ment procedure is as fol lows: The level of the background noise is first set to a given SPLas read by the ISLM.* With the background noise off, a continuoussinusoidal signal of a given frequency is similarly set to a different level to provide a given "signalto-noise" ratio. This continuous signal is then changed in temporal pattern only to a tone-burstwlth a preset duration and period which is then superimposed on top of the background noise. Finally, taken. A.4 Results All the observedand computed data have been tabulated in Table A-2. The the combined signal is fed into the ISLM and readingsare

computations for the values in columns 10 and 11 are explained in the footnotesat the end of the table. The master index in Table A-2 is the pulse _ratlon (PD)t given in the first column, which ranges between 0.4. ms to 4.00ms. The next sorting is on the period (T) in the second caluma, which varies from 2 ms to 4. seconds. The duty cycle is not listed, but is equal to 100 (PD)/T, %. It varies from 0.1 percent to 50 percent. The next sorting is on the center Frequencyof the tone burst in the third column1 which ranged from 20 to 10,000 Hz. The Final sortingwas usually on the slgnal-to-nolse ratio, defined in the table, and listed in column 10, which varied from 5 dB to 50 riB. The measured crest level, which is defined as Lpk - LS , varies from 10 dB to 35 dB. The range of these parametersis considered large enough to embrace mostof the impulsive sounds _'The IS.LM(fi;,K Model 2204/S) performedaccording to the manufacturer's specifications on single and repeatedtone bursl_.137

A-11

Counter Frequency I

Generator Noise

G Fe re nq era uetn or cy I ITon G e-B ene urr a sto t r _

t I l

RayTube Ca (CR thT) ode

SoundLevel I .... MeI te m rp(I uS ls LM e ) 1

Figure A-4.

Block Diagramof the Instrumentation Usedin Generating and Measuring Objective Correction Factors for ArtiFiclal ImloulslveSounds

which are of particular interest. Theobjective correction factor_o = Leq - LAS' is listed in the fourthcolumn. In studying the table, several pointsare of interest. Firsb note that when LAF starts to fluctuate for a particular pulse duroHonand parlod, LASremainssteady,* The second point to be noted is that whenboth the duration and period become longer, LAS starts to fluctuate also. The greatest fluctuation in LASoccurswhen the pulse duration (PD) is on the order of 100 msand the period exceeds2 seconds, impulsive signalsForwhich the duration isover 500 msand the period isover severalseconds have not been included in the measurements since no real sounds which were analyzed fall into thisrange. In FigureA-5 the objective correction facter, 4o = Leq - LAS, hasbeen plotted for a constantFrequency and pulse duration against the measured crest factor Lpk " LS for several values of the dutycycle. Thecorrection factor remained nearly constant in the range of 0 to +2 dB. The average objective correction factor is 0.78 dBwith a standarddeviation of 0.45 dB. The correction facter is plotted against S/N in Figure A-6. The scatteringof the data Poinls is small_but no definitive trend

with varyFngparameterswas observed. The meanand standarddeviation of _o is given on the figure, in the plot of 4 versus frequency (FigureA-7)_ the datascatter has increasedbut still no definitive trendresulted. FromFiguresA-8 _oA-13, the objective correction factor has been plottedversusperiod (T) (for constantpulseduration), pulse duration (PD) (for constantperlod)t and duty cycle, for impulsivesignalswith little or no fluctuation in the SLM (FiguresA-8 to A-10) and for signals which cause substantlal fluctuation in the SLM reading (FiguresA-11 to A-13). Forthe latter, the

data are basedon the maximum meter reading. The scatter of the data rangesbet'_veen +2,0 dB to -3.0 dB. A gross downward trend is evident in FiguresA-8, A-9 and A-12. Although this trend is not c/early definedby the data, it would seem ta suggesta significant decreasein the average value of _o (accompaniedby an increasein data scatter) when the pulse duration substantially exceeds 100 msor the period exceeds ! second. *Significant meter fluctuation for any condition is signified in TableA-2 by two values For the SLM reading(i.e., the maxlmum/minlmum reading). A-13

Table Summary air Experimental Data Obtained from

A-2 Laboratory Synthesized Impulslve Sounds

.,
T

2 i

,= I.,.8 11,5.
tO00 *0,4 _9.1

89.5

,13.5

88.8/89.4 "o

1II7.3 r1 7. !
I 89.9/89.5 J I 96,7/96.6 101.6/]0].3 37.3/86.9 83.3/87.7 91.4/9t.3 96.2/95,9 102.0 109.2 117,0 128.6 137.6 fO2.0 109.4 II7.8 128.8 137.5 101.4 117.8 120.7

1,5

2 .7

P,A, 85 85"* g

I I
i _ ! iJ IO

o.1
0.0 -0 0. .5 8 0.5 0.6

920 92.1
102.1 96 8 7.2 0 88.0 91.2 95,4 82.9 86,8 95.2 105.0 115.0 82.5 85.1 92.4 I02.0 112.0 82.3 90.0 90. I 03 .5 ,0_ 1. 0 I ]01.3 8 96 6..7 5 87.5 90.6 95.6 81,2 85.8 94.2 103,9 114,0 82. I 84,0 92,0 ]01.4 111.4 8_.8 89.2 98.1 03,0 10 8., 83.5 86,5 95.2 104.8 114,8 84,5 86.2 93.0 102,4 IT2.3 84,3 90.5 99,2 85,2 ,09.' .

92.2/,2. ,0 92._/,2.,
101.4/]01.2 96.6 8 6.8/a 96 6.4 4 87.7/87.3 90.8/90.5 95.6/95,4

65
16.5 t1 ].. 5 5 6.5 11.5 16.5 t0 20 30 40 50 10 20 30 40 50 10 30 40 20 6 0 ]8.5 22.7 22.6 23.B 22.0 17.5 23.2 24,8 26.4 25.2 17.1 27.3 29.5 24.4 .0 20 U.A.

_ ]00

-0.2 1.7 1.0 1.0 I.I

82.1 d6

t 200

1,0 0.4 0,8 0,4 0.6

t 500

0.6 0,5 0.0

| . I

iI 1 I.

0 0.5 -0.,

t3 07. 9.5 2 ,

Table A-2 (Continued)

(4)
PD(ml) (11 I T(ml) (2) 1000 F(Hz) (3) 1000 I J / leq'LAs 0.7 0.9 0.8 leq 82.2 82.9 86.8 LAS (5) 81.5 82.0 86.0 LS(6) 04.3 04.5 87.8 LAF(7) LAI(B) Lpkl9) 101.3 109.2 117.6 3/N (10) 10 20 30 C.k. (11) 12.0 24.7 29.8 Ramark U.A. 82.1 dB

v 4 2

1
500

1.2 1 .6 0.5 0.7 0.8 0.9

9 5 .2 1 0 3 .0 89.1 92.8 97.8 : 102.1 86.3 88.0 91.2 95.4 81.8 92.2 10] .0 111.8 81.5 34,7 92.2 101.8 02.5

94.0 103. 4 08.6 92.1 96.5 101.2 85.7 87.4 90,3 95.1 03.8 91.6 101.4 111.8 31.0/00.1 04.5/79.0 92.5/<90 102.5/<100 82.5/82.0

95 10 4.0 .2 83.8/03.4 92.2/9 .9 ! 96.6/96.4 102.2/101. l 85.6/85.9 87.5/87.2 90.4/90.1 95.0/95.2 07.6 95.4 105.2 115.2 84.4/33,5 88,5/82.5 96.8/<90 ! 106.5 85.5/04.0 89.0/88.7 92.4/92.1 96.8/96.6 104.4/101.8 86.4/05.9 83.4/00.0 91.5/91.0 95.9/96.1

128.5 6 137.

40 50 4.7 9.7 14.7 19.7 4.7 9.7 14.7 19.7

33.6 33.3

! P.A. 85dB

20

0.5 0.6 0.9 0.3

200

=2.0 40.6 40.4

109.4 118.6 128.5 T37.6 109.4 118.6 128.6 137.4 108.5

8.2 30 40 50 0 30 40 50 20

31.8 23.2 23.3 22.4 25.0 30.1 31.8 30.9 23.0

P.A. 8t.5 _3

t 2000 I

+0.5 +.5/-I.4 +, 2/-5,7 -.3/+,+1 -.7/++0/+.5

5

U.A.

82.1 dE

/
. v T

+3/+80 as o 0,5 8,2 ,0/840

-.3/+92.2 92.5/<-_t 96.5/<90

,,73
127.4

30
40 30.9

Table A-2 (Continued)

PD{ms) (I) 2 4

T(m,) (2) 2000 400

L

F(Hz)(3) 500 500 IgO0 5000 200

I Loq'LAs I -.2/+= -.5/-.4 -.0/-.7 -.9,/-.8 0.4 0.5

Loq(4) t01.8 111.8 115.0 115.5 91.6 101.2

LAS(5) 102.0/<- 112.3/112.2 115.8/115.8 116.4/116.3 91.2 1003

LS(6) 106.3,/< -_ 115.9/0.0 116.0/l]5.9 115.6/115.5 102.0 III.0

LAF (7)

LAI(6)

Lpk(9) 137.4

J S/N (10) I 50 50 50 50 20 30

C.L. (11) 3I.I 21.8 21.5 21.6 8.4 7.4

Romork U.A. 82.1 df i

112.2/109.5 115.6/I]2.9 115.3/113.0

121.4/121.3 125.2/125.0 ]25,7/125.5

1137.7 137.5 137.2 110.4 119.2

I
v 5

I
v 10 j ]

I I
I

r !
I

*
25

o.,
1.0

,,,.,,,0.6
88.3 87.3

,2,.0
97.4

,2e.2
110.2

40
20

6.4
12.8

50 I r
_' 100 l I

I
I I I
I

, .2 1.2
0.9 1 .3 I.l 0.7 3.9
1.3 1.1 3.6 0.7

107.0 105.8

97.2 96.0
36.2 05.0

10 7.2 I 117.1 I
94.7 1 .2 H0 44 .2 80.7/00.306.7/36.4
90.3/90,1 92.6/. 92.4 87.2/a6.909 1/a0.0 93,2/93.5 92.3 lot,7 96.6/96.3

1 ,9.3 128.1
110.3 119 120.3 1

40

50
20

11.0

,2.,
15.6

,
! I
I

94 .1 9 3.1 1 04 .31 32, 9 36.2 80.5

91,2 95.4 84.6 91,3

3o 40 23 22.6
26,5 31.1 20 30

15 10..1 9

80.3 93,2 07.1

93,3 84.3 90.8

110.2 119.2

17.9 17.5

-I.3 400 1000 -.2/.1 0.7 0.6

71.1 77.2 101.3 82.0

72.4 77.4/77.3 T33.3 81.4

82.1 82.3/32.3 111.6 81,7

72.3/72.1 77.3/77.3 81.3/31 .I

77.1 82.9/82.8 37.2/06.6

98.5 98.3 137,4 98,7 . , 40i v

1.6.4 15.8 15,0 17.0

No BNtt'_
I

I _ I

TabJe

A-2

(Continued)

PD (rn_) (|) 5

T (ms)(2) 100 J t

F (H:) (3) 1 Laq " LAS 2000 4000 10,000 0.6 0.6 0.8

Leq(4) 80.2 83.0 79.5 I

LAS(5) 82,6 82.4 78,7

L0(6) 81.6 81.6 81,4

LAF(7) 82.4/82.3 82.2/82.1 78.5/78.3

LAI(8) 87.6 87.4 83.6

Lpk(9) 98.5 98.3 98,1

S/N (10) ,_

C.L. (I l) 16.9 16.7 16,7

Remark NoBN

r
t

2,t50 8
I

/
b

200 T 000

-0,4 0.2 0.6 l.,/l.6

88,2 9 83.3 7,1 11818

88.6 97. 83.7 1 1ll,7/]11,2

99.3 ]Og 90.3 .O 115,1/114,5 1T4.8/-,24.3/123.0

119,4 lt0.O 128 .0 137.6

00 20 40 50

20.1 2 19.7 0.0 22.5 U.A. 82.1 8B

800

tl i_ 10 I

_ 20 J

50O0 10O

-.2/.4 0 9 0. .1 -0.3

115.5 82,8 9.1 9 97,3 102. T 86.2 88,0 91.2

115.7/110.1 88 .2 92.9 97.6 102,4 85.8 87.7 91,1

115.0/114.5

119.4/- 88..3 88.0 93 0//92,8 97.7/97.0 102.4/102.3 85.9/85.6 87.8/87.5 91.1/90.8

128.8/127.3 89. 0/ 83.3 8.7 93 .5 /9 98.2/98. I

137.3

50 20.6 25,6 30.6 35,6 20.6 20.6 30,6

22.3

t P.A. 8I 0d8

I r
I

t j 100

-0.3 0.4 0,3 0. I

102.9/102.8 87.7/07.4 90.6/90,2 98.0/94,7

10C0 : I 20

I t 500 1000 5 000 50

I

|.0 -.7/.I ,5/ -. 4/.4 .6 -.08

95.4 96.1 III,8 T1 115 5.0 .5 86.2

94.4 98,0 112.5/111.7 11 .8/ 115 8, 4/114.6 114.9 87.0

94.5/94/2 118.2/117.2 115.8/115.0 I1 0 .0/11 4.6 11 5.2 /114 .2

103.5/-"
r

99,3/'99.1 111.9/J10,0 126.1/124,0 .7/ 26, ! 128 129.2 /1 127 .7 0 90.0/89,7

137.9 137.9 1 37 4 10 7, .3

30,6 50 50 5 0 50 20,6

19,7 22.1 22.4 22.1

U.A. 82,1 88

116.8/-1 0.2/ 12 20.4 /= -_ i 87.1/86.8

! 200

P.A, 80dB

I tI

-1. -2 .4 7

88.0 91.2

89,7 93.6

89.8 93. 5/ /93,1 8q.4

98.3 94.2/9 /94 8..0 0

25.6 30,6

Table

A-2

(Continued)

LAS(5) 00.3/80.2 80.0/'/9.0

LS (6) 59.0/83.0 97.1/84.0 106.8/9 _ .0 116.8/ID0.0 120.3/100.0

LAF(7)

LAI(8)

Lpk(9) I09.4 II7.8 127.7 137.8

40 30 40 50 50 50 50 50 50 20 30 40

._ 20.4 20.7 20.9 21.0 17.6 10.6 19.6 19.7 19.3 20.3 20.8 21.0

Remark P.A. 01.5 dB

103.0/98.0 113.2/106.2 120.5/122.5 131.6/126.0 132.2/126.6

137.9 137.8 137.4 137.5 107.2 109.3 110.0 127.0

U.A.

82. I dB

100.0/-= 114.5/<100 117.8/~100 11R.7/_100

119.2/,,,100 117.8/_ 100 117.0/_ 100 117.9/,-,100 89.0/83.0 97.2/84.0 106.8/90.0 1_6.7/I I0.0

'P.A. 95.1/94.6 96.6 105.9 115'.8 126.3 92.7/92.0 102.4/101.5 I T2.3/ITI.4 122.3/121.5 _26.3 93.1/92.3 102.4/101.5 It2.3/111.4 t22.4/121.5 li0.5 119.3 12B.9 137.9 110.2 119.2 129.9 107.8 10.9 13.4 13.1 11.6 _ 16.8 16.6 15.4 --

05d_

l LJ.A. 82.1 dB

--

Table

A-2

(ConHnued)

rpD(ms)(1)

T(ms)(2)

! F(Hz)(3)

Leq-LAs

!Leq (4)

LAS{5)

LS (6)

LAF(7)

LAI (8)

Lpk(9)

]IS/N(IO)C.L!

11)

Remark

I i ! 108

j i l I000 , 20

-3.7/+ -3.8/+= -4.0/*= -.7/-I.0

98.1 I08.0 ,18.0 06.2

101.8/< 1 II.8/<

80 90

lO2.1/< 80 ] II.8/< 90 ! I 98 6/88 2 I 94.9/92.9

119.2 128.7 137.7

30 40

17.1 16.9 15.7 f P.A. 85 dB

122.0/2100:122,0/<100 87.5/87.2 !

I
!

;
I

!
I
J

-52
-2.7/+12.2 -2.7/+= -2.8/+=

880
92.4 1]2,0 ,22.8 71.I

!9 14,9817 j 970/988 r
95.9/82.8 I ]89.7

52

59
28 14.5

;
I
I ! t

2000

1000

95.1/80.2 I]4,7/< 124.8/< 73.6/~ = 79,5/-_ 84 5/ 85. .8 /-_ _ 85.5/-= 82.0/-= 88.3/87.8 89 49/8 /85. 87. 40 .5 108 100

J.A. 82.1 d8

-2.8/_= ;03.8 _04.8/<90 184.8/<98

114,7/< 180 124.8 84.7/-_ I 84.5/-= I8 4.5 84. 5/ /-= 84.5/-_ 84.3/-= 180 79.4/-= 85.3/-= /- I 90.3 91.5/ -_ 91.3/-= 87,8/- 91.4/83.0 83.5/77.8 ! 89.5/83.8 ! .5/8 8.5 _ 9 94 5.8 /90.0 95.6/89.8 92.1/86.2 94.8/93.0 =2.5/+-2.3/+ -2. -2.5/ 6/+ + -2.5/+ -2.5/+*= -1.2/.-6 = ,, =. ,

1,9.2
128,8 137.8 99.] 99.2 9 9.2 99.0 99.3 98.8

8I
4( 5_

;4.4
14. T 13,0 14.4 14.7 1 4 ,4 ..7 5 14.8 14.5 I / i NoBN

288 400 1000 2000 4000 10,000 200 1000 _ 2000 F 20

77.2 8 2.0 83.2 83.0 79.5 87.1 88 86. .0 2

52

P.A. 85 ffi i ,i

-I.7/+t.7

94.9 91. 0/8 /84 3. .0 095.1/90.8 96.8/92.0

52 57

. 400

80 I0 v

0.9 /0. o0.2

82 9.l 9 .8

90.0/92.7 93.0

' 95. 9,.4 // 8 6 .8 I 0 8 7. 8

94.6 /92. 8 96. 8/95 .4

52 8 7

I t

Table A-2 (Concluded)

I LASLS'0 0' L f'7LAI Cp 8 : lS, ('O cL,ii,

400
I , f /T]PD: (2)T: (3)p: _> (4)Leq: (5)LAS: (_)L S: (7)LAF: (8)LAI_ (9)Lpk: (10)S/N: (_ I)C.L,:

2oo0
I I t

28
I _'

-T.8/+1.8 87.1 88.9/_,3 92.0/s3.6 94.7/90.9

-2.2/+2.5 -3.0/+3.2 -3.6/+4.0 88.8 86.2 88.0 91.2/86.3 89.2/83.0 J 91.6/84.0 94.B/84.0 92.0/83.0 [ 94.8/83.5 96.8/93.0 94.8/83.0 97.1/85,0

5_
55,4 52 57

P.A. _5d8
I

Pulse duratlont ms Period of the tmpuha tra;n, ms Cantor frequency of tho svntheslzed pulse, H_ Computed A-wolghted uqu;valont continuous sound pressure leeef_ dB

Impulse Sound Levol Meror rogd_ng Qt "SLOW*' mater _ottlng wi_h A-wulghtlr_g notwork, dB ISLM read_ng at "SLOW" v_it8 no wet_hHn8, dfi JSLM read_ng at "PAST" wHh A-woi_htlng network, d8 I_LM read_ng at "IMPULSE" w_rh A-we_pht_ng nerworkt dfi ]SLM eadlng at **PEAKHOLD" v_ith flat walghtlng, dB tone (prier to tone burstlng)t dB

Difference between IhO unwolghtod rml Iove_fsof the backfaunal no'so and tho continuous slnu_idal Cro_t Loyal. Dat';nad as Lpk - LS_ d8 e: Baekgraurld nolso, USASI, A-wolghted (U.A.),

82. I dB (sea Reference 148 for da_c_iptlan of spectrum)

*'i

A virgule separates upper and lower reodlngs from the samo meter $ottlng (P.A.)_ 85 d8

_'**: 8ackgtound no,so, pink, A-welghted

t: + 6- moans that the d_fference lJ moro than 10 c_ l"t_ _ means that the SLM re_ding is too sm_ll to bo teglstero_ for a partlcular sottlng of attenuators

TT'J'_ No background no_s_ (fiN)

6.0 Frequency= 1000 Hz PulseDurat;on = I ms 4.0 o 0 o 2.0 & Duty Cycle O. 1 Percent 0.2 Percent O. 5 Percent 1 Percent

--_

/_

&

Average = .78 dB 0 S.D. = .45 dB c_b

_a_.

.._
I

Ca o

o O

II 0

-2.0

-4.0

-6.0

I 10

I 20 Lpk - LLS(Crest Level), dB

f 30

40

Figure A-5.

Correction Factor_r a ConstantFrequency and PulseDuret;on Versusthe MeasuredCrest Level

A o=

Leq-LAs,dB

00

00

_.

_,

"el c

i_ .,_ II rl

fo'l

00

(D

3 & n 0
C_ C3 UII_O _ 0 "11 0-

g
< S

Q
0 a_

7-

g
-< O I!

' 4 & u a 3 O o 2 o e , a_ 1

' I J Dural'ion(ms) Period(ms) 5 8 10 20 1O0 50 50 20 100 800 1000 2000 20O0 500 1000 200

'

Average =-1.29 dB S.D. = 1.42 dB

_<o
a.-J_ II "I 0 Q

a
I_ E]

-2
O , O 0 0 0 _ 0 O

-3

-4

I
10 FigureA-7.

I
100

I
1000

,
I0,000

Frequency, Hz Correotion Factor VersusPulseFrequencyfor Steady Readings

A-23

'

'

3 2

Average= .05 dB S.D. =IAO dB

EJ
t .,jog iio_ _

Q C _ &A L

-2"

-3

o _ o

-4

Duration(ms) 0.4 1.0 2.0 4.0 5.0 8.0 10.0 20.0 50.0 I P

o c, 8 o

I 10

t I00 T, Period,ms

I 1000

Figure A-8. CorrectionFactorVersus Period farSteadyReadings

Ao=
4_ I I I

Lq - LAS , dB

I_ 0

_>

UI

CO

N O

i2_ Q

O {3. flip a O _I3EI

O O

i
4

i
dB dB

'

Average = 0.29 S.D. =0,86

0 --

_
I
III II

-|

--2

<1

-3 -4

I
0.1

J
1

I
10

I
100

Duty Cycle, Percent F|gure A-IO. CorrectionFactor Versus Duty Cycle for Steody Readings

....

.................

...........

:..

...........

'
4 Average = -1.42 S.D. = 1.21

'
dB dB

'

'

'

'

2
() C)

1 "O

o-Y
:_
II

=
-I <Io -2 , _ o o Dumtlon (ms) 2 10 20 :.50 100 200 400

_
D

_] o

-3

-4

100 Figure A-11.

1000 T, Period, ms

10,000

Correction FactorVersus Period ForFluctuating Readings

I 4 Average =-1.31 S.D. = 1.42

I dB dB

2
Z_

__ 0 -, 1
.,jIl II

_,
ZX [3

-I

&

&

-3

Period (ms) 1000 ,_ o 2000 2500

-4

I
I0

I
PD, Duration, ms

f
I00

i
I000

Figure A-12.

Correction Factor VersusDuration for Fluctuating Readings

, ,

,,

I 4 Average = -1.45 S.D. 3 1.41 dB dB

o-I O" =,, _I I fl -I *

<1 -2

-a
!

f
0.1 FiguraA-13.

I
1.0

_,

I
10

i
100

Duty Cycle, Percent Correction Falor VersusDutyCycle for Fluctuating Readings

A. 5

Conclusions - Objective Measurementof Impulsive Sounds The ISLM readings have been divided inte two categories:thoseobtained From

the s_gnals with a repetition rate greater than I pulseper second(pps)small Fluctuation region and thosewith a repetition rate From 0.4 to 1 pps. The average value and standarddeviationof the ob]ectlve correction FactorForthesetwo regionsore: Average ._ StandardDeviation o +0.1 dB 1.3 dB 1.4riB

When repetition rate _>I pps When repetlt_onrate =0.4

to 1 pps -1.4dB

The objective correction facter hasatso been studied for various temporal parameters of the impulsivesignals, vlz. slgnal-to-nolse ratio, crest level, pulse duration, period and duty cycle. Theaverage value of _o falls within 1.5 dB over the full range examinedForeach of the above parameters. However, except Forthe decreasein _'o for repetitive rates < I pps(period _"1 sec)t no definitive trend in A with any of the other parameters wasevident. It was mentionedpreviously that a broadbandno_se with normallydistributed instantaneous pressures had a crest level of about 10 d_. Th,,st any impulsive signal evaluatedwith an ]SLM musthave a crest level greater than 10dB before it can producemeaningful test data. This, in turn1 impliesdirectly that theaccuracy in reading the meteris limited by the high crest level of the synthetic signals used in the doraacquisition. Manufacturersof impulsesoundlevel metersestimatethis inaccuracy in the meter reading as_1.0 dB For the highest crest leve/s employedin this study. Thesereadingshad to be made on the lower part of the iSLM scale. Any objective correction Factorwhich is not greatly different than this inherentl dB scale reading error cannot be consldered as significant. It is concluded1 therefare_ that the A-weighted "SLOW" metersetting can be usedto measure dlrectly the A-weighted equivalent soundlevel (Leq) to within an accuracy of::k1.5 dB for an impulsivesignal with a repetition rate greater than 0.4 pps. However_ caution must be exercisedan twofactorsconcerningthis conclusion. First, for ImpulseSoundLevel Meters with a

A -30

conventional (-10 to +10 dB) meter scale, it is necessary to use the lower portion of the scale Fordataacquisitionand to usethe maximumreadingof the soundlevel meter for readlng fluctuating levels. Secondly, the conclusions do not necessarilyapply to the latest state-of-the-art SoundLevel Meters which may en_pl0y even mareaccurate impulse measuring characteristicsand digital readouts or true integration features for measuring an equivalent level (Leq) directly. It is anticipated, however, that the latter typeof instrumentswould, in fact, exhibit even lesserror whenmeasuringthe true equivalent level (Leq) of impulsivesounds in terms of theA-weighted, slow rending. Pulserepetition rates lower than 0.4 Hz were not measured in this sludy. However, at this pulserate, the maximumsoundlevel meter reading for each pulsewill tend to approximatethe readingobtained on a single isolatedpulsewith the same characteristics as each of the repetitive pulses. Young and Cohen 144 have shownthat for single cycle sine burstswith burstfrequencies greater than 100 Hz (i.e., pulse durationsless

than 10 ms), the A-welghted sound exposurelevel forsucha pulsecan be obtained quite accurately by the maximum readingon a soundlevel meterset to A-weightingn SLOW. (For lowerpulse frequencies, this soundlevel meter readingwill tend to exceed the true sound exposurelevel reachingo maximum errorof about +8 dB fo_o : single 20 Hz sineburst.) However, the typeof impulsivenoisesources of concernfor this studyare not expected to involve significantsine pulse components as low as this. For example, theone-third octave bandspectraof the ISO singleevent impulsivesouncb shownlater in Figure B-4 of Appendix B have peak frequencies well above 100 Hz. If spectral contentof an impulseis, Tnfact, domlnontat low frequencies (below 100Hz), then, according to the resultsof Youngand Cohen 144 the A-welghted soundexposure level can also be obtainedwithin a maximum error of about 1 dB for pulsefrequencies down to 20 Hz by using the maximumreadingon the C-welghtlng scale. Thus, for the objective correctionfactor, an interimrecommended procedure is as follows:

A-31

................

..................

. ............

..

....

REC OMME NDA TIO N Until moredefinite data are available, the objective correction factor for the measurement of the equivalent (energyaverage) soundlevel of impulsive noise sourcesshall beassumedequal to zero when the L is basedan the eq maximumreading an the A-scale (SLOW) of an Impulse Precislon Sound Level Meter. For single isolated pulses, the corresponding equivalent sound _such single events_ over a time T (seconds)can he approximated by

LAs(e) TO lagN - 10logT

where LAS(e) = the energy meanvalue of the maximum A-welghted (SLOW) noise level over the N events

(A-2)

When the dominant pulse frequency is below 100 Hz, the C-weighting soale should be used insteadof the A-weightlng. This interim procedureis equivalent to setting the ob]ectlve correction factor (the difference between the Leqof the test signal and the Leq of thereference signal for the same instrumentreading) equal to zero. In any event, a correction factor wouldnot have been required at all if soundlevel measuremento s f transient events were obtainedwith a true rmstime-integratlng meter which measuredsoundexposure level.

A-32

APPENDIX B ISO Round Robin Tests Themostcomplete set of data on loudness of impulsive noisesis provided by the results of on international cooperative RoundRobin test programorganized under the auspices of the international Standards Organization, ISO/TC 43/SC-1, Study Group B (Secrotarlat-]5) 23t "The RoundRobin Teston Evaluation of I.oudness Level of _npulslve Nolse." The final report fromthe organizers, O. Juhl Pedersen_ et al, provides summary data froma portion of the results fromover 22 laboratories covering "close to 400 subjects. "77 More detailed results, from the National Physical-Laboratory (NPL), included in the summaryreport, have been reported by Shlpton, Evensand Robinson. 66 Pertinent resultsfrom thesereports are summarizedhere. The test signals employedfor theseround robin tests consistedof the following three groups= Group 1: Nine quasl-steady impulsive noise signalsrecordedfrom practical noiseseurces t e.g., teletypet pneumatic hammer, outboard motor. Each noisesamplehasa duration of approximately ] seaond and is recordedrepeatedly alternating with the reference signal (I//3 octave bandof noiseat 1 kHz). Intultively judged, the noises of this groupforma continuumrangingfrom highly impulsive to almoststeadynoises. (Their relatlve 1/3 octave bandspectraare shown in Figuresg-I to.B-3). Group 11= Five noisesbasically consisting of a single pulse I e.g., from a

gun or a mechan|cal ram. Thesenoisesare recordedas for Group ! with referencesignals(1 kHz tone pulses)of approximately the same duration as the pulse. (Their relative frequency spectraare shownin Figure B-4). Group lll= Six 1 kHz tone pulsesof durationsfrom 5 msto ]60 ms, The reference signalsare ] kHz tone pulsesof durations]0-320 ms.

B-]

Legend N. 6, ...... ....... 4, 8_ 'Source Outboard Motor Air Hammer (Silenced) Mechanical Ram

-,o:i
20

ii

""

N2 4!t :llI}'IlVt ll 'iil I:I

30

'
Figure _-I.

p;lltIlftF l ilI It
Fraquancyin Hertz Relative One-Thlrd Octave Band Spectraof'lSO Round Robin lrnpuls|ve Naise Samples, Numbers 6, 4, 8 (asMemured By N PL) 66

o iot , :i::I111 ;fli ti t1 t 1tl 1II

B-2

F_gUr_ B-2.

One-Thlrd Octave Band SpectraoF]SO Round Robin lmpuislve Noise Samplest, 2, 5 (AsMec_sureb d y NPL) _

B-3

. ,

Legend No. 3 ...... ....... 9 Y Source Teletype (No Cover) Paper Tape Punch Hammer on Anvll

Figure B-3.

Relative On#-Th]rd

Octave Band Spectra of ISO Round Robin Impulsive

Noise Samples 3, 9, 7 (as Measured by NPL} 66

B-4

Legend No_._. Source 14 SimulatedSofllc Boom ....... ....... ....... 0 13 Gun in Free Field 10 Electric Typewriter 12 Explosion Ram 11 Hammeron Copper Sheet Approximate Duration ~ 0.55 Sec ~ 0.05 Sec ,'_ 0.25 Sec ~ 0.25 Sec ~ 0.6 Sec

o .,.t;ll 0 fJf
" " " log " " Figure B-4,

.....

I00

Iflft; ;;1
1"00t_ i

Frequency in Hertz

Relatlw Onn-Thlrd Octave BandSpectra of ISO RoundRobin Impulsive Noise Samples l0 to M {AsMeasuredby NPL) 66

B-5

The sourceof the first 14 impulsivesounds is identified on the preceding Figures. The sounds were presentedin repeated A-B sequencea st 3 sound levels (55n 751 and 95 dBre 20 _ Pa) to the sub._eots uslngt in each eases loudspeakerpresentations in presumablyflee-fleld of nearly free-field conditions. The subjective datawhich will be reported here consistof average values(over subjects)of the difference in settingsof attenuatorsplaced in the test and reference signalchannels(i.e,t attenua-

tion of test signalminusattenuation for reference signal)required to achieve equal loudness betweenthe two signals. This "Equal Loudness Attenuation" for the subject tests (called ELAsubj by Pederson) provideda basic raw measurement of the relative subjective loudness for each of the test sounds. In order to determine o subjective correction factor As Fromthesetests_it was necessaryto utilize the addfflonal detailed data from Shipton_et a l_66 to correct theseattenuator settings for the additional relative difference in the test signalsbefore any relative attenuation wasapplled. Thus_as illustrated in Figure B-5, an additional small dev_atlonAt accountsfor the difference 3nL of the reference signal and the test signal before the additional eq (EkAsubj) attenuation is applied. Thustas illustrated in the figure below_ As can be defined by As = EkAsubj+ t,t Test Signal
/

(B-t)

Refe re n ce Signal -LR "T---L

....

LR = Leq of Reference Signal

l Leq dB As

"T'E__u b .......

---

Lt =Test Level BeforeAttenuation . Lt =Test Level AFterAttenuaHon (Equal Loudness to LR)

F=gureB-5, Computationof the subjectiveCorrection Factor _ s Fromthe ISO Data. 77

B-6

The valuesfor 6 t were computed from the detailed data on the reference and unattenuated test signal levels in Table 2 of Reference66. It was assumed that these data apply universally to the ISO average values for ELAsuhjfor the corresponding sounds. In other wordst it was necessaryto assumethat the relative unatter_uated signal levels fromone nolso to another were essentially fixed on the RoundRobin tapes and would be reflected in identical variations in each laboratory. Thls clearly is an

approximation but is not considered unreasonableconsldering the expected care each laboratory would take to provide a "fiat" reproduction of the (uniform) test tapes provided by IS0 to each laboratory. Table B-1 summarizes the reported values of ELAsubl from the ISO report for the nine Group ! sounds. Table B-1 [SO RoundRobln Comparisons for EkAsubj (dB)

Parameters Mean Std. Dev.

I 12.2 3.5

2 8.9 3.5

3 7.0 4.9

Sounds (Group i) 4 5 6 7.5 3.9 8.4 3.7 11.5 3.2

7 8.2 3.3

8 8.7 4.1

9 11.4 3.5

Average 9.31

Std. Dev. 1.90

The aomputatlons for At from the NPL data and the corresponding values for As are given in Table B-2. The overall grandoverage of_ s (including oil laboratories in the ISO ffgures_ all 3 levelss nearly 400 subjeotst and for the 9 Group l Impulsive noises) is 12.5 dBo Thestandarddeviation over the 9 average values for eaah noise is 0.9 dB. It must be recognized 0 of courses that thls is a highly smoothed statistical result fort as pointed out in the ISO reporb variation in ELAsub]valuesfrom subject to sub]eot for 77 any one level and test soundcan be 10 to 15 dB_ Neverthelesst the central tendency of the data Is clearly indicated by the above values. Considering the necessary assumptions required to compute_s fromthesedatot it is estlmated that the valuesgiven In Table B-1 are reliable within better than + 1.0 dB.

B-7

Table B-2 Summaryof Computationof _$ , from ]SO Round Robin Data for First Nine (Repetitive) ImpulsiveSounds

Dak_ Source

]mpu!sive Sound

Ref.

__'
(_)

'

I 2 L 3 I 4
/ '

a _9

SiQn_l
1o]

--I
2

,-_o,_lbl 79, 7761 ,3 177.6 7837Ro 78, 7837R2 7R4 i t '

t, LA/c) _0.7 I -0.8 -I.I ! -R.8 -R.I _R.2 0 -0.1 I -0.2 0

3 I ,q_.,,_d, !77:,,1175.6, 1_5:3 _7_170.3 :0, 176 , 176. ., _76.3 ,, 176 , 7_., .2, J I ....... L
4 5 Leq(Test)(8) _t IF) 75.6 I 73.4 69.7 ! 69.9 70,9 75.6 77.R 3.6 72.9 3.4 75,9' 0.3 ' 1.5 2.2 _1 5.6 5.7 5.4 1.0 r_ 6 (77) ELA(_ub){g) 12.2 8.9 7.0 7.5 8.4 11.5 8.2 8.7 11.4 Avg 9.3

7 5,_, ;_._ _,;.Ti_.613.;13.8 ,2._,,.R,2.,

.. (o) (b) L_ve[ of Calibration Tone for Reference SignQI. (e)

:,2._'
*0.9

From To_le 20, Reference 66 (t_ma a_ dR(A) Integrated}"

LAI(Ref) = Impact Sound L_vel from Table 2b, Referenca (so_ al dB(Az)). _ LAI = LAI(Raf) - 78.4 1where 78.4 = LAI oE Calibratlon Tone on Refarence Channel. L (Ref) = 76.4 + P'ow2, EsHmaled Values of eq L for Raferenca Channels. 176.4 = L of eq eq Calibration SignQI on Referance Chonrml. )

(f) (g) (h) (1)

5t = Lcl(Re/) " Leq(Tast)' ._ow 3 - Row 4. From Table 3.3.3.2, _ --ELAsubj+_t, Overall Reference 77. Row 5 + Row 6.'

(c) (d)

/V_an _'s for all 9 Sour<l_.

B-8

Not considered here is the fact that the values of ELAsubjreported by Shlpton et a176t, show a variation with presentation level due to the so-called mld-level bulge in loudness growth.* The effect was relatively small, however_ and hasbeen averaged

out in the above figures. Since L data were not available for the 5 single impulsive eq sounds(Group 11)_ 6s values for Ibese sounds could not be established. CombinedSubjective and Objective Corrections From ISO RoundRobin Tests Analogous to the Equal Loudness Attenuation (ELAsubj) to achieve subjective equallfy of the reference and test signalsr there is also an objective Equal "Loudness" Attenuation (ELAob_) - again adopting Pedersen'sterminology -which ;s the attenuation of the testsfgnal required to achieve the sameresponse on the objective measuring instrumentas for the reference s;gnal. The comparableobjective correction Factor A o which we seekwill be the difference in L between the test and reference s[gnars to eq achieve the same"instrument reading." As with Ast there is the same initial difference

in level _t between the reference s_gnaland the unattenuated slgnal Qnd ;t can be shown that_ for the proceduresemployed in the 150 tests_ Ao =-(ELAob j +At) ,dB (B-2)

Thus_the quantity ELAsubj - EkAob]_ reported for the /SO tests, is the sameas the sumof our objectlve and subjective correction factors (_o +As)* This quanr;ty can be shown to be ec_ual1othe equivalent level of the reference signal, when it is adjusted to the sameloudnessas the test slgnab minusthe equivalent level of this samereference slgnal_.whenit is now adjusted to have the same "Fnstrument"reading as the test signal**" An ideal "instrument" would have a zero value for _ -eA sothat it would
O $

correctly measure the loudnessof on impulsive s_und. However,, a fixed but consistent 6 could be considered as a "error"_ representedby a constantnan-zero value of _o s fixed "instrument" error to be corrected out. The critical parametert thereforet for

* **

At presentation levels of 55*"75,. and 95 dB_ the average values of ELAsuk:'s from Reference66 were 9.3t ]l.2t and 9.51 respectively. When,_ + 6O is added to the "instrument" reading of the test signal the resulting S level is the equivalent sound level_ keqOfan equally loud reference signal. B-9

evaluating

the ablilty of an), "instrument" to measure impulse noise, be it an actual

sound level meter (SLM) or a loudness calculation method, would be the standard deviation of the values of ._ + ,_ about the mean.
O S

Table B-3 summarizes the comparable values of AO + _S from the |SO data. table defines the mean and standard deviation,

The

over noise sources, sub]ectsl and levels measure-

for A + AS for each of the iSO data sources and for the variety of objective O ment "instrument" indlcal_d, it appears from these data that A-weighted

sound level,

slow (LAs) equivalent sound level (Leq)l or some form of loudnesscalculation using, preferably, time-integrated measures of the spectral content, would all have potentially higher utility and vaUdity thunother "instrument"/metric combinations. For the single

event impulsive sounds(Group |I), all the measures with the exception of" B-welghted peak impulse or C-weighted peak-held indicate substantial variation about the mean.

The results of this Round Robin Test can also be compared, in terms of the mean objective correction faator_ o wEth the results from Appendix A. From Table B-2, the mean sum 4 + _ s for A-wefghted slow levels is + ] 1.6. Subtracting the mean subjective correction factor _ S of 12 o5 from Table B-2 gives a mean objective correction

factor _ O of - 0.9 dB. That is, the average A-weighted slow ISLM reading of the 9 impulsive sounds tested would be 0.9 dB above the average L of these sounds. eq This average objective correction factor from the ISO round robin tests of o (].9 dB compares well with the average of A o of + 0.1 and - 1.4 d8 from the two categories of _mpulslve signals (repetition rate > I pps or 0.4 to 1 pps respectively) reported in Appendix A,

B-10

Table 8-3 Evaruafion of ISO RoundRobin Data for f Optlmum "[nsh'ument" to Evaluate Impulsive Noise( ) Sounds1-9 (Group I) A + A
0 S (7 0

Sounds10-14 (Group H) A + _ dB 1.5 1.8 dB_e) +13.4 (6) +12.5 (7)

S O"

't|nsfrumenf" SoundLevel Meter

Metric LAS(a) LAF(a) LAI(a) k (a) eq

dB(e) +11.6 (6) +12.1 (8) +11.2 (11) +10.8 (1)

dB 3.6 4.9 3.3 2.6

2.3 1.3

+ 8.3 (11) +10.5 (1)

I I I

L_,(P[)(c) LpK(4)(d)

+11.9 (1) +11.8 (1)

2.4 2.4

+ 4.5

(I)

1.7 1.6

+ 3.6 (l)

Stevens Mk. V|

' LCS Leq(c)(b)

+ 4.1 (3) - 0.5 (1)

2.2 ].4 + 1.6 (1) 1.8

Ste ns M k.ve VII Zwicker

Leq(C)(b) LCS

1.5 (1)

1.2 2.6

- 0.2 (1) + 8.4 (3)

1.8 7.3

- 0.8 (3)

(a) Maximum peak reading. _o) (a) B-weighted peak impulse. (d) C-weighted peak impulse, One secondintegration Hme.

(e) Number in parenthesess_gnifies number of laboratories who provided data for th_svalue. (f) _o + As defines absolute'accuracy of loudnesspred_cHon, c_is standard deviation about this mean(see text).

B-J1
:j_,_,_._ _ -................... ,,.

APPENDIX C FREQUENCYSPECTRA OF REPEATED TONE BURSTS

Four basic cases for the frequency spectra of transient sounds are illustrated in Figure C-1. The corresponding Fourier spectrafor each oF these eases,where the 135 peak amplitudeoFthe pulse is Po' can be given as follows:

Single SquarePulse

j P(jw) I-_ po T i swT [nw T/2 J /2

(C-l)

RepeatedSquarePulse /P(wt) Jrnex=Po TI | +2'_. n--1 _

I_ o.-_-tT ,Z sin(n tuoT /2) 2-I ] 1/2

(c-2)

Single Tone Burst

- W "l CUl) T T 2j /2 I P(]cu)I = poT sin(cu(w _/

(C-3)

Repeated Tone Burst I P(wt)Jrnax = pT where T = Pulseduration q" = Pulserepetition period uJ 1 = PulseFrequency

'_

1 +2 (nw - w1) T/2 n= _7 jsln(n_u-CUl)T/2

(C-4) 211/2

m o = 2"rt/T, the pulse repetition Frequency n = Order of harmonic

C-1

Time History, P(t)

Fourier Spectra, P(_)

'-7l
, -T/2 _ T/2

Single Square Pulse

I I I 0 2_/T

0 2_/'T

o T - -,,.i/_.

I/r

, t

S_ngleTone Burst

_I

_w

,. t T r T/2 /

RepeatedTone Burst

.... -t-J

_,'_',-".'- :__u

w1

(,*J1

Figure C-1.

Time Historyand Fourier Spectraof FourComrnon impulsive Wave Forms

C-2

The general shapeair the envelope of the frequency spectra is the samein all cases,_ sinx/x. For the single or repeated tonebursts, the spectrumis the sameas For

the corresponding caseof a single Or repeated squarepulse but with the peak frequency shifted to the right to the frequency(_ul)of the pulsedtone. The 7/2 power bandwidth (/, F) of the spectrum for the caseof"the single tone burstcan be expressed as L_ F = 1/I". Thust for a single pulse with only one cycle, T = 21r/w] and the 1/2 power b_ndwidth is equal to the Frequency of the pulse il'salf, Then, a single _mpulse with only one cycle will have a very bread spectrumso that its loudnessw_lJ correspond to the summation of loudness over manycritical bandsin the ear. For o repetitive version of such an impulsivesignal, the frequency separationof the sldebandsis equal to the pulse repetil'ion frequencycu 1 = 2_/'r. The number{N) of harmonics within the same "1/;

power'* point on the spectral envelope wouldbe T N ="_" = ]/duty cycle

jr

C-3

REFERENCES PARTA. ANNOYANCE OF IMPULSIVEOR FLUCTUATING SOUNDS

I.

Plutchik_ Robert, "A Critical Analysis of the Literai_re Dealing with the Effect of Intermittent Sound Stimuli on Performance, Feeling, and Physiology, with Preliminary Work Towardan Experimental Analysis of the Problem ." Report No. AD 149589t October 1957. Broadbent, D.E. and Robinson,D.W.I "Subjective Measurementsof the Relative Annoyanceof Simulated Sonic Bangs and Aircraft No_se." J. SoundVib. 1, 162-174 (1964). Borsky, P.N., "CommunityReactionsto Sonic Booms in the Oklahoma City Area. VOIo II, Data on Community Reactionsand Znterpretat_onSo" AMRL-TR-65-37, Wright Patterson Air Force Base,Oh_o_October 1965.

2.

3.

3a. Peorsons, K.S., "Noisiness Judgments oFHelicopter Flyovers," FederalAviation AgencyReportFAA DS67-1, January 1967. 4. Shepherd,L.J. and Sutherland,W.W., "Relative Annoyanceand Loudnesof s Judgmen_of Various SimulatedSonic BoomWavefarms." NASA CR-1]92, 1968.

40. Rothauser,E.H., Urbanck_G.E. and Pachl, W. P., "Loudness and Annoyance of ImpulsiveOffice Noise." IEEETransactions on Audio and Electroacoustlcs, Vol. AU-16, No. 4, 520-522, December(1968). 5. Nixont C.W.I van Gierket H.E.I and RoslngersG., "ComparativeAnnoyances of Approaching Versus RecedingSound Sources." J.. Acoust. Sac. Am. 45, 330 (1969). Roblnsons D.W., "The Conceptof NoisePollution Level." NPL Aero ReportAc 38, Nat|onaJ PhysicalLaboratory, Teddlngtont Middlesexs England t March 1969. Kryterr K., "The Effectsof Noise on Man." Academic Pressw New York, 304, 423, 1970. Keighley_.E.C. "Acceptability Criteria for Noise in Large Offices." J. Sound& Vib. 11S83-93 (1970)o Roblnsonl D.W., "Towardsa Unified Systemof Noise Assessment. " J. SoundVib. 14 (3)1 279-298 (1971).

6.

7.

8.

9.

R-]

REFERENCES (Continued) PARTA. ANNOYANCE OF IMPULSIVEOR FLUCTUATING SOUNDS (Continued) 10. Anderson, C.M.B. and Robinson, D.W. "The EfFect of Interruption Rate on the Annoyance of an Intermittent Noise. " " NPLAcoustios ReportAc 53, National Physical Laboratoryt Teddington, Middlesex, England, 1971. Eldred, K., "Community Noise." Wyle Laboratories ReportNo. WR 71-17, U.S. EnvironmentalProtection Agency Report No. NTID 300.3, 31 December 1971.

11.

1la. international Organization i:orStendardizationt "Assessment of Noise with Respect to Community Response." ISO Recommendation R 1996, 1971. 11b. International Organization for Stendardizationt "Assessment oFOccupational Noise Exposurefor HearingConservationPurposes." ISO Recommendation R 1999, 1971. 11c. Leverton, J.W., "Helicopter Noise - Blade Slap." Part 2, Experimental Results. National Aeronauticsand SpaceAdmlnislrolion, NASA CR-1983, March 1972. 12. Parry, H.J. and Parry, J.K., "The Interpretation and Meaning oF Laboratory Determinations of the EFFeco t F Duration on the JudgedAcceptabilily oF Noise." J. Soundand Vib. 20(1), 51-57 (1972). Fuehs, G.Lo s "Integration Time Constantfor Annoyanceof Impulsive Noises" Aeustlaa 27, 313-316 (1972). Motsahat, K, MUller, E.A. and Zimmermanl G._ "Further Developments of Noise Indices Taking Into Account the Results of the ATreraft Noises S_dy of the Deutsche Forschungsgemelnschaft in Munich." Max-Planck-lnsfitut Fur Stromungsforschung_ March 1973. Fullers H.Co and Robinson,D.We "SubjectiveReactionsto Steadyand Varying Noise Environments." NPL AcousticsReportAC 62,. National Physicalkaboratorys Teddlngton, Middlesexs England,August 1973. Schemer..Paul D._ "Predicting Community Responst e o Blast Noise." Construction EngineeringResearch LaboratoryeChampaign, lll.t ReportNo.CERL-1R-En17s December 1973. Munch, C.L. and Kings R.J., "CommunityAcceptanceof Helicopter Noise: Criteria and Application." STkorsky A_rcroftfor NASA.. NASA CR-132430,. (1974) Berry. B.F., Rennie t A.J. and Fuller, HC., "RatingHelicopter Noise: the Peasibility ofan "Impulsive NoiseCorrection"." National PhysicalLaboratory memofor [SO/TC43/SC]/WG2 (Berryend Robinson - 2) 67t October 1975.

13.

14.

15.

16.

17.

18.

R-2

REFERENCES (Continued) PARTA. ANNOYANCE OF IMPULSIVEOR FLUCTUATING SOUNDS (Continued) 18a. Man-Acoustics & Noise inc., "Noise CertTfication ConslderaHons for Helicopters Basedon Laboratory investigations." Reort for U.S. Department of Transportation, FAA-RD-76-116, July 1976. 19. Lawton, B.W., "Subjective Assessmenof t Simulated Helicopter Blade-Slap Noise." NASA, Langley ResearchCenter, NASA TN D-8359, December 1976. DIN 45645, "Einheitliche Ermittlung des Beurteilungspegels fL/r Ger_uschlmmissionen" (Uniform Calculation of Judged LeundessLevel of Noise Emission). (Enhvurf 1976.) Notional Research Council, Committee on Hearlng, Bioacoustics and Biomechanics, Assembly of Behavioral and Social Sciences, "Guidelines for Preparing Environmental impact Statementson Noise." June 1977. Patterson, J.H., Jr., i_4ozo_ B.T., Schomer, R.D. and Camp, R.T., Jr., "Subjective Ratingsof Annoyance Produced by Rotary-Wlng Aircraft Noise." U.S. Army Medical Researchand DevelopmentCommand, USAARL Report No. 77-12, May 1977. Galloway, W.J., "Subjective Response to Simulated and Actual Helicopter Blade Slap Noise." Bolt Beranekand Newman Report No. 3573 for the National Aeronautics and SpaceAdministratbn, December 1977. Galanter, E., Popper, R.D. and Perero, T.B., "Annoyance Scales for Simulated VTOL andCTOL OverHights." Paperbefore 94th Meeting, Acoustical Society of America, Miomlt Florida, December 1977.

20.

21,

22.

23.

24.

24o. Galloway, W.J., "Review of the Development of Helicopter impulsive Assessment Proposalsby ISO TC43/SC1/WG2 -Aircraft Noise." MemorandumReport, January 1978. 24b. Levertonr J.W., Southwood, B.J. and Pike, A.C._ "Rating Helicopter Noise." Westland Helicopters, Ltd., England, 1978. 24c. Powell, C.A., "A Subjectlve Field Study of Helicopter Blade-Slop Noise." National Aeronautics and SpaceAdministration, Langely ResearchCenter, NASA Technical Memorandum78758, July 1978. 24d. d'Ambra, F. and Damongeet, A., "Annoyance of Helicopter Impulsive Noise." S.N.I. AerospaHale, Marlgnane, France. Paperfrom "Hellcopter AcousHcs," NASA Conference Publication 2052, Part II, May 22-24, 1978.

R-3

REFERENCES (Continued) PARTA. ANNOYANCE OF IMPULSIVEOR FLUCTUATING SOUNDS (Continued)

24e. Klump, R. G. and Schmidt, D.R., "Annoyance of Helicopter Bladeslap." Naval Ocean SystemsCenter Technical Report247, July 3, 1978. 246. Sternfelcl, H., Jr._ and Doyle1 L.B., "Evaluation of the Annoyance Due to Helicopter Noise." NASA Contractor Report3001, June 1978.

R-4

REFERENCE( S Continued) PART B. NOISINESS AND LOUDNESSOF IMPULSIVEOR FLUCTUATING SOUNDS 25. Hughes, J.W., "The Threshold of Audition for Short Periodsof Stimulation." Proc. Roy. Sac. (London), B133, 293 (1946). Garner, W.R. and M_llerl G.A., "The h,_skedThreshold of"PureTones as a Function oF Duration." J. Exp. Psychol. 37, 293-303 (1947). Munson, W.A._ "The Growth of. Auditory Sensatlon." J. Acoust. Sac. Am. 19, 584-591 (1947). Garner_ W.R., "Auditory Thresholdsof ShortTones as a Function of. Repetition Rates." J. Acoust. Sec. Am. 19, 600-608 (1947). Garner_ W.R., "The Effect of. Frequency Spectrumin Temporal Integration of. Energyin the Ear." J. Acoust. Sac. Am. 19, B08-814 (1947). Miller, G.A., "The Perception of Short Burstsof. Noise." 20, 513-527 (194B). J. Acoust. Sac. Am.

26.

27.

28.

29.

30.

31.

Gamer_ W.R., ''The Loudnessof RepeatedShot,Tones." J. Acoust. Sac. Am. 20, 513-527 (1948). Garner, W. R., "The Loudness and Loudness Matching of ShortTones." J. Acoust. Sec. Am. 21, 398-401 (1949). Niese, H., "Vorschleg fur die Definition und Messungder Deutlichkeit nach subjektiven Grundlagen" (Proposal for Definition and Measurementof Distinctiveness According to a Subjective Basis). Hochfrequenztechn. Eleklroakust. 65, 4. Green, D.M., Birdcall, T.G. and Tanner, W.P., Jr., "Signal Detection asa Function of Signal Intensity and DuroHon." J. Acoust. Sac. Am. 29, 523-531, (1957). Hamilton_ P.M., "Noise Masked Thresholds asa Function oFTonal Duratlanand Masking Noise Band Width. " J. Acoust. Sac. Am. 29, 506-511 (1957). Pollock, l.t "Loudnessof Period;cally Am. 32:4_ 181-185 (1958).
interrupted

32.

33.

34.

35.

36.

White Noise. " J. Acoust. Sac.

37.

Plomp, R. and Bouman, M.A., "Relation Between Hearing Thresholdand Duration ForTone Pulses." J. Acoust. Sac. Am. 31, 749-758 (1959).

R-5

REFERENCES (Continued) PARTB. NOISINESS AND LOUDNESSOF IMPULSIVEOR FLUCTUATING SOUNDS (Continued) 38. Miskolczy-Fodor, R., "Relation Between Loudnessand Duration of Tonal Pulses." H. Responso ef Normal Ears to Sounds with Noise Sensation." J. Acoust. Sac. Am. 32:4, 482-486 (1960). Niese_ H.t "Subjektive Messungden Lautst_rkevan Bandpessrauschen (Subjective Measurementof Intensity of Bandsof Noise). " HockFrequenztechnlk und Elekh'oakustlk 68 (1960). Zwislocki, J., "Theory of Temperal Auditory Summation." J. Acoust. Sac. Am. 32:8, 1046-1060 (1960). SmaJl, A.M., Brandb J.F., and Cox, P.G., "Loudnessasa Function of Signal Duration." J. Acoust. Sac. Am. 34, 513-514 (1962). Patti E., "Uber die Lautst_rke elnzelner kurzer SchalHmpulse(On the Intensity of Short Single Soundlmpulses)." Beih. Akust., No. 1, 212(1963). Sheeley, E.C. and Bilger, R.C., "Temporal IntegraHon asa Function of Frequency," J. Acoust. Sac. Am. 36, 1850-1857(1964). Carter, N.L., "Effect of RepeHHonRateon the Loudness of Trlangular Transients.' J. Acoust. Sac. Am. 37:2, 308-312 (1965). Zw_cker, E., "Temporal Effects in S_multaneous Masking and Luudness." g. Acoust. Sac. Am. 38, 132-141 (1965). Zepler, E.E. and Harel, J.R.P., "The Loudness of Sonic Booms and Other Impulsive Sounds." J. SoundVrb. 2, 249-256 (1965). Garret't, R.M., "DetermlnaHon of the Loudness of Repeated Pulsesof Nobe." d. Sound Vib. 2, 42 (1965). Ekman, G., Berglund, B. and Berglund, V., "Loudness as a Function of th0 Duration of Auditory Stimulation." ScandinavianJour. of Psychology7, 201-208 (1966). Kryter, K. D,, "Review of Research and Methods for Measuring Loudness and Noisiness of CompJoxSounds." NASA CR 442, 1966.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

R.-6

REFERENCES (Continued} PARTB. NOISINESS AND LOUDNESS OF IMPULSIVEOR FLUCTUATING SOUNDS (Continued) 5,0. Stevens, J.C. and Hall, J.W., "Brightness and Loudness as Functions of Stimulus Duration." Perception and Psychophysics 1,.317-327 (1966). Zwicker, E., "Fin Beih'agzur Lautstarkemessung Impulshaltiger Schalle (A Contribution to the Measurement af Intensity oFImpulsive Sounds)." Acustica 17, 11-22 (1966). Pearsons,K.S., "NoisinessJudgmentsoFHelicopter Flyovers." Report No. FAA DS-67-1, January 1967. Peersons, K. and Haronjeff, R., "Category Scallng Judgment Testson Motor Vehicle and Airoraff Noise." ReportNo. FAA DS-67-8, AD 658 755, July 1967. Dubrovskli, N.A. and Tumarklna, L.N., "Investigationof the Human Perception of Amplitude-Modulated No;se." Soviet Physlcs-Acoustlcs13:1, July-September 1967. Johnson, D.R. and Robinson,D.W., Acustica 1B, 241-258 (1967). "The Subjective Evaluation of"Sonic Bangs."

51.

52.

53.

54.

55.

.56.

Bauer, B.B., Torlek, E.L., Rosenheck,A.J. and Allen, R.G., "A LoudnessLevel Monitor for Broadcasting." IEEETrans. in Audio and Eleetroacousflcs, AU-15, t77-182 (1967). Harbert, F., Young, I.M. and Wenner, C.H., "Auditory Flutter Fusionand Envelopeof Signal." J. Acoust. Sac. Am. 44:6,803-806 (1968). Shepherdt L.J. and Sutherland, W.W., "Relative Annoyanceand Loudness of Judgmentsof VariousSimulatedSonic BoomWaveforms." NASA CR-1192, 1968. RothauserlE.H., Urbanek, G.E. and Pachl, W.P., "Loudness and Annoyanceof Impulsive Office Noise." |EEETransactions in Audio and Eleetroacoustics/AU-16, 520-.522(1968). Johnson, D.R. and Mbinson, D.W., "Prol:edurefor Calculating the Loudness of SonicBangs." Aoustica 21,307-318 (1969).

57.

58.

59,

60.

60a. Shipton, M.S. and Robinson,D.W., "Ambient Noise Limits for Industrial Audlametry." Natlonal Physical Laboratory, NPL AcousticsReportAo 69, April 1969.

R-7

REFERENCES (Continued) PART B. NOISINESS AND LOUDNESS OF IMPULSIVE OR FLUCTUATING SOUNDS (Continued)

61.

Reichardb W. and Nieser H.t "Choice of SoundDuration and Silent intervals for Test and Comparison Signals in the Sub]ectlve Measurementof Loudness Level." J. Acoust. Sac. Am. 47:4(2), t083-i090 (1970). Reichardt, W., "Subjective and Obiectlve Measurementof the Loudness Love) of Single and Repeated Impulses." d. Acoust. Sac. Am. 47, 1557-1562 (1970). Fidell, S., Pearsons,K., Grlgnetti, M. and Green, D., "The Nolslnessof Impulsive Sounds." d. Acoust. Sac. Am. 48:6(I), 1304-1310 (1970). Fide)ll S. and Pearsons, K.S., " Studyof the Audibility of Impulsive Sounds." NASA CR 1598, 1970. (Seealso d. Acoust. Sac. Am. 48: 1304-.1310, 1970.) Ollerhead, J., "An Evaluation of Methodsfor Scaling Aircraft Nalse Perception." NASA CR 1883, Wyle LaboratoriesReportNo. WR70-17, 1971. Sh[pton, M.S., Evans, D.H. and Robinson,D.W., "An InvesHgatlonof the Loudness of Noises wlth ImpulsiveCharacteristics." National PhysicalLaboratory, NPL AcousHcs Re_rt Ac 50, duly 1971. Thompson, P.O. and Gales, R.S., "Subjectlve ,Judgment of Loudness Level of ImpulsiveNoises for the Internatlonal Round RoblnTests." Paper FF8, AcousHcal Societyof'Amerlca Meeting, October 1971. Zwlcker, E. and Fastl, H., "On the Developmeno t f the Critlcat Band." J. Acoust. Sac. Am. 52:2 (Part 2), 699-702 (1972). Parry, H.,J. and Parry, J.K., "The Interpretation and Meaning of Laboratory Determinatians of the Effect of Duration on the Judged Acceptability of Noise." d. Sound and Vib. 20(1), 51-57 (1972). Leverton, d.W., "Helicopter Noise-Blade Slap. Part 2, Experimental Results." Institute of Sound and Vibration for National Aeronauticsand SpaceAdrnlnlstrationj NASA CR-1983, 1972. Fuchst G,L.,, "integration Time Constantfor Annoyanceof impulsive Noises." Aeustlca 27, 313-314(L) (1972). Carter, N.L., "Effect of RiseTime andRepetition Rateon the Loudness of Acoustic Transients." .J. Sound and V;b. 21(2), 227-239 (1972),

62.

63.

64.

65.

66.

67.

68.

69.

70.

71,

72.

R8

REFERENCES (Continued) PARTB. NOISINESS AND LOUDNESS OF IMPULSIVEOR FLUCTUATING SOUNDS (Concluded) 73. Stephens, S.D.G., "Auditory Temporal Integration as a Function of Intensity." J. Soundand Vib. 30(I), 109-126 (1973). Carter, N.L. and Dunlop, d.l., "The EFFecto s f RiseTime and Repetition Rote on the ThresholdsForAcoustic Tronsients." J. Sound Vib. 30(3), 359-366 (1973). Boone, M.M., "LoudnessMeasurementson Pure Tone and Broad BandImpulslve Sounds." Acustica 29, 198-204 (1973). Leverton, John W., "Helicopter Noise - Are Existing Methods Adequate For Rating Annoyance or Loudness?" J. American Helicopter Society, 41-441 April 1974. Pederson, O. Juhl, Lyregaard, PE. and Poulsen, T., "The RoundRobin Test on Evoluation of LoudnessLevel of impulsive Nolse." ISO/TC 43/SC/SG'B' (Secretariat- 15) 23, 1977. GustaFsson,B., "The Loudness of Transient Sounds as a Function of SomePhysical Parameters" ,J. Sound and Vib. 37, 389-398 (1974). Torhardt, E., "On the Perception of Periodic SoundFluctuations (Roughness)." Acustica 30, 201-213 (1974). Fuller, H.C. and Robinson,D.W., "TemporalVariablesin the Assessment of an ExperimentalNoise Environment." National Physical Laboratoryt NPL Acoustics ReportAc 72, February 1975. Magliozzb B., Metzger, F.B., RouschlW. and King, R.J._ "A Comprehensive ReviewoF Helicopter Noise Literature." ReportNo. FAA_RD-75-79_ June 1975. Isumi, Kiyoto_ "Two Experiments on the Perceived Noisinessof Periodically IntermittentSounds_"Noise Control Engineering9, 16-23 (1977). Errata Far Fig. 13and 14 by PersonalCommunlcatlon_October 13, 1978.

74.

75.

76.

77.

78.

79.

80.

81.

82.

R-9

REFERENCES (Continued) PARTC. 83. DETECTIONOR PERCEPTION OF IMPULSIVE SOUND

Rnase, T.S., Kryter, K.D. and Stevens, S.S., "The Relative Annoyance Produced by Various Bandsof Noise." Psycho-Acoustic Lab, Harvard University, IC-65, 1944. SherwJn_ C.W., et al, "Detection oF Signalsin Noise: A Comparison Between the HumanDetectorand an Elech'onic Detector." J. Acoust. Sac. Am. 28t 617-622 (1956). Marill, T., "Detection Theoryand Psychophyslcs."MIT ResearchLab, Elech'on Report No. 319, October 1956. Niese, H., Beilrag zur RelationzwischenLautst_rke und L_stigkeitvan Ger6uschen (Conlrlbutlon to the RelationBetween Intensityand Annoyanceof Noises)." Aaustica 15, 236-243 (1965). International Organization ForStandardization, "Method for Calculating Loudness Level." ISO/R-532-1966(E). Parnell, J.E., Nagel, D.C. and Parry, H.J._ "Growth oF NoisinessFor Tones and Bandsof Noise at Different Frequencies." U.S. Department of Transportation, FAA DS 67-211 1967. Zwicker, E. and Feldtkeller, R., "[:)asOhr als Nachrichtenempfonger,2 Auflage (11"_ Earas e a Receptionof Messages,2nd Edition)." S. Hirzel, Stuttgart(Second Edition), 248, 1967. McGill, W.J., "VarTationon Marill's Detection Formula." J. Aeoust. Sac. Am. 43, 70-73 (1968). Jeffress, L.A., "Mathematicaland Electrical Models of Audltory Detection." J. Acoust. Sac. Am. 44, 187-203 (1968). Teranishi, R. and Shaw, E.A.G., "External-Ear Acoustic Modelswith Simple Geometry." J. Acoust. Sac. Am. 44, 257-263 (1968). Stevens, S.S., "Procedure for Calculating LoudnaH: Mark VI." J. Acoust. See. Am. 33, 1577-1585 (1961). (See also, "Perceived Level of Noise by Mark V]l and Decibels(E)." J. Acoust. Sac. Am. 51,575-601 (1972).) Scharf, B., "Loudness" Chapter in Vol. 4, The Hancbookof Perceptions,Carterette, E.C. and Friedman, M.P. (Editors), Academic Press, New York, September1973. Shaw, E.A.G., Transformation of SoundPressure Level from the Free Field to the Eardrumin the Horizontal Piano." J. Acoust. Sac. Am. 56, 1848-1861 (1974). R-IO

84.

85.

86.

87.

88.

89.

90.

91.

92.

93.

94.

95.

REFERENCES (Continued) PARTD. SPEECH INTERFERENCE 96. Miller, G.A. and Lickllder_ J.C.R., "The IntelHglbTllty of interrupted Speech." J. Acaust. Sac. Am. 22, 167-173 (1950). Hawley, M.E. and Kryter, K.D., "Effects of Noise on Speech." Chapter 9 in "Handbook of Noise Control," (C. M. Harris Ed.), McGraw-Hill, N.Y., 1957. Pallack_ I., "MessageProcedures for Unfavorable Communlcat_onConditions." d. Acoust. Sac. Am. 30, 282-285 (1958). Williams, C.E. et al., "The Speech Interference Effects of Aircraft No_se." Report No. FAA DS-67-19, Bolt Beranekand Newman, Inc. 1967. American NaHanal Standards Institute, "Methods far the Calculation of the Art_oulatlon Index." ANSI $3,5-1969. Jeffress, L.A. "Masking." In "Foundation of Modern Audffary Theoryt Vol. 1. II (J. V. Tobias, Editor), Academic Press_ New york, 1970. Scharfl B., "Critical Bands." In "Foundationof Modern Auditory Theory, Vol. 1. " (d. V. Tobias, Editor), Academic Press,New York, 1970. Kryter_ K._ "The Effects of Noise on Man. '_Academic Press,New york, p. 55, 1970. Webstert J.C. t "The Effectsof Noise on the Hearing of Speech." In: Proceedings of the Intemational Congress on Noise as a Public Health Problem." Dubrovnik, Yugoslavia, ReportNo, EPA 550/9-73-008, May 13-18, 1973. van Gierkat H. (Task Group Chalrman)_ "Impact Characterization of Noise Including ImplicaHonsof Identifying and Achieving LevelsofCurnulatlve No_se Exposure." U.S. EnvironmentalProteotFon Agency Aircraft/ATrpart Noise StudyRepartt NT1D 73.4, 27 July 1973. U.S. EnvironmentalProtectionAgency, "Public Health and Welfare Criteria for No_se." Report No. 550/9-73-002, July 27, 1973.

97.

98.

99.

100.

101.

102.

103. 104.

105.

106.

R-11

REFERENCES (Continued) PART E. SLEEP INI_ERENCE

107.

Oswald_ I. t Taylor, A.M. t and Trlesman, M., "Discrlminitive Responses to Stimulation During HumanSleep." Brain, 83t 440-553 (1960). Schleber, J. P, Mery, J._ and Muzet, A. "Etude AnalyHque en Leboratolre de I']nfluence du Bruit sur [e Sommell ." Reportof Centre d'EtudesBioclimatlquesdu CNRS, Strasbourg,France, 1968. Lukas, V. S., and Kryter, K. D,, "Awakening Effectsof SHmulated Sonic Booms and Subsonic A_rcraft Noise." In: "Physiological EFfects of Noise." Welch, B.L. and Welch, A.S. (Editors), PlenumPress_ New York, 283-293 (1970). Berry B. and Thiessen,G.J., "Effects of Impulsive Noise on Sleep." ResearchCouncil of Canada, NRC 11597_ 36, 1970. National

108.

109.

110.

111.

Mery J. Muzet, A., and Schleber, 3. P. _ "Effects du Bru_td4Avionssur le SornmeTI." In: Proceedings of 7th International Congress of AcousHcs_3, Budapest, 509-512, (1971).

(See also numbers103, 106) PARTF. HEARING DAMAGE 112. Ward, W.D., Glorig, A., and Sklarl D., "Relation BetweenRecoveryfrom Temporary ThresholdShift and DuraHonof Exposure." J. Acoust. Sea. Am. 31:5, 600-602 (1959). Ward, W.D.I GIoHg, A., and Sklars D., "Temporary Threshold Shift Producedby Intermittent Exposure to Noise." d. Acoust. Sac. Am. 3h6, 791-794 (1959). Ward W.D., Se/ters_W. and Glorlg, A., "Exploratory Studieson Temporary Threshold Shift from Impulses." J. Acoust. Sac. Am. 33:6,781-793 (1961). Kryter, K., Ward, W.D., Miller, J.D., and Eldredge, D.H., "Hazardous Exposure to Intermittent and Steady-State Noise." J. Acoust. Sac. Am. 39:3, 451-464 (1966). Fletcher, J.L., "Recovery from Impulse Noise Induced AcousticTrauma." Report AD 645 898, U.S. Army Medical ResearchLaboratory, Ft. Knox, Ky. November 221 1966. Botsford,J. "S_mpleMethod for Identifying Acceptable Noise Exposures." J. Acoust. Sac. Am. 42:6, 810-819 (1967).

113.

114.

115.

116.

117.

R-12

REFERENCES (Continued) PARTF. HEARING DAMAGE (Continued) 118. Coles, R., Garlnther, G. i Hedge, D., and Rice, C., "Criteria for Assessing Hearing DamageRisk from Impulse-Nolse Exposure." U.S. Army Technlcal Memo 13-67r Human Engineering Laboratories, AberdeenProvingGround, Md., August 1967. Ward, W.D. (Editor), "Proposed Damage-Risk Criterion For ImpulseNoise (Gunfire)," NAS-NRC CHABA Working Group 57, July 1968. Coles, R., Garinther, G., Lodge, O., and Rico, C., "Hazardous Exposureto Impulse Noise." d. Acoust. Sac. Am. 43:2, 336-343 (1968). Loeb, M. and Fletcher, J.L., "Impulse Duration andTemporary Threshold Shift." J. Acoust. Sac. Am. 44:6_ 1524-1528 (1968)o Ward, W.D., "TemporaryThreshold Shift end Damage-RiskCriteria for Intermittent Noise Exposures." J. Acoust. Sac. Am. 48:2(2), 561-574 (1970). Luz_ G., and Hedger D.C.t "Recovery from Impulse-Noise Induced TTS in Monkeys and Men: A Descriptlve Model." J. Acoust. Sac. Am. 49:6(2)t 1770-1777 (1971). Cohen, A., Anticaglla, J.R., and Carpentert P.L., "Temporary ThresholdShift in Hearing from Exposureto Different Noise Spectra at Equal dBA Level." J. Acoust. Sac. Am. 51:3(2), 503-507 (1972). Burns, W. and Robinson,D.W., "Hearingancl Noise in Industry." Her Maiosty's Stationery OFFice, London (1970). Colas, R.R.A., Rice, C.G.I and Martin, A.M., "Nolse-lnduced Hearing Loss from Impulse No_se: PresentStatus." In: Proceedlngs of the International Congress on Noise as a Public Health Problem, Dubrovn[k, Yugoslavlaj May 13-18, 1973. EPA Report No. 550/9-73-008, 1973. Dieroff, H.G., "Hearing DamageCaused by Very Shorb High-lntensity ImpulseNoise." In: Proceedings of the International Congress on Noise as a Public Health Problemj Dubrovnlk, Yugoslavia, May 13-18, 1973. EPA ReportNo. 550/9-73-008_ 1973. Johansson_ B., Kylln, B., and Reopstarff,S., "Evaluationof the Hearing Damage Risk from Intermittent Noise According to the ISO Recommendations."In: Proceedlngs of the International Congress on Noise as a Public Health Problem, Dubrovnlkt Yugoslavia, May 13-18, 1973.'_ EPAReportNo. 550/9-73-008, 1973.

119.

120.

121.

122.

123.

124.

125.

126.

127.

128.

R-13

REFERENCES (CanHnued) PARTF. HEARING DAMAGE (Concluded)

129.

Passchier-Vermeer, W. r "Noise-Induced Hearing Loss fromExposureto Intermittent and Varying Noise. " ln: "Proceedingsof the international Congress on Noiseas a Public Health Problem, Dubrovnlk_ Yugoslavia, May 13-18_ 1973. EPA ReportNo. 550/9-73-008, 1973. U.S. EnvironmentalProtection Agency, "Information on Levelsof Environmental No_se Requlslteto Protect Public Health and Welfare with an Adequate Margin of Safety." Report No. 550/9-74-004, March 1974.

130.

(See also numbers 103, 105, 106) PARTG. 131. MEASUREMENTOF iMPULSIVE SOUNDS

Beranok, L. L., "Acoustic Measurements." Chapte_ 10and 11..John Wiley &Sans, Inc. New York (1949). Beranek, L. L. (Ed.), "Noise Reduction." ChaptersS, 6, and 7. McGrawHill BookCompany, Inc. New York (1960). Thiessen,G. J. and Subbaraa, K., 'Effect of Reverberationon Reactionto Impact Noise. " FourthIntarnatlonal Congress on Acoustics, Copenhagen, 21-28 August 1962. Internotianal Electratechnical Commission,"Precision SoundLevel Meters. " Publ_cotion 179_ 1965. Crocker, M. and Sutherland, L. C., "InstrumentationRequirements far Measurement of Sonic Boomand BlastWaves- A Theoretical Study. " J. Sound Vib. 7:3 (1968). Hewlett Packard_ "Acaust;esHandbookApplication Note 100. " November 1968. 8ruel & Kjoans "Instructionsand Applications - Impulse PrecisionSoundLevel Meter Type 2204/S "June (1970). Petersont A. and Gross, E. t "Handbookof Noise Measurement." General Radio, 1972. HempstaoktT. I., Pawell_ J. A._ and Etse, D._ "A Note an the Limitations in the Useof the ImpulsePrecision Sound Level Meter. " Applied Acoustics (5), 141-144 (1972). R-14

132.

133.

134.

135.

136.

t37.

138.

139.

REFERENCES (Concluded) PART G. 140. MEASUREMENT OF iMPULSIVESOUNDS (Concluded)

Olesen, H.P. and Zaveri, K., "Measurements of Averaging Timesof Level Recorders Types2305 and 2307." B & K instruments Technical Review, No. 1, 1974. Kundert, W.R., "The ImpulseSound-LevelMeter - What's It All About?" Sound and Vibration, 50_53, March 1974. Wahrmann, C.G. and Broch, J.T., "On the Averaging Time of RMS Measurements." B& K Insltument_ TechnicalReview, No. 2, 1975. Zaver[, K., "Averaging Timeof Level Recorder Type2306 and 'Fast' and 'Slow' Responso ef Level Recorders 2305/06/07." B & K Insrrumenls Technical Review, No. 2, 1975. Young, R.W. end Cowen, S.J., "Responses oF SoundLevel Maters to Impulsive Sounds." Paperbefore 91st Meeting, Acoustical Social,/oFAmerica, Washington, D.C., April 1976. Martin, R., "The ImpulseSoundLevel Meter and Prope|alsForIts Use in Germany." Paperbefore Inter-Noise 76, Wash[ngten, D.C., 5-7 April 1976. Brael, Per V., "Noise, Do We MeasureIt Correctly?" Noise Control Engineering 8, 52-60 (1977). Schamer, P., "EvaluationoFC-Weighted Ldnfor Assessment of ImpulseNoise." J. Acoust. Soc. Am. 62, 396399 (1977). Referencefor USASI Noise - General Radio, "Type 1382, Random-NoiseGenerator 20 Hz-50 kHz," April 1968.

141.

142.

143.

144.

145.

146.

147.

148.

:"

R-15

TECHNICAL REPORT DATA (Ph'a_" read In_lntctio_lx on the rererse b_/or_" comp/clin1,,}
I,'RI2POF_T NO, 12, ' 3. RECiPiENT'5 ACCESSION NO,

EPA 550/9-79-I03
4,TITLE AND SUBTITLE

5,REPORT

DATE

Annoyance,Loudness,and rfeasurement of Repetitive Type Impulsivetloise Sources

7, AUTHOR{S)

November1979 _.PERFOHMINGOR_ANJZATIONCODE
l L PERFOIIMING ORGANIZATION REPORT NO,

L.C. Sutherland, R.E. Burke

I), PERFORMING ORGANIZATION NAME AND ADDRESS I0. PROGI_AM ELEMENT NO.

Wyle Research El Segundo, California

II.CONTRACT/GRANTNO. 90245 68-01-4694

U.S. Environmental Protection Agency Office of NoiseAbatementand Control (ANR-471) Washington,D.C. 20460
15, 5UPPL_'MENTARY NOTES

,4. SPONSORING AGENCY CODE

_6,ABSTRACT

This study was undertaken to evaluate subjective and objective aspects of moderatelevels of noise from impulsivesources. The study excludedevaluation of hearingdamage risk or annoyancefrom buildlngvibrationby high level impulsive noise, which were coveredby recent r:coJr_endations of che NationalResearchCouncil, Committeeon HearingBioacnustics and Biomechanics, Working Group 69. While the study includedoriginalinvestigations into some of the objectiveaspectsof impulsivenoise,a detailedreview of the literature on the subjective aspectswas emphasized. Based on this availableliterature, _he annoyanceand loudness from a wide variety of repetitive impulsenoises were evaluated. These resultswere applied to the evaluation of impulsivenoise from a numberof specificnoise sources. Based on the most pertinentliterature, it is tentatively concludedthat a subjective impulsecorrection factor of +7 dB applied to theA-weightedequivalent sound levels of these types of repetitiveimpulsivenoise sources would betterdefine their effective level in tems of annoyancereactions, rio additional correctionis identified at this time for crest level or repetition rate. Researchon subjective correction factors for helicopter blade slap is also reviewed and potentialreasonsfor the smallersubjective correctionfactors (i.e.,0 to 6 dB) for annoyanceresponseto (CONTINUED ON BACK OF PAGE)
7, a, DESCRIPTORS KEY WORDS AND DOCUMENT ANALYSIS !I,IDENTIFIEHSIOPEN ENDED TERMS C. COSATI I:Jcl_Group

Noise,Annoyance,ImpulseNoise, Loudness, CalculationMethods

18. DISTRI

BUTION

STATEMENT

19, *_ECURITY

CLASS

(Y_JiI I_'potlj"

21. NO, OF PACES

Limited supplyavailableat ANR-471, I_ashington, D.C. 20460 EPA-ONAC Also ava_lah2p a_ NTI(;
EPA _Ofl_ 222D-I (9-_*_) w Ul gfl_lRl'MIml_r4ltWG01f lIPS ll,[ .3LI-L_/I)O

Unclassified 2D._ECURITyCLAS$(Thi$pa_) 22.PRICE Unclassified

this type of soundare discussed. It is recommended that refinements to this subjective correctionfactorbe based on the use of standardloudnesscalculation methods (Stevens Mark VII or Zwicker)modifiedto include provision for a shortertime constantto reflect subjective responseto short durationimpulsivesounds. The studyalso includeda brief experimental evaluation of the measurement of a wide varietyof simulatedrepetitive impulslve-type signalsvaryinoin duty cycle, repetitionrate, pulse frequency, and ratio of peak impulsesignal levelto continuous backgroundnoise level. When repetitiveimpulses are measuredusing maximumvalues of A-weighted(slow)readingson an ImpulseSoundLevel Meter, no objective correctionis necessary in order to measure,with an accuracyof +I.5 dB, the equivalent level (Leq) of the wide varietyof impulsive signals investigate_.

United SIl+ll+| Environmonlal Pratection AQency WxZshlngtonDC 2044_ ANR-4 71. Olflcial Businezs Penalty fol Plivale LJr_5300

poltage and FlllZ PO_ Environmental PtOlOCllOn Agoncy _PA+3QS

Spatial Fourth.Clau R|te 8OOk