208 7 8MB
English Pages 329 Year 2009
Electroacoustic Devices: Microphones and Loudspeakers
This page intentionally left blank
Electroacoustic Devices: Microphones and Loudspeakers Edited by Glen Ballou
AMSTERDAM•BOSTON•HEIDELBERG•LONDON NEWYORK•OXFORD•PARIS•SANDIEGO SANFRANCISCO•SINGAPORE•SYDNEY•TOKYO FocalPressisanimprintofElsevier
FocalPressisanimprintofElsevier 30CorporateDrive,Suite400,Burlington,MA01803,USA LinacreHouse,JordanHill,OxfordOX28DP,UK © 2009 ELSEVIER Inc. All rights reserved. Nopartofthispublicationmaybereproducedortransmittedinanyformorbyanymeans, electronicormechanical,includingphotocopying,recording,oranyinformationstorageand retrievalsystem,withoutpermissioninwritingfromthepublisher.Detailsonhowtoseek permission,furtherinformationaboutthePublisher’spermissionspoliciesandourarrangements withorganizationssuchastheCopyrightClearanceCenterandtheCopyrightLicensingAgency, canbefoundatourwebsite:www.elsevier.com/permissions. Thisbookandtheindividualcontributionscontainedinitareprotectedundercopyrightbythe Publisher(otherthanasmaybenotedherein). Notices Knowledgeandbestpracticeinthisfieldareconstantlychanging.Asnewresearchandexperience broadenourunderstanding,changesinresearchmethods,professionalpractices,ormedical treatmentmaybecomenecessary. Practitionersandresearchersmustalwaysrelyontheirownexperienceandknowledgein evaluatingandusinganyinformation,methods,compounds,orexperimentsdescribedherein.In usingsuchinformationormethodstheyshouldbemindfuloftheirownsafetyandthesafetyof others,includingpartiesforwhomtheyhaveaprofessionalresponsibility. Tothefullestextentofthelaw,neitherthePublishernortheauthors,contributors,oreditors, assumeanyliabilityforanyinjuryand/ordamagetopersonsorpropertyasamatterofproducts liability,negligenceorotherwise,orfromanyuseoroperationofanymethods,products, instructions,orideascontainedinthematerialherein. Library of Congress Cataloging-in-Publication Data Applicationsubmitted British Library Cataloguing-in-Publication Data AcataloguerecordforthisbookisavailablefromtheBritishLibrary. ISBN:978-0-240-81267-0 ForinformationonallFocalPresspublications visitourwebsiteatwww.elsevierdirect.com Typeset by:diacriTech,Chennai,India 0910111254321 PrintedintheUnitedStatesofAmerica
Contents
Part I Electroacoustic Devices 1. Microphones Glen Ballou, Joe Ciaudelli, Volker Schmitt 1.1 Introduction 1.2 Pickup Patterns 1.2.1 Omnidirectional Microphones 1.2.2 Bidirectional Microphones 1.2.3 Unidirectional Microphones 1.3 Types of Transducers 1.3.1 Carbon Microphones 1.3.2 Crystal and Ceramic Microphones 1.3.3 Dynamic Microphones 1.3.4 Capacitor Microphones 1.3.5 Electret Microphones 1.4 Microphone Sensitivity 1.4.1 Open-Circuit Voltage Sensitivity 1.4.2 Maximum Power Output Sensitivity 1.4.3 Electronic Industries Association (EIA) Output Sensitivity 1.4.4 Various Microphone Sensitivities 1.4.5 Microphone Thermal Noise 1.5 Microphone Practices 1.5.1 Placement 1.5.2 Grounding 1.5.3 Polarity 1.5.4 Balanced or Unbalanced 1.5.5 Impedance 1.6 Miscellaneous Microphones 1.6.1 Pressure Zone Microphones (PZM) 1.6.2 Lavalier Microphones 1.6.3 Head-Worn Microphones 1.6.4 Base Station Microphones 1.6.5 Differential Noise-Canceling Microphones 1.6.6 Controlled-Reluctance Microphones 1.6.7 Handheld Entertainer Microphones 1.6.8 Pressure-Gradient Condenser Microphones 1.6.9 Interference Tube Microphones 1.6.10 Rifle Microphones 1.6.11 Parabolic Microphones
4 5 5 9 10 23 23 24 26 33 48 49 49 50 51 53 53 54 54 56 57 59 59 61 61 69 72 75 76 77 77 79 82 89 90
v
vi
Contents
1.6.12 Zoom Microphones 1.6.13 Automatic Microphone Systems 1.6.14 PolarFlex™ Microphone System 1.7 Stereo Microphones 1.7.1 Coincident Microphones 1.7.2 XY Stereo Technique 1.7.3 The ORTF Technique 1.7.4 The M/S Stereo Technique 1.7.5 The Stereo Boom Microphone 1.7.6 SASS Microphones 1.7.7 Surround Sound Microphone System 1.8 Microphones for Binaural Recording 1.8.1 Artificial Head Systems 1.8.2 In the Ear Recording Microphones 1.9 USB Microphones 1.10 Wireless Communication Systems 1.10.1 Criteria for Selecting a Wireless Microphone 1.10.2 Receiving Antenna Systems 1.10.3 Companding 1.10.4 Waterproof Wireless Microphone Systems 1.11 Multichannel Wireless Microphone and Monitoring Systems 1.11.1 Introduction 1.11.2 Frequencies 1.11.3 Spacing 1.11.4 Frequency Deviation 1.11.5 Frequency Coordination 1.11.6 Transmitter Considerations 1.11.7 Receiver Considerations 1.11.8 Antennas 1.11.9 Wireless Monitor Systems 1.11.10 System Planning for Multichannel Wireless Systems 1.11.11 Future Considerations: Digital Wireless Transmission 1.11.12 Conclusion 1.12 Microphone Accessories 1.12.1 Inline Microphone Processors 1.12.2 Windscreens and Pop Filters 1.12.3 Shock Mounts 1.12.4 Stands and Booms 1.12.5 Attenuators and Equalizers 1.13 Microphone Techniques 1.13.1 Stereo Micing Techniques 1.13.2 Microphone Choice 1.13.3 Microphone Characteristics 1.13.4 Specific Micing Techniques 1.13.5 Conclusion Acknowledgments References Bibliography
91 94 95 98 98 100 103 104 111 113 115 120 120 124 126 127 129 134 137 138 139 139 140 140 140 140 143 146 147 151 152 154 155 156 156 160 165 169 170 171 173 174 174 176 192 192 192 193
Contents
vii
2. Loudspeakers Jay Mitchell 2.1 Introduction 2.1.1 Uses of Loudspeakers 2.1.2 Loudspeaker Components 2.2 Transducer Types 2.2.1 Electrodynamic Transducers 2.2.2 Diaphragm Types 2.2.3 Suspension Methods 2.2.4 Mechanical Construction 2.3 Compression Drivers 2.4 Electrostatic Transducers 2.5 Piezoelectric Loudspeakers 2.6 Motor Design Considerations 2.6.1 Output Limitations 2.6.2 Heat Transfer Designs for High-Power Woofers 2.7 Radiator Types 2.7.1 Direct Radiators 2.7.2 Cone Radiators 2.7.3 Dome Radiators 2.7.4 Ring Radiators 2.7.5 Panel Radiators 2.7.6 Horns 2.8 Loudspeaker Systems 2.8.1 Configuration Choices 2.8.2 Types of Loudspeaker Systems 2.8.3 Performance Issues in Multiway Systems 2.8.4 Line Arrays 2.8.5 Crossovers 2.8.6 Acoustic Boundaries 2.8.7 Conclusion 2.9 Characterization of Loudspeaker Performance 2.9.1 Motivation 2.9.2 Efficiency and Sensitivity 2.9.3 Network Transfer Function 2.9.4 Loudspeaker Transfer Function 2.9.5 Impedance 2.9.6 Distortion 2.9.7 Characterization for Design Purposes 2.9.8 Characterization for the User 2.10 Direct Radiation of Sound 2.10.1 Acoustics of Radiators 2.10.2 Direct Radiator Enclosure Design 2.10.3 Horns 2.11 Loudspeaker Testing and Measurement 2.11.1 Linear Transfer Function 2.11.2 Chart Recorders
196 196 197 197 198 199 201 201 203 206 210 213 215 219 221 221 221 222 222 223 223 234 235 236 237 241 243 249 251 252 252 252 253 253 255 257 257 258 261 261 265 270 271 271 272
viii
Contents
2.11.3 Real Time Analyzers 2.11.4 Time-Windowed Measurements 2.11.5 Swept Sine Measurements Reference Bibliography
273 274 276 277 277
3. Loudspeaker Cluster Design Ralph Heinz 3.1 3.2 3.3 3.4 3.5
Why Array? Array Problems and Partial Solutions: A Condensed History Conventional Array Shortcomings Conventional Array Shortcoming Analysis Coincident Acoustical Centers: A Practical Approach 3.5.1 TRAP Horns: A New Approach 3.5.2 TRAP Performance 3.6 Low Frequency Arrays: Beneficial Interference 3.6.1 Horizontal Woofer Arrays: Maintaining Wide Dispersion 3.6.2 Vertical Woofer Arrays 3.7 Line Arrays and Digitally Steerable Loudspeaker Column Arrays 3.7.1 What Affects Intelligibility 3.7.2 Measuring Intelligibility 3.7.3 Architecture and Room Acoustics 3.7.4 Line Arrays 3.7.5 DSP-Driven Vertical Arrays 3.7.6 Multichannel DSP Can Control Array Height 3.7.7 Steerable Arrays May Look Like Columns But They Are Not
Index
279 280 280 284 286 286 289 289 289 291 292 293 294 295 296 300 306 306
311
Part I
Electroacoustic Devices
This page intentionally left blank
Section 1
Microphones Glen Ballou, Joe Ciaudelli, Volker Schmitt
1.1 Introduction 1.2 Pickup Patterns 1.2.1 Omnidirectional Microphones 1.2.2 Bidirectional Microphones 1.2.3 Unidirectional Microphones 1.3 Types of Transducers 1.3.1 Carbon Microphones 1.3.2 Crystal and Ceramic Microphones 1.3.3 Dynamic Microphones 1.3.4 Capacitor Microphones 1.3.5 Electret Microphones 1.4 Microphone Sensitivity 1.4.1 Open-Circuit Voltage Sensitivity 1.4.2 Maximum Power Output Sensitivity 1.4.3 Electronic Industries Association (EIA) Output Sensitivity 1.4.4 Various Microphone Sensitivities 1.4.5 Microphone Thermal Noise 1.5 Microphone Practices 1.5.1 Placement 1.5.2 Grounding
1.5.3 Polarity 1.5.4 Balanced or Unbalanced 1.5.5 Impedance 1.6 Miscellaneous Microphones 1.6.1 Pressure Zone Microphones (PZM) 1.6.2 Lavalier Microphones 1.6.3 Head-Worn Microphones 1.6.4 Base Station Microphones 1.6.5 Differential NoiseCanceling Microphones 1.6.6 Controlled-Reluctance Microphones 1.6.7 Handheld Entertainer Microphones 1.6.8 Pressure-Gradient Condenser Microphones 1.6.9 Interference Tube Microphones 1.6.10 Rifle Microphones 1.6.11 Parabolic Microphones 1.6.12 Zoom Microphones 1.6.13 Automatic Microphone Systems 1.6.14 PolarFlex™ Microphone System 1.7 Stereo Microphones 1.7.1 Coincident Microphones 1.7.2 XY Stereo Technique
Electroacoustic Devices: Microphones and Loudspeakers Copyright©2009byFocalPress,Inc.Allrightsofreproductioninanyformreserved. Sections1,2,and3wereoriginallypublishedinHandbookforSoundEngineers,FourthEditionbyGlenBallou.
3
4
PART | I
1.7.3 The ORTF Technique 1.7.4 The M/S Stereo Technique 1.7.5 The Stereo Boom Microphone 1.7.6 SASS Microphones 1.7.7 Surround Sound Microphone System 1.8 Microphones for Binaural Recording 1.8.1 Artificial Head Systems 1.8.2 In the Ear Recording Microphones 1.9 USB Microphones 1.10 Wireless Communication Systems 1.10.1 Criteria for Selecting a Wireless Microphone 1.10.2 Receiving Antenna Systems 1.10.3 Companding 1.10.4 Waterproof Wireless Microphone Systems 1.11 Multichannel Wireless Microphone and Monitoring Systems 1.11.1 Introduction 1.11.2 Frequencies 1.11.3 Spacing 1.11.4 Frequency Deviation 1.11.5 Frequency Coordination 1.11.6 Transmitter Considerations
Electroacoustic Devices
1.11.7 Receiver Considerations 1.11.8 Antennas 1.11.9 Wireless Monitor Systems 1.11.10 System Planning for Multichannel Wireless Systems 1.11.11 Future Considerations: Digital Wireless Transmission 1.11.12 Conclusion 1.12 Microphone Accessories 1.12.1 Inline Microphone Processors 1.12.2 Windscreens and Pop Filters 1.12.3 Shock Mounts 1.12.4 Stands and Booms 1.12.5 Attenuators and Equalizers 1.13 Microphone Techniques 1.13.1 Stereo Micing Techniques 1.13.2 Microphone Choice 1.13.3 Microphone Characteristics 1.13.4 Specific Micing Techniques 1.13.5 Conclusion Acknowledgments References Bibliography
1.1 INTRODUCTION All sound sources have different characteristics; their waveform varies, their phasecharacteristicsvary,theirdynamicrangeandattacktimevary,andtheir frequencyresponsevaries,justtonameafew.Noonemicrophonewillreproduce allofthesecharacteristicsequallywell.Infact,eachsoundsourcewillsoundbetterormorenaturalwithonetypeorbrandofmicrophonethanallothers.Forthis reasonwehaveandalwayswillhavemanytypesandbrandsofmicrophones.
Section | 1
Microphones
5
Microphonesareelectroacousticdevicesthatconvertacousticalenergyinto electricalenergy.Allmicrophoneshaveadiaphragmormovingsurfacethatis excitedbytheacousticalwave.Thecorrespondingoutputisanelectricalsignal thatrepresentstheacousticalinput. Microphones fall into two classes: pressure and velocity. In a pressure microphonethediaphragmhasonlyonesurfaceexposedtothesoundsource sotheoutputcorrespondstotheinstantaneoussoundpressureoftheimpressed soundwaves.Apressuremicrophoneisazero-order gradient microphone,and isassociatedwithomnidirectionalcharacteristics. Thesecondclassofmicrophoneisthe velocitymicrophone,alsocalleda first-order gradient microphone,wheretheeffectofthesoundwaveisthedifferenceorgradientbetweenthesoundwavethathitsthefrontandtherearof thediaphragm.Theelectricaloutputcorrespondstotheinstantaneousparticle velocityintheimpressedsoundwave.Ribbonmicrophonesaswellaspressure microphonesthatarealteredtoproducefront-to-backdiscriminationareofthe velocitytype. Microphones are also classified by their pickup pattern or how they discriminatebetweenthevariousdirectionsthesoundsourcecomesfrom,Fig.1.1. Theseclassificationsare: ● ●
●
Omnidirectional—pickupisequalinalldirections. Bidirectional—pickupisequalfromthetwooppositedirections(180°)apart andzerofromthetwodirectionsthatare90°fromthefirst. Unidirectional—pickup is from one direction only, the pickup appearing cardioidorheart-shaped.
Theairparticlerelationshipsoftheairparticledisplacement,velocity,and accelerationthatamicrophoneseesasaplanewaveinthefarfield,areshown inFig.1.2.
1.2 PICKUP PATTERNS Microphonesaremadewithsingle-ormultiple-pickuppatternsandarenamed bythepickuppatterntheyemploy.Thepickuppatternsanddirectionalresponse characteristicsofthevarioustypesofmicrophonesareshowninFig.1.1.
1.2.1 Omnidirectional Microphones Theomnidirectional,orspherical,polarresponseofthepressuremicrophones, Fig.1.3,iscreatedbecausethediaphragmisonlyexposedtotheacousticwave on the front side. Therefore, no cancellations are produced by having sound waveshittingboththefrontandrearofthediaphragmatthesametime. Omnidirectionalmicrophonesbecomeincreasinglydirectionalasthediameterofthemicrophonereachesthewavelengthofthefrequencyinquestion,as showninFig.1.4;therefore,themicrophoneshouldhavethesmallestdiameter
Hypercardioid
2
FIGURE 1.1 Performancecharacteristicsofvariousmicrophones.
1808
1568
1408
PART | I
1208
1008
Pickup angle (2 u) for 6 dB attenuation
1168
2
Pickup angle (2 u) for 3 dB attenuation
1308
2 1.9
1.7
1.7
1
Equivalent distance 908
7 14
7
0.87
1
0.93
2
1
0.67
3.8
Front random response Back random response
0.5
1
1
Supercardioid
E E E 5 0 (1 1 cos u) E 5E0 [(321) 1(323)cos u] E 5 0(1 13cos u) 2 2 4 33 25 27
Directional
0.5
33
E 5 E0 cos u
Bidirectional
100
E 5 E0
Omnidirectional
Front random response Total random response
Random energy efficiency (%) Front response Back response
Voltage output
Directional Response Characteristics
Microphone
6 Electroacoustic Devices
Section | 1
7
Microphones
b a 5 Displacement 0 b 5 Displacement max e c 5 Displacement 0 d 5 Displacement max e 5 Displacement 0
c
a Particle displacement 908 1808
d
a 5 Velocity max b 5 Velocity 0 c 5 Velocity max d 5 Velocity 0 e 5 Velocity max
Particle velocity
2708 3608
Particle acceleration
a 5 Acceleration 0 b 5 Acceleration max c 5 Acceleration 0 d 5 Acceleration max e 5 Acceleration 0
FIGURE 1.2 Air particle motion in a sound field, showing relationship to velocity and acceleration.
FIGURE 1.3
Omnidirectionalpickuppattern.CourtesyShureIncorporated.
possibleifomnidirectionalcharacteristicsarerequiredathighfrequencies.The characteristicthatallowswavestobendaroundobjectsisknownasdiffraction andhappenswhenthewavelengthislongcomparedtothesizeoftheobject.
8
PART | I
Electroacoustic Devices
1kHz 10 kHz Sealed microphone 1kHz 10 kHz FIGURE 1.4
High-frequencydirectivityofanomnidirectionalmicrophone.
As the wavelength approaches the size of the object, the wave cannot bend sharplyenoughand,therefore,passesbytheobject.Thevariousresponsesstart todivergeatthefrequencyatwhichthediameterofthediaphragmofthemicrophone,D,isapproximatelyone-tenththewavelength,λ,ofthesoundarriving asequation D=
λ 10
(1.1)
Thefrequency,f,atwhichthevariationbeginsis f=
v 10D
(1.2)
where, visthevelocityofsoundinfeetpersecond,ormeterspersecond, Disthediameterofthediaphragminfeetormeters. Forexample,a1⁄2inch(1.27cm)microphonewillbegintovaryfromomnidirectional,thoughonlyslightly,at 1130 0.5 c 10 ´ m 12 = 2712 Hz
f=
andwillbedownapproximately3dBat10,000Hz. Omnidirectionalmicrophonesarecapableofhavingaveryflat,smoothfrequencyresponseovertheentireaudiospectrumbecauseonlythefrontofthe diaphragmisexposedtothesource,eliminatingphasecancellationsfoundin unidirectionalmicrophones. For smoothness of response the smaller they are, the better.The problem usually revolves around the smallest diaphragm possible versus the lowest signal-to-noiseratio,SNR,orputanotherway,thesmallerthediaphragm,the lowerthemicrophonesensitivity,therefore,thepoorertheSNR. Omnidirectionalmicrophoneshaveverylittleproximityeffect.SeeSection 1.2.3.1foradiscussiononproximityeffect.
Section | 1
Microphones
9
Because the pickup pattern is spherical, the random energy efficiency is 100%,andtheratiooffrontresponsetobackorsideis1:1;therefore,signals fromthesidesorrearwillhavethesamepickupsensitivityasfromthefront, giving a directivity index of 0dB. This can be helpful in picking up wanted roomcharacteristicsorconversationsaroundatableaswhenrecordingasymphony.However,itcanbedetrimentalwheninanoisyenvironment. Omnidirectional microphones are relatively free from mechanical shock becausetheoutputatallfrequenciesishigh;therefore,thediaphragmcanbe stiff.This allows the diaphragm to follow the magnet or stationary system it operatesagainstwhensubjectedtomechanicalmotion(seeSection1.3.3).
1.2.2 Bidirectional Microphones A bidirectional microphoneisonethatpicksupfromthefrontandbackequally wellwithlittleornopickupfromthesides.Thefieldpattern,Fig.1.5,iscalled a figure 8. Becausethemicrophonediscriminatesbetweenthefront,back,andsides, randomenergyefficiencyis33%.Inotherwords,backgroundnoise,ifitisina reverberantfield,willbe67%lowerthanwithanomnidirectionalmicrophone. The front-to-back response will still remain one; however, the front-to-side responsewillapproachinfinity,producingadirectivityindexof4.8.Thiscan beextremelyusefulwhenpickinguptwoconversationsonoppositesidesof a table. Because of the increased directional capabilities of the microphone, pickup distance is 1.7 times greater before feedback in the direct field than foranomnidirectionalmicrophone.Theincludedpickupconeangleshownin Fig.1.6for6dBattenuationonaperfectbidirectionalmicrophoneis120°off thefrontofthemicrophoneand120°offtherearofthemicrophone.Because of diffraction, this angle varies with frequency, becoming narrower as the frequencyincreases.
FIGURE 1.5
Bidirectionalpickuppattern.CourtesySennheiserElectronicCorporation.
10
PART | I
108
08
108 208
208 308 408
508
Electroacoustic Devices
0 24 28
308 408
508
Decibels 212 608
608 216
708
708 220 808
808
908
908
1008 1108
Mic Directional characteristic of a typical velocity microphone
1008 1108
10,000 Hz 6000 Hz 1000 Hz
FIGURE 1.6 Polar pattern of a typical bidirectional ribbon velocity microphone showing the narrowingpatternathighfrequencies.
1.2.3 Unidirectional Microphones Unidirectional microphoneshaveagreatersensitivitytosoundpickupfromthe frontthananyotherdirection. Theaverageunidirectionalmicrophonehasafront-to-backratioof20–30dB; thatis,ithas20–30dBgreatersensitivitytosoundwavesapproachingfromthe frontthanfromtherear. Unidirectional microphones are usually listed as cardioid or directional, Fig.1.7,supercardioid,Fig.1.8,orhypercardioid,Fig.1.9.Thepickuppattern iscalledcardioidbecauseitisheartshaped.Unidirectionalmicrophonesarethe most commonly used microphones because they discriminate between signal andrandomunwantednoise.Thishasmanyadvantagesincluding: ● ● ●
Lessbackgroundnoise, Moregainbeforefeedbackespeciallywhenusedinthedirectfield, Discriminationbetweensoundsources.
Section | 1
Microphones
FIGURE 1.7
Cardioidpickuppattern.CourtesyShureIncorporated.
FIGURE 1.8
Supercardioidpickuppattern.CourtesyShureIncorporated.
FIGURE 1.9
Hypercardioidpickuppattern.CourtesySennheiserElectronicCorporation.
11
12
PART | I
Electroacoustic Devices
Thecardioidpatterncanbeproducedbyoneoftwomethods: 1. The first method combines the output of a pressure diaphragm and a pressure-gradient diaphragm, as shown in Fig.1.10. Since the pressuregradientdiaphragmhasabidirectionalpickuppatternandthepressurediaphragm has an omnidirectional pickup pattern, the wave hitting the front ofthediaphragmsadds,whilethewavehittingtherearofthediaphragm cancelsasitis180°out-of-phasewiththerearpickuppatternofthepressure diaphragm.Thismethodisexpensiveandseldomusedforsoundreinforcementorgeneral-purposemicrophones. 2. Thesecondandmostwidelyusedmethodofproducingacardioidpattern istouseasinglediaphragmandacousticallydelaythewavereachingthe rearofthediaphragm.Whenawaveapproachesfromthefrontofthediaphragm,itfirsthitsthefrontandthentherearofthediaphragmaftertravelingthroughanacousticaldelaycircuit,asshowninFig.1.11A.Thepressure onthefrontofthediaphragmisat0°whileontherearofthediaphragmitis someanglebetween0°and180°,asshowninFig.1.11B.Iftherearpressure wasat0°,theoutputwouldbe0.Itwouldbeidealiftherearpressurewere at180°sothatitcouldaddtotheinput,doublingtheoutput. Thephaseinversioniscausedbytheextradistancethewavehastotravel toreachthebackofthediaphragm.Whenthewaveiscomingfromtherearof
Input
Front
Output
Diaphragm movement
Sound wave
Input
Output
Rear Sound wave
Pressure gradient cartridge Pressure cartridge
FIGURE 1.10
Two-diaphragmcardioidmicrophone.
Section | 1
13
Microphones
Acoustical delay
Front Output
Rear A. Ideal
Front
Output
Rear B. Normal FIGURE 1.11
Cardioidmicrophoneemployingacousticaldelay.
themicrophone,ithitsthefrontandbackofthediaphragmatthesametimeand withthesamepolarity,thereforecancelingtheoutput. The frequency response of cardioid microphones is usually rougher than an omnidirectional microphone due to the acoustical impedance path and its effectsonthefrontwaveresponse.Thefrontandrearresponsesofacardioid microphonearenotthesame.Althoughthefrontpatternmaybeessentiallyflat overtheaudiospectrum,thebackresponseusuallyincreasesatlowandhigh frequencies,asshowninFig.1.12. Discriminationbetweenthefrontandbackresponseisbetween15and30dB inthemidfrequenciesandcouldbeaslittleas5–10dBattheextremeends,as showninFig.1.12.
1.2.3.1 Proximity Effects Asthesourceismovedclosertothediaphragm,thelow-frequencyresponse increasesduetotheproximityeffect,Fig.1.13.Theproximityeffect1iscreated because at close source-to-microphone distance, the magnitude of the
14
PART | I
110 0 210 dB 220 230 240
Front Rear 100
50
500 1K Frequency–Hz
5K 10K 20K
Frequencyresponseofatypicalcardioidmicrophone.
Relative response–dB
FIGURE 1.12
Electroacoustic Devices
15 0 25
60
Microphone A 240 120 720
10.5 dB
120
15 0
60
Microphone B 249
25
720
9.5 dB
20
50
100
1K Frequency–Hz
FIGURE 1.13 Proximityeffectvariationsinresponsetodistancebetweensourceandmicrophone forcardioidmicrophones.CourtesyTelexElectro-Voice.
soundpressureonthefrontisappreciablygreaterthanthesoundpressureon therear.InthevectordiagramshowninFig.1.14A,thesoundsourcewasa distancegreaterthan2ftfromthemicrophone.Theangle2KDisfoundfrom D,whichistheacousticdistancefromthefronttotherearofthediaphragm and K = 2π/λ. Figure1.14B shows the vector diagram when used less than 4inchestothesoundsource. Inbothcases,forceF1,thesoundpressureonthefrontofeachdiaphragm, isthesame.ForceF2istheforceonthebackofthediaphragmwhenthemicrophoneisusedatadistancefromthesoundsource,andF0istheresultantforce. TheforceF0′onthebackofthediaphragmiscreatedbyaclosesoundsource. Laterally,thevectorsumF0′isconsiderablylargerinmagnitudethanF0and thereforeproducesgreateroutputfromthemicrophoneatlowfrequencies.This canbeadvantageousordisadvantageous.Itisparticularlyusefulwhenvocalistswanttoaddlow-frequencytotheirvoiceoraninstrumentalisttoaddlow frequencies to the instrument. This is accomplished by varying the distance betweenthemicrophoneandthesoundsource,increasingbassasthedistance decreases.
Section | 1
15
Microphones
Fg
F0 2KD 0 F1 A. With the sound source at a distance from the microphone.
Fg
F0 F0 F2 F1 0 B. With the sound source close to the microphone.
FIGURE 1.14
Vectordiagramofaunidirectionalmicrophone.CourtesyTelexElectro-Voice.
1.2.3.1.1 Frequency Response Frequencyresponseisanimportantspecificationofunidirectionalmicrophones andmustbecarefullyanalyzedandinterpretedintermsofthewaythemicrophoneistobeused.Ifajudgmentastothesoundqualityofthemicrophoneis madestrictlyfromasingleon-axisresponse,theinfluenceoftheproximityeffect andoff-axisresponsewouldprobablybeoverlooked.Acomparisonoffrequency responseasafunctionofmicrophone-to-sourcedistancewillrevealthatallunidirectionalmicrophonesexperienceacertainamountofproximityeffect.Inorder toevaluateamicrophone,thisvariationwithdistanceisquiteimportant. Whenusingaunidirectionalmicrophone2inahandheldorstand-mounted configuration, it is conceivable that the performer will not always remain exactlyonaxis.Variationsof±45°oftenoccur,andsoaknowledgeoftheuniformityofresponseoversucharangeisimportant.Thenatureoftheseresponse variations is shown in Fig.1.15. Response curves such as these give a better indicationofthistypeofoff-axisperformancethanpolarresponsecurves.The polarresponsecurvesarelimitedinthattheyareusuallygivenforonlyafew frequencies,therefore,thecompletespectrumisdifficulttovisualize. For applications involving feedback control or noise rejection, the polar response or particular off-axis response curves, such as at 135° or 180°, are important.Thesecurvescanoftenbemisleadingduetotheacousticconditions andexcitationsignalsused.Suchmeasurementsareusuallymadeunderanechoic conditionsatvariousdistanceswithsine-waveexcitation.Lookingsolelyata rearresponsecurveasafunctionoffrequencyismisleadingsincesuchacurve doesnotindicatethepolarcharacteristicatanyparticularfrequency,butonly thelevelatoneangle.Suchcurvesalsotendtogivetheimpressionofarapidly fluctuating high-frequency discrimination. This sort of performance is to be expectedsinceitisvirtuallyimpossibletodesignamicrophoneofpracticalsize withaconstantangleofbestdiscriminationathighfrequencies,Fig.1.16.The principalfactorinfluencingthisvariationinrearresponseisdiffraction,which
16 Re at ve response–dB
PART | I
Electroacoustic Devices
5dB 08 108 208 308 408 508
20
50 100
1000 Frequency–Hz
10K20K
Re at ve response–dB
FIGURE 1.15 Variations in front response versus angular position. Note: Curves have been displacedby2.5dBforcomparisonpurposes.
0°
0 210 220
180°
230 20
50 100
1000 Frequency–Hz
10K 20K
FIGURE 1.16 Typical fluctuations in high-frequency rear response for a cardioid microphone. CourtesyShureIncorporated.
iscausedbythephysicalpresenceofthemicrophoneinthesoundfield.This diffractioneffectisfrequencydependentandtendstodisrupttheidealperformanceoftheunidirectionalphase-shiftelements. To properly represent this high-frequency off-axis performance, a polar responsecurveisofvalue,butit,too,canbeconfusingathighfrequencies.The reason for this confusion can be seen in Fig.1.17, where two polar response curves only 20Hz apart are shown.The question that arises then is how can such performance be properly analyzed?A possible solution is to run polar responsecurveswithbandsofrandomnoisesuchas 1⁄3octavesofpinknoise. Randomnoiseisusefulbecauseofitsaveragingabilityandbecauseitsamplitudedistributioncloselyresemblesprogrammaterial. Anechoicmeasurementsareonlymeaningfulaslongasnolargeobjectsare incloseproximitytothemicrophone.Thepresenceofthehumanheadinfront ofamicrophonewillseriouslydegradetheeffectivehigh-frequencydiscrimination.AnexampleofsuchdegradationcanbeseeninFig.1.18whereahead objectwasplaced2inches(5cm)infrontofthemicrophone.(Thetwocurves havenotbeennormalized.)Thissortofperformanceresultsfromtheheadasa reflectorandisacommoncauseoffeedbackasoneapproachesamicrophone. Thisshouldnotbeconsideredasashortcomingofthemicrophone,butrather
Relative response–dB
Section | 1
17
Microphones
0 3130 Hz
210
3150 Hz
220 230 0
40
80
120 160 200 240 280 Angular position–degrees
320 360
Relative response–dB
FIGURE 1.17 An example of rapid variations in high-frequency polar response for singlefrequencyexcitation.CourtesyShureIncorporated.
0 �10 �20 �30 0
40
80
120 160 200 240
280
320 360
Angular position–degrees FIGURE 1.18 Anexampleofaheadobstacleonapolarresponse.CourtesyShureIncorporated.
asanunavoidableresultofthesoundfieldinwhichitisbeingused.At180°, forexample,themicrophonesees,inadditiontothesourceitistryingtoreject, areflectionofthatsourcesome2inches(5cm)infrontofitsdiaphragm.This phenomenonisgreatlyreducedatlowfrequenciesbecausetheheadisnolonger anappreciableobstacletothesoundfield.Itisthusclearthattheeffectivediscriminationofanyunidirectionalmicrophoneisgreatlyinfluencedbythesound fieldinwhichitisused. 1.2.3.1.2 Types of Cardioid Microphones Cardioidmicrophonesarenamedbythewaysoundenterstherearcavity.The sound normally enters the rear of the microphone’s cavity through single or multipleholesinthemicrophonehousing,asshowninFig.1.19. 1.2.3.1.3 Single-Entry Cardioid Microphones Allsingle-entrantcardioidmicrophoneshavetherearentranceportlocatedat onedistancefromtherearofthediaphragm.Theportlocationisusuallywithin 11⁄2 inches(3.8cm)ofthediaphragmandcancausealargeproximityeffect. TheElectro-VoiceDS35isanexampleofasingle-entrantcardioidmicrophone, Fig.1.20.
18
PART | I
Electroacoustic Devices
Source A. Single-entry microphone. Front
Rear B. Three-entry microphone.
C. Multiple-entry microphone. FIGURE 1.19 Threetypesofcardioidmicrophones.
FIGURE 1.20 Electro-VoiceDS35single-entrantmicrophone.CourtesyElectro-Voice,Inc.
The low-frequency response of the DS35 varies as the distance from the soundsourcetothemicrophonedecreases,Fig.1.21.Maximumbassresponse isproducedinclose-upusewiththemicrophone11⁄2 inches(3.8cm)fromthe soundsource.Minimumbassresponseisexperiencedatdistancesgreaterthan 2ft(0.6m).Usefuleffectscanbecreatedbyimaginativeapplicationofthevariablelow-frequencyresponse. Anothersingle-entrantmicrophoneistheShureSM81.3Theacousticalsystemofthemicrophoneoperatesasafirst-ordergradientmicrophonewithtwo soundopenings,asshowninFig.1.22.Figure1.22showsasimplifiedcrosssectionalviewofthetransducer,andFig.1.23indicatesthecorrespondingelectricalanalogcircuitofthetransducerandpreamplifier. Onesoundopening,whichisexposedtothesoundpressurep1,isrepresented bythefrontsurfaceofthediaphragm.Theothersoundopening,orrearentry, consistsofanumberofwindowsinthesideofthetransducerhousingwhere thesoundpressurep2prevails.ThediaphragmhasanacousticalimpedanceZ0, whichalsoincludestheimpedanceofthethinairfilmbetweenthediaphragm and backplate.The sound pressure p2 exerts its influence on the rear surface
19
Microphones
Response–dB
Section | 1
20
100 200 500 1k 2k Frequency–Hz
5k 10k 20k
FIGURE 1.21 FrequencyresponseversusdistanceforanElectro-VoiceDS35single-entrantcardioidmicrophone.CourtesyElectro-Voice,Inc.
V1
Blackplate
Air film R2L2
R1L1
P1
Diagram
Spacer
V2
Transducer housing
P2
FIGURE 1.22 Simplifiedcross-sectionalviewoftheShureSM81condensertransducer.Courtesy ShureIncorporated.
CA ET CP
C1
EP CP
Transducer Connector
RP
Preamplifier
FIGURE 1.23 ElectricalequivalentcircuitoftheShureSM81condensertransducerandpreamplifier.CourtesyShureIncorporated.
20
PART | I
Electroacoustic Devices
ofthediaphragmviaascreenmountedinthesidewindowsofthetransducer housing, having a resistance R1 and inertance L1, through the cavity V1 with complianceC1.AsecondscreenhasaresistanceR2andinertanceL2,through asecondcavityV2withcomplianceC2,andfinallythroughtheperforationsin thebackplate. The combination of circuit elements L1, R1, C1, L2, R2, C2 forms a laddernetworkwithlossyinertances,andiscalledalossy ladder network.The transfercharacteristicofthisnetworkenforcesatimedelayonthepressurep2 impartingdirectional(cardioid)characteristicsforlowandmediumfrequencies.Athighfrequenciestheattenuationcausedbythenetworkislarge,and theresultingpressurearrivingatthebackofthediaphragmduetop2issmall. Themicrophonethenoperatesmuchlikeanomnidirectionalsystemunderthe predominant influence of p1.At these frequencies directional characteristics are attained by diffraction of the sound around a suitably shaped transducer housing. A rotary low-frequency response shaping switch allows the user to select between a flat and a 6dB/octave roll-off at 100Hz or an 18dB/octave cutoff at80Hz.The100Hzroll-offcompensatesfortheproximityeffectassociated witha6inches(15cm)sourcetomicrophonedistance,whilethe80Hzcutoff significantlyreducesmostlow-frequencydisturbanceswithminimaleffecton voicematerial.Intheflatpositionthemicrophonehasa6dB/octaveelectronic infrasonicroll-off,with−3dBat10Hztoreducetheeffectsofinaudiblelowfrequencydisturbancesonmicrophonepreamplifierinputs.Attenuationisprovidedforoperationathighsoundpressurelevels(to145dBSPL)bymeansof arotarycapacitiveswitch(seeSection1.3.4.1). AfinalexampleofasingleentrycardioidmicrophoneistheShureBeta57 supercardioiddynamicmicrophone,Fig.1.24.BoththeShureBeta57andthe Beta58useneodymiummagnetsforhotteroutputandincorporateanimproved shockmount. 1.2.3.1.4 Three-Entry Cardioid Microphones The Sennheiser MD441 is an example of a three-entry cardioid microphone, Fig.1.25.Thelow-frequencyrearentryhasad(distancefromcenterofdiaphragmtotheentryport)ofabout2.8inches(7cm),themid-frequencyentryd isabout2.2inches(5.6cm),andthehigh-frequencyentrydisabout1.5inches (3.8cm),withthetransitioninfrequencyoccurringbetween800Hzand1kHz. Eachentryconsistsofseveralholesaroundthemicrophonecaseratherthana singlehole. Thisconfigurationisusedforthreereasons.Byusingamultiplearrangementofentryholesaroundthecircumferenceofthemicrophonecaseintothe low-frequency system, optimum front response and polar performance can be maintained, even though most of the entries may be accidentally covered whenthemicrophoneishandheldorstandmounted.Themicrophonehasgood proximity performance because the low-frequency entry ports are far from
Section | 1
Microphones
21
FIGURE 1.24 ShureBeta57dynamicmicrophone.CourtesyShureIncorporated.
FIGURE 1.25 Sennheiser MD441 three-entry cardioid microphone. Courtesy Sennheiser ElectronicCorporation.
thediaphragm(4.75in)aswellasthehigh-frequencyentryhavingverylittle proximityinfluenceatlowfrequencies.Thetwo-entryconfigurationhasacardioidpolarresponsepatternthatprovidesawidefrontworkingangleaswellas excellentnoiserejectionandfeedbackcontrol.
22
PART | I
Electroacoustic Devices
1.2.3.1.5 Multiple-Entry Cardioid Microphones The Electro-Voice RE20 ContinuouslyVariable-D microphone, Fig.1.26, is anexampleofmultiple-entrymicrophones.Multiple-entrymicrophoneshave manyrearentranceports.Theycanbeconstructedassingleports,allatadifferent distance from the diaphragm, or as a single continuous opening port. Eachentranceistunedtoadifferentbandoffrequencies,theportclosestto thediaphragmbeingtunedtothehighfrequencies,andtheportfarthestfrom thediaphragmbeingtunedtothelow-frequencyband.Thegreatestadvantage ofthisarrangementisreduced-proximityeffectbecauseofthelargedistance between the source and the rear entry low-frequency port, and mechanical crossovers are not as sharp and can be more precise for the frequencies in question. Asinmanycardioidmicrophones,theRE20hasalow-frequencyroll-off switch to reduce the proximity effect when close micing. Figure 1.27 shows thewiringdiagramoftheRE20.Bymovingtheredwiretoeitherthe150Ω or50Ωtap,themicrophoneoutputimpedancecanbechanged.Notethe“bass tilt”switchthat,whenopen,reducestheseriesinductanceand,therefore,the low-frequencyresponse.
FIGURE 1.26 Electro-Voice RE20 multiple-entry (variable-D cardioid microphone). Courtesy TelexElectro-Voice.
Section | 1
23
Microphones
Yel 0.56 mF Blu
Case gnd
Blk
Shield
Red
150 V
0.2 mH
Red
50 V Blk
Yel
Red
Bass tilt
Blu
Cartridge Grn
140 mH Yel
Grn
FIGURE 1.27 Electro-VoiceRE20cardioidwiringdiagram.Note“basstilt”switchcircuitand outputimpedancetaps.CourtesyTelexElectro-Voice.
1.2.3.1.6 Two-Way Cardioid Microphones Inatwo-waymicrophonesystem,thetotalresponserangeisdividedbetween ahigh-frequencyandalow-frequencytransducer,eachofwhichisoptimally adjustedtoitsspecificrangesimilartoatwo-wayloudspeakersystem.Thetwo systemsareconnectedbymeansofacrossovernetwork. TheAKGD-222EBschematicallyshowninFig.1.28employstwocoaxiallymounteddynamictransducers.Oneisdesignedforhighfrequenciesandis placedclosesttothefrontgrilleandfacingforward.Theotherisdesignedfor lowfrequenciesandisplacedbehindthefirstandfacingrearward.Thelowfrequencytransducerincorporatesahum-buckingwindingtocanceltheeffects ofstraymagneticfields.Bothtransducersarecoupledtoa500Hzinductive-capacitive-resistivecrossovernetworkthatiselectroacousticallyphasecorrected and factory preset for linear off-axis response. (This is essentially the same designtechniqueusedinamoderntwo-wayloud-speakersystem.) Thetwo-waymicrophonehasapredominantlyfrequency-independentdirectionalpattern,producingmorelinearfrequencyresponseatthesidesofthemicrophoneandfarmoreconstantdiscriminationattherearofthemicrophone.Proximity effectatworkingdistancesdownto6inches(15cm)isreducedbecausethedistance betweenthemicrophonewindscreenandthelow-frequencytransducerislarge. The D-222EB incorporates a three-position bass-roll-off switch that provides 6dB or 12dB attenuation at 50Hz. This feature is especially useful in speechapplicationsandinacousticallyunfavorableenvironmentswithexcessivelow-frequencyambientnoise,reverberation,orfeedback.
1.3 TYPES OF TRANSDUCERS 1.3.1 Carbon Microphones Oneoftheearliesttypesofmicrophones,thecarbon microphone,isstillfound inoldtelephonehandsets.Ithasverylimitedfrequencyresponse,isverynoisy, has high distortion, and requires a hefty dc power supply. A carbon microphone4isshowninFig.1.29andoperatesinthefollowingmanner.
24
PART | I
Electroacoustic Devices
Replacement transducer/crossover module Red (in phase)
Red Front YEL facing HF transducer
212dB
1
2
2
Rear facing LF 1 transducer
Factory preset for most linear response at 908 off axis
Hum-bucking coil Black
Bassrolloff switch 26dB 3
2
1 White
FIGURE 1.28 Schematic of anAKG D-222EB two-way cardioid microphone. CourtesyAKG Acoustics,Inc.
Rubber mount Metal diaphragm
Insulation Carbon granules
Brass button
Carbon disk
Brass cup
Rubber mount
FIGURE 1.29
Rheostat
Step-up transformer
Output
Battery
Connectionandconstructionofasingle-buttoncarbonmicrophone.
Severalhundredsmallcarbongranulesareheldinclosecontactinabrass cupcalledabuttonthatisattachedtothecenterofametallicdiaphragm.Sound wavesstrikingthesurfaceofthediaphragmdisturbthecarbongranules,changingthecontactresistancebetweentheirsurfaces.Abatteryordcpowersource isconnectedinserieswiththecarbonbuttonandtheprimaryofanaudioimpedance-matchingtransformer.Thechangeincontactresistancecausesthecurrent fromthepowersupplytovaryinamplitude,resultinginacurrentwaveform similartotheacousticwaveformstrikingthediaphragm. Theimpedanceofthecarbonbuttonislowsoastep-uptransformerisused toincreasetheimpedanceandvoltageoutputofthemicrophoneandtoeliminatedcfromtheoutputcircuit.
1.3.2 Crystal and Ceramic Microphones Crystalandceramicmicrophoneswereoncepopularbecausetheywereinexpensive and their high-impedance high-level output allowed them to be connecteddirectlytotheinputgridofatubeamplifier.Theyweremostpopularin
Section | 1
25
Microphones
usewithhometaperecorderswheremicrophonecableswereshortandinput impedanceshigh. Crystal and ceramic microphones5 operate as follows: piezoelectricity is “pressure electricity” and is a property of certain crystals such as Rochelle salt, tourmaline, barium titanate, and quartz. When pressure is applied to thesecrystals,electricityisgenerated.Present-daycommercialmaterialssuch as ammonium dihydrogen phosphate (ADP), lithium sulfate (LN), dipotassiumtartrate(DKT),potassiumdihydrogenphosphate(KDP),leadzirconate, andleadtitanate(PZT)havebeendevelopedfortheirpiezoelectricqualities. Ceramicsdonothavepiezoelectriccharacteristicsintheiroriginalstate,but thecharacteristicsareintroducedinthematerialsbyapolarizingprocess.In piezoelectricceramicmaterialsthedirectionoftheelectricalandmechanical axesdependsonthedirectionoftheoriginaldcpolarizingpotential.During polarization a ceramic element experiences a permanent increase in dimensionsbetweenthepolingelectrodesandapermanentdecreaseindimension paralleltotheelectrodes. The crystal element can be cut as a bender element that is only affected byabendingmotionorasatwisterelementthatisonlyaffectedbyatwisting motion,Fig.1.30.
A. Crystal twister bimorph.
B. Ceramic bender bimorph.
C. Crystal bender bimorph. Ceramic strip
Silver electrodes Through holes covered with graphite D. Multimorph. FIGURE 1.30 Curvatures of bimorphs and multimorph. Courtesy Clevite Corp., Piezoelectric Division.
26
PART | I
Electroacoustic Devices
The internal capacitance of a crystal microphone is about 0.03µF for the diaphragm-actuatedtypeand0.0005–0.015µFforthesound-celltype. The ceramic microphone operates like a crystal microphone except it employsabariumtitanateslabintheformofaceramic,givingitbettertemperatureandhumiditycharacteristics. Crystalandceramicmicrophonesnormallyhaveafrequencyresponsefrom 80to6500Hzbutcanbemadetohaveaflatresponseto16kHz.Theiroutput impedance is about 100kΩ, and they require a minimum load of 1–5MΩ to producealevelofabout−30dBre1V/Pa.
1.3.3 Dynamic Microphones The dynamic microphoneisalsoreferredtoasapressureormoving-coilmicrophone.Itemploysasmalldiaphragmandavoicecoil,movinginapermanent magnetic field. Sound waves striking the surface of the diaphragm cause the coil to move in the magnetic field, generating a voltage proportional to the soundpressureatthesurfaceofthediaphragm. In a dynamic pressure unit, Fig.1.31, the magnet and its associated parts (magneticreturn,polepiece,andpoleplate)produceaconcentratedmagnetic fluxofapproximately10,000Ginthesmallgap. Thediaphragm,akeyitemintheperformanceofamicrophone,supportsthe voicecoilcentrallyinthemagneticgap,withonly0.006inchclearance. An omnidirectional diaphragm and voice-coil assembly is shown in Fig.1.32.Thecompliancesectionhastwohingepointswiththesectionbetween themmadeupoftangentialcorrugatedtriangularsectionsthatstiffenthisportionandallowthediaphragmtomoveinandoutwithaslightrotatingmotion. Thehingepointsaredesignedtopermithigh-complianceaction.Aspacersupportsthemovingpartofthediaphragmawayfromthetoppoleplatetoprovide roomforitsmovement.Thecementingflatisbondedtothefaceplate.Astiff
Diaphragm
Pole piece Voice coil
Magnet
FIGURE 1.31
Asimplifieddrawingofadynamicmicrophone.CourtesyShureIncorporated.
Section | 1
27
Microphones
Hinge point Spacer
Dome Coil seat
Cementing Tangential compliance flat section A. Sectional view
Voice coil
7/8"
B. Top view FIGURE 1.32
Omnidirectionaldiaphragmandvoicecoilassembly.
hemispherical dome is designed to provide adequate acoustical capacitance. The coil seat is a small step where the voice coil is mounted, centered, and bondedonthediaphragm. Early microphones had aluminum diaphragms that were less than 1mil (0.001in)thick.Aluminumislight-weight,easytoform,maintainsitsdimensional stability, and is unaffected by extremes in temperature or humidity. Unfortunately,beingonly1milthickmakesthediaphragmsfragile.Whenitis touchedorotherwisedeformedbyexcessivepressure,analuminumdiaphragm isdead. Mylar™,apolyesterfilmmanufacturedbytheDuPontCompany,iscommonlyusedfordiaphragms.Mylar™isauniqueplastic.Extremelytough,ithas hightensilestrength,highresistancetowear,andoutstandingflexlife.Mylar™ diaphragmshavebeencycletestedwithtemperaturevariationsfrom−40°Fto +170°F(−40°Cto+77°C)overlongperiodswithoutanyimpairmenttothediaphragm.SinceMylar™isextremelystable,itspropertiesdonotchangewithin thetemperatureandhumidityrangeinwhichmicrophonesareused. ThespecificgravityofMylar™isapproximately1.3ascomparedto2.7for aluminumsoaMylar™diaphragmmaybemadeconsiderablythickerwithout upsettingtherelationshipofthediaphragmmasstothevoice-coilmass.
28
PART | I
Electroacoustic Devices
Mylar™diaphragmsareformedunderhightemperatureandhighpressure, aprocessinwhichthemolecularstructureisformedpermanentlytoestablish a dimensional memorythatishighlyretentive.Unlikealuminum,Mylar™diaphragmswillretaintheirshapeanddimensionalstabilityalthoughtheymaybe subjectedtodrasticmomentarydeformations. Thevoicecoilweighsmorethanthediaphragmsoitisthecontrollingpart ofthemassinthediaphragmvoice-coilassembly.Thevoicecoilanddiaphragm mass(analogoustoinductanceinanelectricalcircuit)andcompliance(analogoustocapacitance),maketheassemblyresonateatagivenfrequencyasany tunedelectricalcircuit.Thefree-coneresonanceofatypicalundampedunitis intheregionof350Hz. Ifthevoicecoilwereleftundamped,theresponseoftheassemblywould peakat350Hz,Fig.1.33.Theresonantcharacteristicisdampedoutbytheuse ofanacousticresistor,afeltringthatcoverstheopeningsinthecenteringring behindthediaphragm.Thisisanalogoustoelectricalresistanceinatunedcircuit. While this reduces the peak at 350Hz, it does not fix the droop below 200Hz.Additionalacousticalresonantdevicesareusedinsidethemicrophone casetocorrectthedrooping.Acavitybehindtheunit(analogoustocapacitance) helps resonate at the low frequencies with the mass (inductance) of the diaphragmandvoice-coilassembly. Another tuned resonant circuit is added to extend the response down to 35Hz.Thiscircuit,tunedtoabout50Hz,isoftenatubethatcouplestheinside cavityofthemicrophonehousingtotheoutside,Fig.1.34. Thecurvatureofthediaphragmdomeprovidesstiffness,andtheaircavity betweenitandthedomeofthepolepieceformanacousticcapacitance.This capacitanceresonateswiththemass(inductance)oftheassemblytoextendthe responseupto20kHz. Tocontrolthehigh-frequencyresonance,anonmagneticfilterisplacedin frontofthediaphragm,creatinganacousticresistance,Fig.1.34.Thefilteris alsoaneffectivemechanicalprotectiondevice.Thefilterpreventsdirtparticles, magneticchips,andmoisturefromgravitatingtotheinsideoftheunit.Magneticchips,ifallowedtoreachthemagneticgaparea,eventuallywillaccumulateontopofthediaphragmandimpairthefrequencyresponse.Itispossible
Undamped
110 0
Damped
210
40 50
100
200
500
1k
2k
5k
Frequency–Hz FIGURE 1.33
Diaphragmandvoice-coilassemblyresponsecurve.
10 k
20 k
Section | 1
29
Microphones
Resonator Sintered bronze filter Magnet Transformer
Case cavity
Tube
Connector
FIGURE 1.34
Omnidirectionalmicrophonecross-sectionview.
forsuchchipstopinthediaphragmtothepolepiece,renderingthemicrophone inoperative. Figure1.35illustratestheeffectofavaryingsoundpressureonamoving-coil microphone.Forthissimplifiedexplanation,assumethatamasslessdiaphragm voice-coilassemblyisused.Theacousticwaveform,Fig.1.35A,isonecycleof anacousticwaveform,whereaindicatesatmosphericpressureAT;andbrepresentsatmosphericpressureplusaslightoverpressureincrement∆orAT + ∆. The electrical waveform output from the moving-coil microphone, Fig.1.35B, does not follow the phase of the acoustic waveform because at maximumpressure,AT + ∆orb,thediaphragmisatrest(novelocity).Further, thediaphragmanditsattachedcoilreachmaximumvelocity,hencemaximum electricalamplitude—atpointcontheacousticwaveform.Thisisofnoconsequenceunlessanothermicrophoneisbeingusedalongwiththemoving-coil microphonewheretheothermicrophonedoesnotseethesame90°displacement. Due to this phase displacement, condenser microphones should not be mixedwithmoving-coilorribbonmicrophoneswhenmicingthesamesource atthesamedistance.(Soundpressurecanbeproportionaltovelocityinmany practicalcases.)6 A steady overpressure which can be considered an acoustic square wave, Fig.1.35C, would result in the output shown in Fig.1.35D.As the acoustic pressurerisesfroma to b,ithasvelocity,Fig.1.35,creatingavoltageoutput fromthemicrophone.Oncethediaphragmreachesitsmaximumdisplacement
30
PART | I
Electroacoustic Devices
b
c a
e
A. Acoustical sine waveform.
d
a e B. Electrical sine waveform.
d
b
c b
c
a
d
g
e
f
C. Acoustical square waveform.
c-d-e a-b
FIGURE 1.35
f-g
D. Electrical square waveform.
Effectofavaryingsoundpressureonamoving-coilmicrophone.
at b,andstaysthereduringthetimeintervalrepresentedbythedistancebetween bandc,voice-coilvelocityiszerosoelectricaloutputvoltageceasesandthe outputreturnstozero.Thesamesituationrepeatsitselffromc to eandfrome to fontheacousticwaveform.Ascanbeseen,amoving-coilmicrophonecannot reproduceasquarewave. Another interesting theoretical consideration of the moving-coil microphonemechanismisshowninFig.1.36.Assumeasuddentransientcondition. Startingataontheacousticwaveform,thenormalatmosphericpressureissuddenlyincreasedbythefirstwavefrontofanewsignalandproceedstothefirst
Section | 1
31
Microphones
b a9
c a
e
A. Acoustical Waveform.
d e
a9 a
d
b
B. Electrical waveform.
c FIGURE 1.36
Effectofatransientconditiononamoving-coilmicrophone.
overpressurepeak,AT + ∆orb.Thediaphragmwillreachamaximumvelocity halfwaytobandthenreturntozerovelocityatb.Thiswillresultinapeak,a′, intheelectricalwaveform.Frombon,theacousticwaveformandtheelectrical waveformwillproceedasbefore,cycleforcycle,but90°apart. Inthisspecialcase,peaka′doesnotfollowtheinputpreciselysoitissomething extra. It will probably be swamped out by other problems (especially mass)encounteredinapracticalmoving-coilmicrophone.Itdoesillustratethat evenwitha“perfect,”massless,moving-coilmicrophone,“perfect”electrical waveformswillnotbeproduced. When sound waves vibrate the diaphragm, the voice coil has a voltage inducedinitproportionaltothemagnitudeandatthefrequencyofthevibrations.Thevoicecoilanddiaphragmhavesomefinitemassandanymasshas inertiathatcausesittowanttostayintheconditionitisin—namely,inmotion oratrest.Ifthestationarypartofthediaphragm-magnetstructureismovedin space,theinertiaofthediaphragmandcoilcausesthemtotrytoremainfixed inspace.Therefore,therewillberelativemotionbetweenthetwopartswitha resultant electrical output.An electrical output can be obtained in two ways, bymotionofthediaphragmfromairborneacousticalenergyorbymotionof themagnetcircuitbystructure-bornevibration.Thediaphragmmotionisthe desiredoutput,whilethestructure-bornevibrationisundesired.
32
PART | I
Electroacoustic Devices
Severalthingsmaybetriedtoeliminatetheundesiredoutput.Themassof thediaphragmandvoicecoilmaybereduced,buttherearepracticallimits,or thefrequencyresponsemaybelimitedmechanicallywithstifferdiaphragmsor electronicallywithfiltercircuits.However,limitedresponsemakesthemicrophoneunsuitableforbroad-rangeapplications.
Relative response–dB
1.3.3.1 Unidirectional Microphones Torejectunwantedacousticalnoisesuchassignalsemanatingfromthesidesor rearofthemicrophone,unidirectionalmicrophonesareused,Fig.1.7.Unidirectionalmicrophonesaremuchmoresensitivetovibrationrelativetotheiracousticsensitivitythanomnidirectionaltypes.Figure1.37showsaplotofvibration sensitivity versus frequency for a typical omnidirectional and unidirectional microphonewiththelevelsnormalizedwithrespecttoacousticalsensitivity. The vibration sensitivity of the unidirectional microphone is about 15dB higherthantheomnidirectionalandhasapeakatabout150Hz.Thepeakgives acluetohelpexplainthedifference. Unidirectional microphones are usually differential microphones; that is, the diaphragm responds to a pressure differential between its front and back surfaces.Theoncomingsoundwaveisnotonlyallowedtoreachthefrontofa diaphragmbut,throughoneormoreopeningsandappropriateacousticalphaseshiftnetworks,reachestherearofthediaphragm.Atlowfrequencies,thenet instantaneous pressure differential causing the diaphragm to move is small compared to the absolute sound pressure, Fig.1.38. CurveA is the pressure wavethatarrivesatthefrontofthediaphragm.CurveBisthepressurewave thatreachestherearofthediaphragmafteraslightdelayduetothegreaterdistancethesoundhastotraveltoreachtherearentryandsomeadditionalphase shiftitencountersafterentering.Thenetpressureactuatingthediaphragmis curveC,whichistheinstantaneousdifferencebetweenthetwouppercurves. Inatypicalunidirectionalmicrophone,thedifferentialpressureat100Hzwill beaboutone-tenthoftheabsolutepressureor20dBdownfromthepressurean omnidirectionalmicrophonewouldexperience.
0 dB � 0.001 V for l g 30 20 10 0 �10 20
FIGURE 1.37
Unidirectional
Omnidirectional
50 100
1000 Frequency–Hz
Vibrationsensitivityofmicrophonecartridge.
10k 20k
33
Microphones
Relative pressure amplitude
Section | 1
10”
A
B
C FIGURE 1.38
Differentialpressureatlowfrequenciesonunidirectionalmicrophones.
Toobtaingoodlow-frequencyresponse,areasonablelow-frequencyelectricaloutputisrequiredfromaunidirectionalmicrophone.Toaccomplishthis, the diaphragm must move more easily for a given sound pressure. Some of thisisaccomplishedbyreducingthedampingresistancetolessthanone-tenth used in an omnidirectional microphone.This reduction in damping increases themotionofthemechanicalresonantfrequencyofthediaphragmandvoice coil, around 150Hz in Fig.1.37, making the microphone much more acceptabletostructure-bornevibrations.Sincethediaphragmofanomnidirectional microphone is much more heavily damped, it will respond less to inertial or mechanicalvibrationforces. Toeliminateunwantedexternallow-frequencynoisefromeffectingaunidirectionalmicrophone,somekindofisolationsuchasamicrophoneshockmount isrequiredtopreventthemicrophonecartridgefromexperiencingmechanical shockandvibration.
1.3.4 Capacitor Microphones Inacapacitororcondensermicrophonethesoundpressurelevelvariesthehead capacitanceofthemicrophonebydeflectingoneortwoplatesofthecapacitor, causing an electrical signal that varies with the acoustical signal. The varyingcapacitancecanbeusedtomodulateanRFsignalthatislaterdemodulated orcanbeusedasonelegofavoltagedivider,Fig.1.39,whereRandCform thevoltagedividerofthepowersupply++ to−. The head of most capacitor microphones consists of a small two-plate 40–50pFcapacitor.Oneofthetwoplatesisastretcheddiaphragm;theotheris aheavybackplateorcenterterminal,Fig.1.39.Thebackplateisinsulatedfrom thediaphragmandspacedapproximately1mil(0.001in)from,andparallelto, therearsurfaceofthediaphragm.Mathematicallytheoutputfromtheheadmay becalculatedas
34
PART | I
Electroacoustic Devices
1
C
R 11 2 FIGURE 1.39
Voltagedividertypeofcapacitormicrophone.
EO =
Ep a2 P 8dt
(1.3)
where, EOistheoutputinvolts, Episthedcpolarizingvoltageinvolts, aistheradiusofactiveareaofthediaphragmincentimeters, Pisthepressureindynespersquarecentimeter, disthespacingbetweenthebackplateanddiaphragmincentimeters, tisthediaphragmtensionindynespercentimeter. Many capacitor microphones operate with an equivalent noise level of 15–30dBSPL.Althougha20–30dBSPLisintherangeofawell-constructed studio,a20–30dBmicrophoneequivalentnoiseisnotmaskedbyroomnoise becauseroomnoiseoccursprimarilyatlowfrequenciesandmicrophonenoise athigh-frequencyashiss. Inthepastthequalityofsoundrecordingswaslimitedbythecharacteristicsoftheanalogtapeandrecordmaterial,apartfromlossesinducedbythe copyingandpressingprocedures.Tapesaturation,forinstance,createdadditional harmonic and disharmonic distortion components, which affected the recordingfidelityathighlevels,whereasthelinearityatlowandmediumlevels was quite acceptable. The onset of these distortions was rather soft and extendedtoawidelevelrangethatmakesitdifficulttodeterminethethreshold ofaudibility. Thedistortioncharacteristicsofthestandardstudiocondensermicrophones isadequateforoperationwithanalogrecordingequipment.Althoughexhibiting ahighdegreeoftechnicalsophistication,thesemicrophonesshowindividual variations in the resolution of complex tonal structures, due to their specific frequencyresponsesanddirectivitypatternsandnonlineareffectsinherentto thesemicrophones. Thesepropertiesweremostlyconcealedbythedistortionssuperimposedby the analog recording and playback processing. But the situation has changed essentially since the introduction of digital audio. The conversion of analog
Section | 1
35
Microphones
signals into digital information and vice versa is carried out very precisely, especiallyathighsignallevels.Duetothelinearquantizationprocesstheinherentdistortionsofdigitalrecordingsvirtuallydecreaseatincreasingrecording levels,whichturnsformerdistortionbehaviorupsidedown.Thisnewreality, whichisintotalcontrasttoformerexperiencewithanalogrecordingtechnique, contributesmostlytothefactthatthespecificdistortioncharacteristicsofthe microphonemaybecomeobvious,whereastheyhavebeenmaskedpreviously bythemoresignificantdistortionsofanalogrecordingtechnique. Anotherfeatureofdigitalaudioistheenlargeddynamicrangeandreduced noisefloor.Unfortunately,duetothisimprovement,theinherentnoiseofthe microphonesmaybecomeaudible,becauseitisnolongercoveredupbythe noiseoftherecordingmedium. The capacitor microphone has a much faster rise time than the dynamic microphonebecauseofthesignificantlylowermassofthemovingparts(diaphragm versus diaphragm/coil assembly). The capacitor rise time rises from 10%ofitsrisetimeto90%inapproximately15µs,whiletherisetimeforthe dynamicmicrophoneisintheorderof40µs. Capacitor microphones generate an output electrical waveform in step or phasewiththeacousticalwaveformandcanbeadaptedtomeasureessentially dcoverpressures,Fig.1.40. Someadvantagesofcapacitormicrophonesare: ● ● ●
● ●
Small,low-massrigiddiaphragmsthatreducevibrationpickup. Smooth,extended-rangefrequencyresponse. Rugged—capable of measuring very high sound pressure levels (rocket launches). Lownoise(whichispartiallycancelledbytheneedforelectronics). Smallheadsize,whichprovideslowdiffractioninterference. B
B
C
C E
A
D
A
D
E
F
Acoustical waveform B
B
C
C E
A
A
D
D E
F
Electrical waveform FIGURE 1.40 Capacitor microphone acoustic wave and electrical signals. Note the in-phase condition.
36
PART | I
Electroacoustic Devices
1.3.4.1 Voltage Divider Capacitor Microphone Voltage divider-type capacitor microphonesrequireapreamplifierasanintegralpartofthehousingandasourceofpolarizingvoltageforthehead,anda sourceofpower. A high-quality capacitor microphone, the Sennheiser K6 Modular CondenserMicrophoneSeriesissuitableforrecordingstudios,televisionandradio broadcast, motion picture studios, and stage and concert hall applications, as wellashigh-qualitycommercialsoundinstallations. TheK6/ME62seriesisacapacitor microphone system,Fig.1.41,thatusesAF circuitrywithfield-effecttransistorssoithasalownoiselevel(15dBperDINIEC 651),highreliability,andlifelongstability.Lowcurrentconsumptionatlowvoltageandphantomcircuitpoweringpermitfeedingthemicrophonesupplyvoltage viaastandardtwo-conductorshieldedaudiocableoraninternalAAbattery. TheK6offersinterchangeablecapsules,allowingtheselectionofdifferent response characteristics from omnidirectional to cardioid to shotgun to adapt themicrophonetovarioustypesofenvironmentsandrecordingapplications. BecauseofthenewPCMrecorders,signal-to-noiseratio(SNR)hasreached alevelof90 dB,requiringcapacitormicrophonestoincreasetheirSNRlevel to match the recorder. The shotgun K6 series microphone, Fig.1.42, has an equivalentnoiselevelof16 dB(DINIEC651).
FIGURE 1.41 Modularmicrophonesystemwithanomnidirectionalcartridgeutilizingavoltage dividercircuit.CourtesySennheiserElectronicCorporation.
Section | 1
Microphones
37
FIGURE 1.42 ThesamemicrophoneshowninFig.1.41butwithashotguncartridge.Courtesy SennheiserElectronicCorporation.
As in most circuitry, the input stage of a voltage divider-type capacitor microphonehasthemosteffectonnoise,Fig.1.43.Itisimportantthatthevoltageonthetransducerdoesnotchange.Thisisnormallyaccomplishedbycontrollingtheinputcurrent.InthecircuitofFig.1.43,thevoltagesVin, Vo, and VD arewithin0.1%ofeachother.Noise,whichmightcomeintothecircuitasVin throughtheoperationalamplifier,isonly1⁄91ofthevoltageVo. Preattenuation,thatis,attenuationbetweenthecapacitorandtheamplifier, canbeachievedbyconnectingparallelcapacitorstotheinput,byreducingthe inputstagegainbymeansofcapacitorsinthenegativefeedbackcircuit,orby reducing the polarizing voltage to one-third its normal value and by using a resistivevoltagedividerintheaudioline.Figure1.44istheschematicforthe AKGC-460Bmicrophone.
1.3.4.2 Phantom Power for Capacitor Microphones Acommonwaytosupplypowerforcapacitormicrophonesiswithaphantom circuit. Phantom or simplex powering is supplying power to the microphone fromtheinputofthefollowingdevicesuchasapreamplifier,mixer,orconsole.
38
PART | I
2
E1 D G
Vin
Vnoise S
VD
R
V9
1
Electroacoustic Devices
E2 Vout
FIGURE 1.43 SimplifiedschematicofanAKGC-460Bmicrophoneinputcircuit.CourtesyAKG Acoustics.
FIGURE 1.44
SchematicofanAKGC-460Bmicrophone.CourtesyAKGAcoustics.
Mostcapacitormicrophonepreamplifierscanoperateonanyvoltagebetween 9Vdc and 52Vdc because they incorporate an internal voltage regulator.The preamplifier supplies the proper polarizing voltage for the capacitor capsule plusimpedancematchesthecapsuletothebalancedlow-impedanceoutput. Standard low-impedance, balanced microphone input receptacles are easily modifiedtosimplexbothoperatingvoltageandaudiooutputsignal,offeringthe followingadvantagesinreducedcostandeaseofcapacitormicrophoneoperation: ●
●
●
●
●
Special external power supplies and separate multiconductor cables formerlyrequiredwithcapacitormicrophonescanbeeliminated. The B+ supply in associated recorders, audio consoles, and commercial soundamplifierscanbeusedtopowerthemicrophonedirectly. Dynamic,ribbon,andcapacitormicrophonescanbeusedinterchangeably onstandard,low-impedance,balancedmicrophonecircuits. Dynamic, ribbon, and self-powered capacitor microphones may be connected to the modified amplifier input without defeating the microphone operatingvoltage. Anyrecording,broadcast,andcommercialinstallationcanbeinexpensively upgradedtocapacitormicrophoneoperationusingexisting,two-conductor microphonecablesandelectronics.
Section | 1
39
Microphones
Phantomcircuituserequiresonlythatthemicrophoneoperatingvoltagebe appliedequallytopins2and3oftheamplifierlow-impedance(normallyan XLR input) receptacle. Pin 1 remains ground and circuit voltage minus.The polarity of standard microphone cable wiring is not important except for the usual audio polarity requirement (see Section 1.5.3). Two equally effective methodsofamplifierpoweringcanbeused: 1. ConnectanamplifierB+supplyof9–12Vdirectlytotheungroundedcenter tap of the microphone input transformer, as shown in Fig.1.45.A seriesdroppingresistorisrequiredforvoltagesbetween12and52V.Figure1.46 isatypicalresistorvaluechart.Achartcanbemadeforanymicrophoneif thecurrentisknownforaparticularvoltage. 2. A two-resistor, artificial center powering circuit is required when the microphoneinputtransformerisnotcenter-tapped,orwheninputattenuationnetworks areusedacrosstheinputtransformerprimary.ConnectaB+supplyof9–12V directlytotheartificialcenteroftwo332Ω,1%toleranceprecisionresistors, asshowninFig.1.47.Anytransformercentertapshouldnotbegrounded.For voltagesbetween12and52V,doublethechartresistorvalueofFig.1.46.
Cable
Microphone 2
2
3 1
3 1
1
1
Amplifier input transformer
Rv
B1
2
FIGURE 1.45 Directcenter-tapconnectionmethodofphantompoweringcapacitormicrophones. CourtesyAKGAcoustics. RV (kV) 14
RV max
12 10 8 6 4 RV min
2 0
0
10
20
30
40
50
FIGURE 1.46 DroppingresistorvaluechartforphantompoweringAKGC-451Emicrophones. CourtesyAKGAcoustics.
40
PART | I
Electroacoustic Devices
Cable
Microphone 2
2
3
3
1
1
Amplifier input transformer 2Rv 2Rv
B1
FIGURE 1.47 Artificial center tap connected method of powering capacitor microphones. CourtesyAKGAcoustics.
Anynumberofcapacitormicrophonesmaybepoweredbyeithermethod fromasingleB+sourceaccordingtothecurrentavailable.Usethelargestresistorvalueshown(Rv max)forvariousvoltagesinFig.1.46forminimumcurrent consumption.
1.3.4.3 Capacitor Radio-Frequency, Frequency-Modulated Microphones A frequency-modulated microphone is a capacitor microphone that is connected to a radio-frequency (RF) oscillator. Pressure waves striking the diaphragmcausevariationsinthecapacityofthemicrophoneheadthatfrequency modulatestheoscillator.Theoutputofthemodulatedoscillatorispassedtoa discriminatorandamplifiedintheusualmanner. Capacitor microphones using an RF oscillator are not entirely new to the recordingprofession,butsincetheadventofsolid-statedevices,considerable improvement has been achieved in design and characteristics. An interestingmicrophoneofthisdesignistheSchoepsModelCMT26Umanufactured inWest Germany by Schall-Technik, and named after Dr. Carl Schoeps, the designer. ThebasiccircuitryisshowninFig.1.48.Bymeansofasingletransistor, twooscillatorycircuitsareexcitedandtunedtotheexactsamefrequencyof 3.7MHz. The output voltage from the circuits is rectified by a phase-bridge detector circuit, which operates over a large linear modulation range with verysmallRFvoltagesfromtheoscillator.Theamplitudeandpolarityofthe output voltage from the bridge depend on the phase angle between the two high-frequency voltages. The microphone capsule (head) acts as a variable capacitanceinoneoftheoscillatorcircuits.Whenasoundwaveimpingeson the surface of the diaphragm of the microphone head, the vibrations of the diaphragmaredetectedbythephasecurveoftheoscillatorcircuit,andanaudio frequencyvoltageisdevelopedattheoutputofthebridgecircuit.Themicrophone-headdiaphragmismetaltoguaranteealargeconstantcapacitance.An automaticfrequencycontrol(afc)withalargerangeofoperationisprovided
Section | 1
41
Microphones
29 Vdc
4
Oscillator Capsule
Phase demodulator
Capacitor diode FIGURE 1.48
AFC
Modulation output
19 Vdc
1
BasiccircuitfortheSchoepsradio-frequencycapacitormicrophone,seriesCMT.
bymeansofcapacitancediodestoprecludeanyinfluencecausedbyagingor temperaturechangesonthefrequency-determiningelements,thatmightthrow thecircuitryoutofbalance. Internaloutputresistanceisabout200Ω.Thesignal,feddirectlyfromthe bridge circuit through two capacitors, delivers an output level of −51dB to −49dB (depending on the polar pattern used) into a 200Ω load for a sound pressurelevelof10dynes/cm2.TheSNRandthedistortionareindependentof theloadbecauseofthebridgecircuit;therefore,themicrophonemaybeoperatedintoloadimpedancesrangingfrom30to200Ω. 1.3.4.3.1 Capacitor Radio-Frequency Microphones A capacitor microphone of somewhat different design, manufactured by Sennheiser and also employing a crystal-controlled oscillator, is shown in Fig.1.49.Intheconventionalcapacitormicrophone(withoutanoscillator)the inputimpedanceofthepreamplifierisintheorderof100MΩsoitisnecessary to place the capacitor head and preamplifier in close proximity. In the Sennheisermicrophone,thecapacitiveelement(head)usedwiththeRFcircuitryis amuchlowerimpedancesincetheeffectofasmallchangeincapacitanceat radiofrequenciesisconsiderablygreaterthanataudiofrequencies.Insteadof thecapacitorheadbeingsubjectedtoahighdcpolarizingpotential,thehead issubjectedtoanRFvoltageofonlyafewvolts.Anexternalpowersupplyof 12Vdcisrequired. Referring to Fig.1.49, the output voltage of the 8MHz oscillator is periodicallyswitchedbydiodesD1andD2tocapacitorC.Theswitchingphaseis shifted90°fromthatoftheoscillatorbymeansofloosecouplingandindividuallyaligningtheresonanceofthemicrophonecircuitMunderano-soundcondition.Asaresult,thevoltageacrosscapacitorCiszero.Whenasoundimpinges on the diaphragm, the switching phase changes proportionally to the sound
42
PART | I
Electroacoustic Devices
1
8 MHz Crystal-controlled oscillator
Q1
D1
Q2
D2 Feedback
+
2
M
FIGURE 1.49 BasiccircuitfortheSennheisermodel105,405,and805capacitormicrophones. CourtesySennheiserElectronicCorporation.
pressure, and a corresponding audio voltage appears across capacitor C.The outputoftheswitchingdiodesisdirectlyconnectedtothetransistoramplifier stage,whosegainislimitedto12dBbytheuseofnegativefeedback. AhighQoscillatorcircuitisusedtoeliminatetheeffectsofRFoscillator noiseasnoiseinanoscillatorycircuitisinverselyproportionaltotheQofthe circuit.BecauseofthehighQofthecrystalanditsstability,compensatingcircuitsarenotrequired,resultinginlowinternalnoise. The output stage is actually an impedance-matching transformer adjusted for100Ω,foraloadimpedanceof2000Ωorgreater.RFchokesareconnected intheoutputcircuittopreventRFinterferenceandalsotopreventexternalRF fieldsfrombeinginducedintothemicrophonecircuitry. 1.3.4.3.2 Symmetrical Push-Pull Transducer Microphone Investigationsonthelinearityofcondensermicrophonescustomarilyusedin therecordingstudioswerecarriedoutbySennheiserusingthedifference frequency methodincorporatingatwintonesignal,Fig.1.50.Thisisaveryreliable testmethodastheharmonicdistortionsofbothloudspeakersthatgeneratethe testsoundsseparatelydonotdisturbthetestresult.Thus,differencefrequency signalsarisingatthemicrophoneoutputarearisingfromnonlinearitiesofthe microphoneitself. Figure1.51showsthedistortioncharacteristicsofeightunidirectionalstudio condenser microphones which were stimulated by two sounds of 104dB SPL (3Pa).Thefrequencydifferencewasfixedto70Hzwhilethetwintonesignalwas sweptthroughtheupperaudiorange.Thecurvesshowthatunwanteddifference frequencysignalsofconsiderablelevelsweregeneratedbyallexaminedmicrophones.Althoughthecurvesareshapedratherindividually,thereisageneraltendencyforincreaseddistortionlevels(upto1%andmore)athighfrequencies. The measurement results can be extended to higher signal levels simply by linear extrapolation.This means, for instance, that 10 times higher sound
43
Microphones
f12f2 5 Const
Section | 1
f1 f2 f1f2
f1
f1f2
f2 FIGURE 1.50 Differencefrequencytest.CourtesySennheiserElectronicCorporation.
Distortion–%
3
23104 dB SPL Df 5 70 Hz
2
1
0 200
500
1k 2k 5k Frequency–Hz
10k
20 k
FIGURE 1.51 Frequency distortion of eight unidirectional microphones. Courtesy Sennheiser ElectronicCorporation.
pressureswillyield10timeshigherdistortions,aslongasclippingofthemicrophonecircuitisprevented.Thus,twosoundsof124dBSPLwillcausemore than 10% distortion in the microphones. Sound pressure levels of this order arebeyondthethresholdofpainofhumanhearingbutmayariseatclose-up micing.Despitethefactthattheaudibilityofdistortionsdependssignificantly onthetonalstructureofthesoundsignals,distortionfiguresofthisorderwill considerablyaffectthefidelityofthesoundpickup. The Cause of Nonlinearity. Figure 1.52showsasimplifiedsketchofacapacitivetransducer.Thediaphragmandbackplateformacapacitor,thecapacityof whichdependsonthewidthoftheairgap.Fromtheacousticalpointofview theairgapactsasacompleximpedance.Thisimpedanceisnotconstantbut dependsontheactualpositionofthediaphragm.Itsvalueisincreasedifthe diaphragm is moved toward the backplate and it is decreased at the opposite movement,sotheairgapimpedanceisvariedbythemotionofthediaphragm. Thisimpliesaparasiticrectifyingeffectsuperimposedtotheflowofvolume velocitythroughthetransducer,resultinginnonlinearity-createddistortion. Solving the Linearity Problem. Apush-pulldesignofthetransducerhelpsto improvethelinearityofcondensermicrophones,Fig.1.53.Anadditionalplate equaltothebackplateispositionedsymmetricallyinfrontofthediaphragm,
44
PART | I
Electroacoustic Devices
Diaphragm Air gap Backplate
FIGURE 1.52 Conventionalcapacitormicrophonetransducer.
Diaphragm
Backplate
Backplate
Symmetrical transducer FIGURE 1.53 Symmetricalcapacitormicrophonetransducer.
so two air gaps are formed with equal acoustical impedances as long as the diaphragm is in its rest position. If the diaphragm is deflected by the sound signal,thenbothairgapimpedancesaredeviatedoppositetoeachother.The impedanceofonesideincreaseswhiletheotherimpedancedecreases.Thevariationeffectscompensateeachotherregardlessofthedirectionofthediaphragm motion,andthetotalairgapimpedanceiskeptconstant,reducingthedistortion ofacapacitivetransducer. Figure1.54 shows the distortion characteristics of the Sennheiser MKH series push-pull element transformer less RF condenser microphones. The improvementonlinearityduetothepush-pulldesigncanbeseenbycomparing Fig.1.51toFig.1.54.
3
D stort on–%
2 3104 dB SPL Df 5 70 Hz 2
1
0 200
500
5k 2k 1k Frequency–Hz
10k
20k
FIGURE 1.54 Distortioncharacteristicsofthesymmetricalcapacitormicrophonetransducer.
Section | 1
Microphones
45
1.3.4.3.3 Noise Sources Theinherentnoiseofcondensermicrophonesiscausedpartlybytherandom incidenceoftheairparticlesatthediaphragmduetotheirthermalmovement. Thelawsofstatisticsimplythatsoundpressuresignalsatthediaphragmcan be evaluated by a precision that improves linearly with the diameter of the diaphragm. Thus, larger diaphragms yield better noise performance than smallerones. Anothercontributionofnoiseisthefrictionaleffectsintheresistivedampingelementsofthetransducer.Thenoisegenerationfromacousticalresistorsis basedonthesameprinciplesasthenoisecausedbyelectricalresistors,sohigh acousticaldampingimpliesmorenoisethanlowdamping. Noiseisalsoaddedbytheelectricalcircuitofthemicrophone.Thisnoise contributiondependsonthesensitivityofthetransducer.Hightransducersensitivityreducestheinfluenceofthecircuitnoise.Theinherentnoiseofthecircuit itself depends on the operation principle and on the technical quality of the electricaldevices. Noise Reduction. Large-diameter diaphragms improve noise performance. Unfortunately, a large diameter increases the directivity at high frequencies. A1inch(25mm)transducerdiameterisusuallyagoodchoice. A further method to improve the noise characteristics is the reduction of theresistivedampingofthetransducer.Inmostdirectionalcondensermicrophones,ahighamountofresistivedampingisusedinordertorealizeaflatfrequencyresponseofthetransduceritself.Withthisdesigntheelectricalcircuit ofthemicrophoneisrathersimple.However,itcreatesreducedsensitivityand increasednoise. Keeping the resistive damping of the transducer moderate will be a more appropriate method to improve noise performance; however, it leads to the transducerfrequencyresponsethatisnotflatsoequalizationhastobeapplied byelectricalmeanstoproduceaflatfrequencyresponseofthecompletemicrophone.Thisdesigntechniquerequiresamoresophisticatedelectricalcircuitbut producesgoodnoiseperformance. Theelectricaloutputofatransduceractsasapurecapacitance.Itsimpedancedecreasesasthefrequencyincreasessothetransducerimpedanceislowin anRFcircuitbuthighinanAFcircuit.Moreover,inanRFcircuittheelectrical impedanceofthetransducerdoesnotdependontheactualaudiofrequencybut isratherconstantduetothefixedfrequencyoftheRFoscillator.Contraryto this, in anAF design, the transducer impedance depends on the actual audio frequency, yielding very high values especially at low frequencies. Resistors of extremely high values are needed at the circuit input to prevent loading of the transducer output. These resistors are responsible for additional noise contribution. The RF circuit features a very low output impedance which is comparable to that of dynamic-type microphones. The output signal can be applied
46
PART | I
Electroacoustic Devices
directlytobipolartransistors,yieldinglownoiseperformancebyimpedance matching. TheSennheiserMKH20,Fig.1.55,isapressuremicrophonewithomnidirectional characteristics.The MKH 30 is a pure pressure-gradient microphonewithahighlysymmetricalbidirectionalpatternduetothesymmetry ofthepush-pulltransducer.TheMKH40,Fig.1.56,operatesasacombined pressureandpressure-gradientmicrophoneyieldingaunidirectionalcardioid pattern. ●
●
●
●
Themicrophonesarephantompoweredby48Vdcand2mA.Theoutputs aretransformerlessfloated,Fig.1.57. The SPLmax is 134dB at nominal sensitivity and 142dB at reduced sensitivity. TheequivalentSPLofthemicrophonesrangefrom10–12dBAcorrespondingtoCCIR-weightedfiguresof20–22dB. Thedirectionalmicrophonesincorporateaswitchablebassroll-offtocancel the proximity effect at close-up micing. The compensation is adjusted to about5cm(2in)distance.
Aspecialfeatureoftheomnidirectionalmicrophoneisaswitchablediffuse fieldcorrectionthatcorrectsforbothdirectanddiffusesoundfieldconditions.
FIGURE 1.55 Omnidirectionalpressurecapacitormicrophone.Notethelackofrearentryholes inthecase.CourtesySennheiserElectronicCorporation.
FIGURE 1.56 Unidirectionalpressure/pressure-gradientcapacitormicrophone.CourtesySennheiser ElectronicCorporation.
FIGURE 1.57
1 3 L3
C1
3
2
100pF
5mh
24V
24V
C2
L1
R5
330 V
R3 10nF
470 pF
L5
R7
C7 68 nF
1.5 kV
10 nF
T2 BC8508B
C R1 470 kV 4
28.5V
0.55V
0.55V R4 330 V
C1 B(2)
42V
A(1)
S1 42V 210dB
37.5V
D4
T1 BCB468
330 V
R2
D3
46.5V
1 22mF
C3 68V 1.25V
R1
26.5V
L2
Bat 83
D2
C5
Bat83
D1
1 nF
39 mh
L4
1 mF
1 mF
C11
8.2 kV
C1 150 nF
+
R16
R10
R14
R19
27 mF
+
C12
DS
R18
390 V R17 S2 39 kV R11 3 kW
C10 +
1 mh
22W
R9 470W 470 PF
39 kV 28.5 V
T4 BC860B
LL4148
6.8 kV T3 BCB508 33 V
C13
37.5V
D6 C14 68 mF
3.3 kV
R21
ZMM15
68 mF D8
3.3 kV C16 +
R20
ZMM68
150 mF D7
C15 +
39 mh
L7
1 nF
L4 39 mh
C13 1 nF
ZMM15
62 V
R23
C17
62 V
R22 41 V
41 V
C13 1 nF
C13 1 nF
SchematicofaSennheiserMKH20P48U3capacitormicrophone.CourtesySennheiserElectronicCorporation.
2
KS20-0
1
3
St XLR 3
1
2
Section | 1 Microphones
47
48
PART | I
Electroacoustic Devices
Thenormalswitchpositionisrecommendedforaneutralpickupwhencloseup micing and the diffuse field position is used if larger recording distances are usedwherereverberationsbecomesignificant. The distinction between both recording situations arises because omnidirectionalmicrophonestendtoattenuatelateralandreverseimpingingsound signalsathighfrequencies.Diffusesoundsignalswithrandomincidencecause alackoftrebleresponse,whichcanbecompensatedbytrebleemphasisatthe microphone. Unfortunately, frontally impinging sounds are emphasized also, butthiseffectisnegligibleifthereverberantsoundisdominant.
1.3.5 Electret Microphones An electret microphoneisacapacitormicrophoneinwhichtheheadcapacitor is permanently charged, eliminating the need for a high-voltage bias supply. From a design viewpoint a microphone intended to be used for critical recording, broadcast, or sound reinforcement represents a challenge involvingminimalperformancecompromise.Earlyelectretsofferedthemicrophone designer a means of reducing the complexity of a condenser microphone by eliminating the high-voltage bias supply, but serious environmental stability problems negated this advantage.Well-designed electret microphones can be storedat50°C(122°F)and95%relativehumidityforyearswithasensitivity lossofonly1dB.Undernormalconditionsoftemperatureandhumidity,electret transducers will demonstrate a much lower charge reduction versus time thanunderthesevereconditionsindicated.Evenifaproperelectretmaterialis used,therearemanystepsinthefabricating,cleaning,andchargingprocesses thatgreatlyinfluencechargestability.
FIGURE 1.58
ShureSM81electretcapacitormicrophone.CourtesyShureIncorporated.
Section | 1
49
Microphones
TheShureSM81cardioidcondensermicrophone3Fig.1.58,usesanelectret materialasameansofestablishingabiasvoltageonthetransducer.ThebackplatecarriestheelectretmaterialbaseduponthephysicalpropertiesofhalocarbonmaterialssuchasTeflon™andAclar,whichareexcellentelectrets,and materialssuchaspolypropyleneandpolyesterterephthalate(Mylar™),which aremoresuitablefordiaphragms. TheoperationoftheShureSM81microphoneisexplainedinSection1.2.3.1.3.
1.4 MICROPHONE SENSITIVITY7 Microphone sensitivityisthemeasureoftheelectricaloutputofamicrophone withrespecttotheacousticsoundpressurelevelinput. Sensitivityismeasuredinoneofthreemethods: Open-circuitvoltage Maximumpoweroutput ElectronicIndustries Association(EIA)sensitivity
0dB=1V/µbar 0dB=1mW/10µbar =1mW/Pa 0dB=EIAstandard SE-105
Thecommonsoundpressurelevelsusedformeasuringmicrophonesensitivityare: 94dBSPL10dyn/cm2SPL 10µbaror1Pa 74dBSPL1dyn/cm2SPL 1µbaror0.1Pa 0dBSPL0.0002dyn/cm2SPL 0 .0002Paor20µPa—thresholdofhearing 94dBSPLisrecommendedsince74dBSPListooclosetotypicalnoise levels.
1.4.1 Open-Circuit Voltage Sensitivity Thereareseveralgoodreasonsformeasuringtheopen-circuitvoltage: ●
●
●
Iftheopen-circuitvoltageandthemicrophoneimpedanceareknown,the microphoneperformancecanbecalculatedforanyconditionofloading. Itcorrespondstoaneffectiveconditionofuse.AmicrophoneshouldbeconnectedtoahighimpedancetoyieldmaximumSNR.A150–250Ωmicrophoneshouldbeconnectedto2kΩorgreater. When the microphone is connected to a high impedance compared to its own, variations in microphone impedance do not cause variations in response.
Theopen-circuitvoltagesensitivity(Sv)canbecalculatedbyexposingthe microphonetoaknownSPL,measuringthevoltageoutput,andusingthefollowingequation: Sv = 20 log E o − dB SPL + 94
(1.4)
50
PART | I
Test loudspeaker Random noise generator
Filter set Power amplifier
Electroacoustic Devices
Sound level meter
94 dB SPL Microphone to be tested Microvoltmeter
FIGURE 1.59 Reference7.)
Method of determining open-circuit voltage sensitivity of a microphone. (From
where, Svistheopen-circuitvoltagesensitivityindecibelsre1Vfora10dyn/cm2SPL (94dBSPL)acousticinputtothemicrophone, Eoistheoutputofthemicrophoneinvolts, dBSPListheleveloftheactualacousticinput. ThemicrophonesurementsystemcanbesetupasshowninFig.1.59.The setup requires a random-noise generator, a microvoltmeter, a high-pass and alow-passfilterset,apoweramplifier,atest-loudspeaker,andasoundlevel meter(SLM).TheSLMisplacedaspecificmeasuringdistance(about5–6ft or1.5–2m)infrontoftheloudspeaker.ThesystemisadjusteduntiltheSLM reads94dBSPL(abandofpinknoisefrom250to5000Hzisexcellentforthis purpose).ThemicrophonetobetestedisnowsubstitutedfortheSLM. ItisoftennecessarytoknowthevoltageoutputofthemicrophoneforvariousSPLstodeterminewhetherthemicrophonewilloverloadthepreamplifier circuitortheSNRwillbeinadequate.Todeterminethis,use E o = 10 e
Sv + dB SPL − 94 20
o
(1.5)
where, Eoisthevoltageoutputofmicrophone, Svistheopen-circuitvoltagesensitivity, dBSPListhesoundpressurelevelatthemicrophone.
1.4.2 Maximum Power Output Sensitivity7 The maximum power output sensitivity form of specification gives the maximumpoweroutputindecibelsavailablefromthemicrophoneforagivensound pressureandpowerreference.Suchaspecificationcanbecalculatedfromthe internalimpedanceandtheopen-circuitvoltageofthemicrophone.Thisspecificationalsoindicatestheabilityofamicrophonetoconvertsoundenergyinto electricalpower.Theequationis
Section | 1
51
Microphones
Sp = 10 log
Vo2 Ro
+ 44 dB’
(1.6)
where, Spisthepowerlevelmicrophonesensitivityindecibels, Voistheopen-circuitvoltageproducedbya1µbar(0.1Pa)soundpressure, Roistheinternalimpedanceofthemicrophone. Theformofthisspecificationissimilartothevoltagespecificationexcept thatapowerasopposedtoavoltagereferenceisgivenwiththesoundpressure reference.A1mWpowerreferenceanda10µbar(1Pa)pressurereferenceare commonlyused(asforthepreviouscase).Thisformofmicrophonespecificationisquitemeaningfulbecauseittakesintoaccountboththevoltageoutput andtheinternalimpedanceofthemicrophone. Spcanalsobecalculatedeasilyfromtheopencircuitvoltagesensitivity Sp = Sv − 10 log Z + 44 dB
(1.7)
where, Spisthedecibelratingforanacousticalinputof94dBSPL(10dyn/cm2)or1Pa, Z is the measured impedance of the microphone (the specifications of most manufacturersusetheratedimpedance). Theoutputlevelcanalsobedetermineddirectlyfromtheopen-circuitvoltage Sp = 10 log
E2o 0.001Z
− 6 dB
(1.8)
where, Eoistheopen-circuitvoltage, Zisthemicrophoneimpedance. Becausethequantity10log(E2/0.001Z)treatstheopen-circuitvoltageasif itappearsacrossaload,itisnecessarytosubtract6dB.(Thereadingis6dB higherthanitwouldhavebeenhadaloadbeenpresent.)
1.4.3 Electronic Industries Association (EIA) Output Sensitivity The Electronic IndustriesAssociation (EIA) Standard SE-105,August 1949, defines the system rating (GM) as the ratio of the maximum electrical output fromthemicrophonetothesquareoftheundisturbedsoundfieldpressureina planeprogressivewaveatthemicrophoneindecibelsrelativeto1mW/0.0002 dyn/cm2.Expressedmathematically, GM = 20 log
Eo P
− 10 log Zo − 50 dB
(1.9)
52
PART | I
Electroacoustic Devices
where, Eoistheopen-circuitvoltageofthemicrophone, Pistheundisturbedsoundfieldpressureindyn/cm2, Zoisthemicrophone-ratedoutputimpedanceinohms. Forallpracticalpurposes,theoutputlevelofthemicrophonecanbeobtained byaddingthesoundpressurelevelrelativeto0.0002dyn/cm2 to GM. BecauseGM,SV,andSParecompatible,GMcanalsobecalculated GM = Sv − 10 log RMR − 50 dB
(1.10)
where, GMistheEIArating, RMRistheEIAcentervalueofthenominalimpedancerangeshownbelow. Ranges (ohms) 20–80 80–300 300–1250 1250–4500 4500–20,000 20,000–70,000
Values Used (ohms) 38 150 600 2400 9600 40,000
= = = = = =
TheEIAratingcanalsobedeterminedfromthechartinFig.1.60.
SPL 620
130 120
100
110
2110 Threshold of pain 2100
290
Dynes/square centimeter
Reference sound field
90
290 0.1
280
270
Peak rms sound pressure at 1 ft from man's mouth, conversational speech
260
Average rms sound
270 1.0
70 Ordinary conversation at 3 ft
0.1
0.01
Average medium office
50
Decibels
40
Theater with audience
240
20
Rustle of leaves in gengle breeze
230
0.001
0 Threshold of hearing 0.002 Sound pressure level
FIGURE 1.60
2,400
10,000 8,000 6,000
2,000
2160 2150
600
1,000 800 600 400
2140 200
2130
150
SV 5 20 LOG10 E/P dB E 5 Open circuit voltage P 5 Sound pressure in dB referred to 1 mBAR Open Circuit power response SP 5 SV 2 LOG R 1 44 dB R 5 Nominal impedance
100.0
40,000
4,000
2170
38
10
220
2180
2120 Open-circuit voltage response
10.0 Country house
9,600
230
250
30
250 240
260
60
280
100,000 80,000 60,000
20,000
2190
80
1.0
40,000 Gm
D 0.1% atmosphere Peak rms sound with lips at microphone; riveter at 35 ft.
100
10
Impedance RMR
SV Millivolts 0.01
Sensitivity response GM 5 SV 2 10 LOG10RMR 2 50 dB RMR 5 Center value of nominal impedance range
Microphonesensitivityconversionchart.
100 80 60 40 20 10 8 6 4 2 1
Nominal impedance in ohms
Section | 1
53
Microphones
1.4.4 Various Microphone Sensitivities Microphonesaresubjectedtosoundpressurelevelsanywherefrom40dBSPL whendistantmicingto150dBSPLwhenextremelyclosemicing(i.e., 1⁄4inch fromtherocksinger’smouthorinsideadrumorhorn). Varioustypesofmicrophoneshavedifferentsensitivities,whichisimportant to know if different types of microphones are intermixed since gain settings,SNR,andpreamplifieroverloadwillvary.Table1.1givesthesensitivities ofavarietyofdifferenttypesofmicrophones.
TABLE 1.1 Sensitivities of Various Types of Microphones Type of Microphone
Sp
Carbon-button
−60 to −50 dB
Sv
Crystal
−50 to −40 dB
Ceramic
−50 to −40 dB
Dynamic (moving coil)
−60 to −52 dB
−85 to −70 dB
Capacitor
−60 to −37 dB
−85 to −45 dB
Ribbon-velocity
−60 to −50 dB
−85 to −70 dB
Transistor
−60 to −40 dB
Sound power
−32 to −20 dB
Line level
−40 to 0 dB
−20 to 0 dB
Wireless
−60 to 0 dB
−85 to 0 dB
1.4.5 Microphone Thermal Noise Sinceamicrophonehasanimpedance,itgeneratesthermalnoise.Evenwithout anacousticsignal,themicrophonewillstillproduceaminuteoutputvoltage. Thethermalnoisevoltage,En,producedbytheelectricalresistanceofasound sourceisdependentonthefrequencybandwidthunderconsideration,themagnitudeoftheresistance,andthetemperatureexistingatthetimeofthemeasurement.Thisvoltageis E n = 4ktR (bw ) where, kistheBoltzmann’sconstant,1.38×10–23J/K, tistheabsolutetemperature,273°+roomtemperature,bothin°C, Ristheresistanceinohms, bwisthebandwidthinhertz.
(1.11)
54
PART | I
Electroacoustic Devices
TochangethistodBvuse EINdBv = 20 log
En 0.775
(1.12)
Thethermalnoiserelativeto1Vis−198dBfora1Hzbandwidthand1Ω impedance.Therefore, TN = - 198 dB + 10 log (bw) + 10 log Z 1V
(1.13)
where, TNisthethermalnoiserelativeto1V, bwisthebandwidthinhertz, Zisthemicrophoneimpedanceinohms. Thermal noise relative to 1V can be converted to equivalent input noise (EIN)by EINdBm = −198 dB + 10 log (bw) + 10 log Z − 6 − 20 log 0.775 V
(1.14)
SincetheEINisindBmanddBmisreferencedto600Ω,theimpedanceZ is600Ω.
1.5 MICROPHONE PRACTICES 1.5.1 Placement Microphonesareplacedinvariousrelationshipstothesoundsourcetoobtain varioussounds.Whateverpositiongivesthedesiredeffectthatiswantedisthe correctposition.Therearenoexactrulesthatmustbefollowed;however,certainrecommendationsshouldbefollowedtoassureagoodsound.
1.5.1.1 Microphone-to-Source Distance Microphonesarenormallyusedinthedirectfield.Underthiscondition,inverse squarelawattenuationprevails,meaningthateachtimethedistanceisdoubled, the microphone output is reduced 6dB. For instance, moving from a microphone-to-sourcedistanceof2.5to5cm(1to2in)hasthesameeffectasmoving from15to30cm(6to12in),1to2ft(30to60cm),or5to10ft(1.5to3m). Distancehasmanyeffectsonthesystem.Inareinforcementsystem,doublingthedistancereducesgainbeforefeedback6dB;inallsystems,itreduces theeffectofmicrophone-to-sourcevariations. Usingtheinverse-square-lawequationforattenuation, attenuationdB = 20 log
D1 D2
(1.15)
Section | 1
55
Microphones
itcanbeseen,atamicrophone-to-sourcedistanceof2.5cm(1in),movingthe microphoneonly1.25cm(1⁄2in)closerwillincreasethesignal6dBand1.25cm (1⁄2in)fartherawaywilldecreasethesignal3.5dBforatotalsignalvariationof 9.5dBforonly2.5cm(1in)oftotalmovement!Atasource-to-microphonedistanceof30cm(12in),amovementof2.5cm(1in)willcauseasignalvariationof only0.72dB.Bothconditionscanbeusedadvantageously;forinstance,closemicingisusefulinfeedback-proneareas,highnoiselevelareas(rockgroups),orwhere thetalentwantstousethesource-to-microphonevariationstocreateaneffect. Thefartherdistancesaremostusefulwherelecternsandtablemicrophones areusedorwherethetalkerwantsmovementwithoutlevelchange. The microphone-to-source distance also has an effect on the sound of a microphone,particularlyonewithacardioidpattern.Asthedistancedecreases, the proximity effect increases creating a bassy sound (see Section 1.2.3.1). Closinginonthemicrophonealsoincreasesbreathnoiseandpopnoise.
1.5.1.2 Distance from Large Surfaces Whenamicrophoneisplacednexttoalargesurfacesuchasthefloor,6dBof gaincanberealized,whichcanbeahelpwhenfarmicing. Asthemicrophoneismovedawayfromthelargesurfacebutstillinproximityofit,cancellationofsomespecificfrequencieswilloccur,creatinganotchof upto30dB,Fig.1.65.Thenotchiscreatedbythecancellationofafrequency that,afterreflectingoffthesurface,reachesthemicrophonediaphragm180°out ofpolaritywiththedirectsound. Thefrequencyofcancellation,fc,canbecalculatedfromtheequation fc =
0.5c Dr1 + Dr2 −Dd
(1.16)
where, cisthespeedofsound,1130feetpersecondor344meterspersecond, 0.5istheout-of-polarityfrequencyratio, Dr1isthereflectedpathfromthesourcetothesurfaceinfeetormeters, Dr2isthereflectedpathfromthesurfacetothemicrophoneinfeetormeters, Ddisthedirectpathfromthesourcetothemicrophoneinfeetormeters. Ifthemicrophoneis10ftfromthesourceandbothare5ftabovethefloor, thecanceledfrequencyis 1130 ´ 0.5 7.07 + 7.07 − 10 = 136.47 Hz
fc =
(1.17)
Ifthemicrophoneismovedto2ftabovethefloor,thecanceledfrequencyis 319.20Hz.Ifthemicrophoneis6inchesfromthefloor,thecanceledfrequency is1266.6Hz.Ifthemicrophoneis1inchfromthefloor,thecanceledfrequency is7239.7Hz.
56
PART | I
Electroacoustic Devices
1.5.1.3 Behind Objects Sound,likelight,doesnotgothroughsolidoracousticallyopaqueobjects.It does, however, go through objects of various density. The transmission loss orabilityofsoundtogothroughthistypeofmaterialisfrequencydependent; therefore,ifanobjectofthistypeisplacedbetweenthesoundsourceandthe microphone,thepickupwillbeattenuatedaccordingtothetransmissioncharacteristicsoftheobject. Low-frequencysoundbendsaroundobjectssmallerthantheirwavelength, whichaffectsthefrequencyresponseofthesignal.Thenormaleffectofplacing the microphone behind an object is an overall reduction of level, a lowfrequencyboost,andahigh-frequencyroll-off. 1.5.1.4 Above the Source When the microphone is placed above or to the side of a directional sound source (i.e., horn or trumpet), the high-end frequency response will roll off because high frequencies are more directional than low frequencies, so less high-frequencySPLwillreachthemicrophonethanlow-frequencySPL. 1.5.1.5 Direct versus Reverberant Field Micinginthereverberantfieldpicksupthecharacteristicoftheroombecause themicrophoneispickingupasmuchormoreoftheroom,asitisthedirect soundfromthesource.Whenmicinginthereverberantfield,onlytwomicrophonesarerequiredforstereosinceisolationoftheindividualsoundsources is impossible. When in the reverberant field, a directional microphone will losemuchofitsdirectivity.Therefore,itisoftenadvantageoustouseanomnidirectionalmicrophonethathassmootherfrequencyresponse.Tomicsources individually,youmustbeinthedirectfieldandusuallyveryclosetothesource toeliminatecross-feed.
1.5.2 Grounding Thegroundingofmicrophonesandtheirinterconnectingcablesisofextreme importancesinceanyhumornoisepickedupbythecableswillbeamplified along with the audio signal. Professional systems generally use the method showninFig.1.61.Herethesignalispassedthroughatwo-conductorshielded cabletothebalancedinputofapreamplifier.Thecableshieldisconnectedto pinnumber1,andtheaudiosignaliscarriedbythetwoconductorsandpins2 and3oftheXLR-typeconnector.Theactualphysicalgroundisconnectedat thepreamplifierchassisonlyandcarriedtothemicrophonecase.Innoinstance isasecondgroundeverconnectedtothefarendofthecable,becausethiswill causetheflowofgroundcurrentsbetweentwopointsofgrounding. In systems designed for semiprofessional and home use, the method in Fig.1.62isoftenused.Notethatonesideoftheaudiosignaliscarriedoverthe
Section | 1
57
Microphones
2-conductor shielded cable 3-Pin female 3-pin male microphone microphone connector connector Preamplifier
Microphone
Black 3 �
2 �
White/Red Shield
Typicallow-impedancemicrophonetopreamplifierwiring.
Microphone
Microphone
Preamplifier
Black Shield
FIGURE 1.62
3 �
1
1 FIGURE 1.61
2 �
1/4 phone plug
Typicalsemiprofessional,hi-fimicrophonetopreamplifierwiring.
cableshieldtoapin-typeconnector.Thebodiesofboththemaleandfemale connectoraregrounded:thefemaletotheamplifiercaseandthemaletothe cableshield.Themicrophoneendisconnectedinasimilarmanner;hereagain thephysicalgroundisconnectedonlyatthepreamplifierchassis.Humpicked up on the shield and not on the center conductor is added to the signal and amplifiedthroughthesystem.
1.5.3 Polarity Microphone polarity,orphaseasitisoftencalled,isimportantespeciallywhen multiplemicrophonesareused.Whentheyareinpolaritytheyaddtoeachother ratherthanhavecancelingeffects.Ifmultiplemicrophonesareusedandoneis outofpolarity,itwillcausecombfilters,reducingqualityandstereoenhancement.TheEIAstandardRS-221.A,October1979,states“Polarityofamicrophoneoramicrophonetransducerelementreferstoin-phaseorout-of-phase conditionofvoltagedevelopedatitsterminalswithrespecttothesoundpressureofasoundwavecausingthevoltage.” Note:Exact in-phase relationshipcanbetakentomeanthatthevoltageis coincidentwiththephaseofthesoundpressurewavecausingthevoltage.In practicalmicrophones,thisperfectrelationshipmaynotalwaysbeobtainable.
58
PART | I
Electroacoustic Devices
Thepositiveorin-phaseterminalisthatterminalthathasapositivepotential andaphaseanglelessthan90°withrespecttoapositivesoundpressureatthe frontofthediaphragm. Whenconnectedtoathree-pin XLR connectorasperEIAstandardRS-297, thepolarityshallbeasfollows: ● ● ●
Out-of-phase—terminal3(black). In-phase—terminal2(redoranycolorotherthanblack). Ground—terminal1(shield).
Figure1.63showstheproperpolarityforthree-pinandfive-pinXLRconnectorsandforthree-pinandfive-pinDINconnectors. Asimplemethodofdeterminingmicrophonepolarityisasfollows: If two microphones have the same frequency response and sensitivity and are placednexttoeachotherandconnectedtothesamemixer,theoutputwill doubleifbothareused.However,iftheyareoutofpolaritywitheachother,the totaloutputwillbedown40–50dBfromtheoutputofonlyonemicrophone. The microphones to be tested for proper polarity are placed alongside each otherandconnectedtotheirrespectivemixerinputs.Withasingleacoustic sourceintothemicrophones(pinknoiseisagoodsource),onemixervolume
3
3 4 2
2 1
5
1
3-contact XLR:
5-contact XLR:
1 - Ground 1 - Ground 2 - Audio (In-phase) 2 - Audio (In-phase) 1 3 - Audio (Return) 3 - Audio (Return) } Channel 4 - Audio (In phase) } Channel 2 5 - Audio (Return) A. XLR connectors. 2 3 3
1
4
1
3-contact DIN:
5 2 5-contact DIN:
1 - Ground 2 - Left (Return) 3 - Right (Return) 4 - Left (In-phase) 5 - Right (In-phase) B. DIN connectors.
1 - Audio (In phase) 2 - Ground 3 - Audio (Return)
FIGURE 1.63
Microphoneconnectorpolarity.
Section | 1
Microphones
59
control is adjusted for a normal output level as indicated on aVU meter. Notethevolumecontrolsettingandturnitoff.Makethesameadjustment forthesecondmicrophone,andnotethesettingofthisvolumecontrol.Now openbothcontrolstothesesettings.Ifthemicrophonesareoutofpolarity, thequalityofreproductionwillbedistorted,andtherewillbeadistinctdrop inlevel.Reversingtheelectricalconnectionstoonemicrophonewillbring themintopolarity,makingthequalityaboutthesameasonemicrophone operatingandtheoutputlevelhigher. If the microphones are of the bidirectional type, one may be turned 180° to bringitintopolarityandlatercorrectedelectrically.Ifthemicrophonesare ofthedirectionaltype,onlytheoutputorcableconnectionscanbereversed. Afterpolarizingabidirectionalmicrophone,therearshouldbemarkedwith awhitestripeforfuturereference.
1.5.4 Balanced or Unbalanced Microphonescanbeconnectedeitherbalancedorunbalanced.Allprofessional installationsuseabalancedsystemforthefollowingreasons: ● ● ●
Reducedpickupofhum. Reducedpickupofelectricalnoiseandtransients. Reducedpickupofelectricalsignalsfromadjacentwires.
These reductions are realized because the two signal conductors shown in Fig.1.64pickupthesamestraysignalwithequalintensityandpolarity,sothenoise isimpressedevenlyoneachendofthetransformerprimary,eliminatingapotential acrossthetransformerandcancelinganyinputnoise.Becausethebalancedwires areinashieldedcable,thesignaltoeachconductorisalsogreatlyreduced. Wheninstallingmicrophonesintoanunbalancedsystem,anynoisethatgets totheinnerunbalancedconductorisnotcanceledbythenoiseintheshield,so the noise is transmitted into the preamplifier. In fact, noise impressed on the microphoneendoftheshieldaddstothesignalbecauseoftheresistanceofthe shieldbetweenthenoiseandtheamplifier. Balancedlow-impedancemicrophonelinescanbeaslongas500ft(150m) butunbalancedmicrophonelinesshouldneverexceed15ft(4.5m).
1.5.5 Impedance Mostprofessionalmicrophonesarelowimpedance,200Ω,andaredesignedto workintoaloadof2000Ω.High-impedancemicrophonesare50,000Ωandare designedtoworkintoanimpedanceof1–10MΩ.Thelow-impedancemicrophonehasthefollowingadvantages: ●
Lesssusceptibletonoise.Anoisesourceofrelativelyhighimpedancecannot“drive”intoasourceofrelativelylowimpedance(i.e.,themicrophone cable).
60
PART | I
Microphone � � �
�
+ FIGURE 1.64
●
� �
Electroacoustic Devices
Preamplifier � �
+
Noisecancellationonbalanced,shieldedmicrophonecables.
Capableofbeingconnectedtolongmicrophonelineswithoutnoisepickup andhigh-frequencyloss.
Allmicrophonecablehasinductanceandcapacitance.Thecapacitanceis about40pF(40×10−12)/ft(30cm).Ifacableis100ftlong(30m),thecapacitancewouldbe(40×10−12)×100ftor4×10–9For0.004µF.Thisisequivalent toa3978.9Ωimpedanceat10,000Hzandisfoundwiththeequation Xc =
1 2π fC
(1.18)
Thishaslittleeffectonamicrophonewithanimpedanceof200Ωasitdoes notreducetheimpedanceappreciablyasdeterminedby Zr =
Xc Zm Xc + Zm
(1.19)
Foramicrophoneimpedanceof200Ω,thetotalimpedanceZT =190Ωor lessthan0.5dB. Ifthissamecablewereusedwithahigh-impedancemicrophoneof50,000Ω, 10,000Hzwouldbedownmorethan20dB. Makingtheloadimpedanceequaltothemicrophoneimpedancewillreduce themicrophonesensitivity6dB,whichreducestheoverallSNRby6dB.For thebestSNR,theinputimpedanceoflow-impedancemicrophonepreamplifiers isalways2000Ωorgreater. Iftheloadimpedanceisreducedtolessthanthemicrophoneimpedance, ortheloadimpedanceisnotresistive,themicrophonefrequencyresponseand outputvoltagewillbeaffected. Changingtheloadofahigh-impedanceorceramicmicrophonefrom10MΩ to100kΩreducestheoutputat100Hzby27dB.
Section | 1
61
Microphones
1.6 MISCELLANEOUS MICROPHONES 1.6.1 Pressure Zone Microphones (PZM) The pressure zone microphone,referredtoasaPZ MicrophoneorPZM,isa miniaturecondensermicrophonemountedface-downnexttoasound-reflecting plateorboundary.Themicrophonediaphragmisplacedinthepressurezone justabovetheboundarywheredirectandreflectedsoundscombineeffectively in-phaseovertheaudiblerange. In many recording and reinforcement applications, the sound engineer is forcedtoplacemicrophonesnearhardreflectivesurfacessuchaswhenrecordinganinstrumentsurroundedbyreflectivebaffles,reinforcingdramaoropera withthemicrophonesnearthestagefloor,orrecordingapianowiththemicrophoneclosetotheopenlid. In these situations, sound travels from the source to the microphone via two paths: directly from the source to the microphone, and reflected off the surface to the microphone. The delayed sound reflections combine with the directsoundatthemicrophone,resultinginphasecancellationsofvariousfrequencies,Fig.1.65.Thiscreatesaseriesofpeaksanddipsinthenetfrequency responsecalledthecomb-filter effect,affectingtherecordedtonequalityand givinganunnaturalsound. The PZM was developed to avoid the tonal coloration caused by microphoneplacementnearasurface.Themicrophonediaphragmisarrangedparallelwithandveryclosetothereflectingsurfaceandfacingit,sothatthedirect
Microphone
R so efle un ct d e
d
Direct Sound
Reflective surface or boundary
Leve –dB
A. Microphone receiving direct and delayed sound simultaneously.
Frequency (linear scale) B. Resulting frequency response. FIGURE 1.65
Effectsofcancellationcausedbynearreflections(combfilters).
62
PART | I
Electroacoustic Devices
andreflectedwavescombineatthediaphragmin-phaseovertheaudiblerange, Fig.1.66. Thisarrangementcanprovideseveralbenefits: ●
●
● ●
●
●
●
●
● ●
●
Wide,smoothfrequencyresponse(naturalreproduction)becauseofthelack ofphaseinterferencebetweendirectandreflectedsound. A6dBincreaseinsensitivitybecauseofthecoherentadditionofdirectand reflectedsound. HighSNRcreatedbythePZM’shighsensitivityandlowinternalnoise. A3dBreductioninpickupofthereverberantfieldcomparedtoaconventionalomnidirectionalmicrophone. Lackofoff-axiscolorationasaresultofthesoundentry’ssmallsizeand radialsymmetry. Good-sounding pickup of off-mic instruments due to the lack of off-axis coloration. Identical frequency response for random-incidence sound (ambience) and directsoundduetothelackofoff-axiscoloration. Consistenttonequalityregardlessofsound-sourcemovementormicrophoneto-sourcedistance. Excellentreach(clearpickupofquietdistantsounds). Hemisphericalpolarpattern,equalsensitivitytosoundscomingfromany directionabovethesurfaceplane. Inconspicuouslow-profilemounting.
Di
re
ct
Reflected Close up view PZM
Level–dB
A. PZM receiving direct and reflected sound simultaneously.
Frequency B. Resulting frequency response. FIGURE 1.66
Effectsofreceivingdirectandreflectedsoundsimultaneously.
Section | 1
Microphones
FIGURE 1.67
PZM30Dpressurezonemicrophone.CourtesyCrownInternational,Inc.
63
The Crown PZM-30 series microphones, Fig.1.67, one of the original PZMs, are designed for professional use and built to take the normal abuse associated with professional applications. Miniaturized electronics built into themicrophonecantileverallowthePZM-30seriestobepowereddirectlyby simplexphantompowering.
1.6.1.1 Phase Coherent Cardioid (PCC) The phase coherent cardioid microphone(PCC)isasurface-mountedsupercardioidmicrophonewithmanyofthesamebenefitsasthePZM.UnlikethePZM, however,thePCCusesasubminiaturesupercardioidmicrophonecapsule. Technically,thePCCisnotapressurezonemicrophone.Thediaphragmof aPZMisparalleltotheboundary;thediaphragmofthePCCisperpendicular totheboundary.UnlikeaPZM,thePCCaimsalongtheplaneonwhichitis mounted.Inotherwords,themainpickupaxisisparallelwiththeplane. The Crown PCC-200 microphone, a Phase Coherent Cardioid surfacemountedboundarymicrophone,Fig.1.68,isintendedforuseonstagefloors, lecterns, and conference tables wherever gain-before-feedback and articulation are important. Figure 1.69 shows the horizontal polar response for this microphone. ThePCC-160canbedirectlyphantompowered.Abass-tiltswitchisprovidedfortailoringlowendresponse. 1.6.1.2 Directivity The PZM picks up sounds arriving from any direction above the surface it is mounted on (hemispherical). It is often necessary to discriminate against sounds arriving from certain directions.To make the microphone directional orhemicardioid(rejectsoundsfromtherear)thecapsulecanbemountedwith thecantileverinacornerboundarymadeof1⁄4inch(6mm)thickPlexiglas.The larger the boundary, the better it discriminates against low-frequency sounds fromtherear.
64
PART | I
Electroacoustic Devices
Decibels
FIGURE 1.68 Crown PCC-200 Phase Coherent Cardioid microphone. Courtesy Crown International,Inc.
15 10 5 0 �5 �10 �15 �20 �25 �30 �35 20
Boost 0� Cut
130� 50 100 200 500 1k 2k Frequency–Hz
5k 10k 20k
FIGURE 1.69 The horizontal plane polar response of the PCC-160 Phase Coherent Cardioid microphonewiththesource30°abovetheinfiniteboundary.CourtesyCrownInternational,Inc.
Forbestresults,acornerboundary12in×24inwide(0.3m×0.6m)isrecommendedandisnearlyinvisibletotheaudience,Fig.1.70. Aboom-mountedorsuspendedPZMcanbetapedtothecenterofa1⁄4inch (6mm)thick2ft×2ft(0.6m×0.6m)or4ft×4ft(1.2m×1.2m)panel.The microphoneshouldbeplaced4inches(10cm)off-centerforasmootherfrequencyresponse.Usingclearacrylicplastic(Plexiglas)makesthepanelnearly invisiblefromadistance.IftheedgesofthePlexiglaspickuplight,theycanbe tapedorpaintedblack. 1.6.1.2.1 Sensitivity Effects IfaPZMcapsuleisplacedverynearasinglelargeboundary(within0.020in or0.50mm),suchasalargeplate,floor,orwall,incomingsoundreflectsoff thesurface.Thereflectedsoundwaveaddstotheincomingsoundwaveinthe
Section | 1
Microphones
FIGURE 1.70
Cornerboundaryusedtocontroldirectivityofpressurezonemicrophones.
65
Pressure Zone next to the boundary. This coherent addition of sound waves doublesthesoundpressureatthemicrophone,effectivelyincreasingthemicrophonesensitivityoroutputby6dBoverastandardmicrophone. IfthePZMcapsuleisplacedatthejunctionoftwoboundariesatrightangles toeachother,suchasthefloorandawall,thewallincreasessensitivity6dB, andthefloorincreasessensitivityanother6dB.Addingtwoboundariesatright anglesincreasessensitivity12dB. WiththePZMelementatthejunctionofthreeboundariesatrightangles, suchasinthecornerofthefloorandtwowalls,microphonesensitivitywillbe 18dBhigherthanwhatitwasinopenspace. Notethattheacousticsensitivityofthemicrophonerisesasboundariesare added,buttheelectronicnoiseofthemicrophonestaysconstant,sotheeffectiveSNRofthemicrophoneimproves6dBeverytimeaboundaryisaddedat rightanglestopreviousboundaries. 1.6.1.2.2 Direct-to-Reverberant Ratio Effects Directsoundsensitivityincreases6dBperboundaryadded,whilereverberant or random-incidence sound increases only 3dB per boundary added. Consequently,thedirect-to-reverberantratioincreases3dB(6dBdir−3dBrev)wheneveraboundaryisaddedatrightanglestopreviousboundaries. 1.6.1.2.3 Frequency-Response Effects The low-frequency response of the PZM or PCC depends on the size of the surface it is mounted on. The larger the surface, the more the low-frequency responseisextended.Thelow-frequencyresponseshelvesdowntoalevel6dB belowthemid-frequencylevelatthefrequencywherethewavelengthisabout 6timestheboundarydimension.Forexample,thefrequencyresponseofaPZM ona2ft×2ft(0.6m×0.6m)panelshelvesdown6dBbelow94Hz.Ona5in× 5in(12cm×12cm)plate,theresponseshelvesdown6dBbelow376Hz.
66
PART | I
Electroacoustic Devices
For best bass and flattest frequency response, the PZM or PCC must be placedonalargehardboundarysuchasafloor,wall,table,orbaffleatleast 4ft×4ft(1.2m×1.2m). Toreducebassresponse,thePZMorPCCcanbemountedonasmallplate wellawayfromotherreflectingsurfaces.Thisplatecanbemadeofthinplywood,Masonite,clearplastic,oranyotherhard,smoothmaterial.Whenused onacarpetedfloorthePZMorPCCshouldbeplacedonahard-surfacedpanel atleast1ft×1ft(0.3m×0.3m)forflattesthigh-frequencyresponse. Todeterminethefrequencyf−6dBwheretheresponseshelvesdown6dB,use f− 6 dB =
188* D
(1.20)
*57.3forSIunits where, Distheboundarydimensioninfeetormeters. Forexample,iftheboundaryis2ft(0.6m)square,the6dBdownpointis 188 D 188 = 2 = 94Hz
f− 6 dB =
Below94Hz,theresponseisaconstant6dBbelowtheuppermid-frequency level.Notethatthereisaresponseshelf,notacontinuousroll-off. WhenthePZMisonarectangularboundary,twoshelvesappear.Thelongside oftheboundaryisDmaxandtheshortsideDmin.Theresponseisdown3dBat f− 3 dB =
188* Dmax
(1.21)
f− 3 dB =
188* Dmin
(1.22)
*57.3forSIunits andisdownanother3dBat
*57.3forSIunits Thelow-frequencyshelfvarieswiththeangleofthesoundsourcearound theboundary.At90°incidence(soundwavemotionparalleltotheboundary), thereisnolow-frequencyshelf. Thedepthoftheshelfalsovarieswiththedistanceofthesoundsourceto thepanel.Theshelfstartstodisappearwhenthesourceiscloserthanapanel dimensionaway.IfthesourceisveryclosetothePZMmountedonapanel, thereisnolow-frequencyshelf;thefrequencyresponseisflat.
Section | 1
67
Microphones
If the PZM is at the junction of two or more boundaries at right angles toeachother,theresponseshelvesdown6dBperboundaryattheabovefrequency.Forexample,atwo-boundaryunitmadeof2ft(0.6m)squarepanels shelvesdown12dBbelow94Hz. Thereareotherfrequency-responseeffectsinadditiontothelow-frequency shelf.Forsoundsourceson-axistotheboundary,theresponserisesabout10dB above the shelf at the frequency where the wavelength equals the boundary dimension. Forasquarepanel, Fpeak =
0.88c D
(1.23)
where, cisthespeedofsound(1130ft/sor344m/s)Distheboundarydimensionin feetormeters. Foracircularpanel Fpeak =
c D
(1.24)
As an example, a 2ft (0.6m) square panel has a 10dB rise above the shelf at 0.88c D 0.88 ´ 1130 = 2 = 497 Hz
Fpeak =
Notethatthisresponsepeakisonlyforthedirectsoundofanon-axissource. Theeffectismuchlessifthesoundfieldatthepanelispartlyreverberant,or ifthesoundwavesstrikethepanelatanangle.Thepeakisalsoreducedifthe microphonecapsuleisplacedoff-centerontheboundary. Figure1.71showsthefrequencyresponseatvariousanglesofsoundincidenceofaPZMmountedona2ftsquarepanel.Notetheseveralphenomena showninthefigure: ● ● ● ●
● ●
Thelow-frequencyshelf(mostvisibleat30°and60°). Thelackoflow-frequencyshelvingat90°(grazingincidence). The10dBriseinresponseat497Hz. Theedge-interferencepeaksanddipsabove497Hz(mostvisibleat0°or normalincidence). Thelesseningofinterferenceatincreasingangles. Thegreaterrearrejectionofhighfrequenciesthanlowfrequencies.
68
PART | I
Electroacoustic Devices
Sound source
90°
Decibels
180� A. Layout. 15 10 5 0 �5 �10 �15 �20 �25 �30 �35 20
0� 30� 60� 120� 150� 180�
50 100 200 500 1k 2k Frequency–Hz B.Response.
5k 10k 20k
FIGURE 1.71 Frequencyresponseofapressurezonemicrophone.Notethe6dBshelfat94Hz.
1.6.1.2.4 Frequency-Response Anomalies Caused by Boundaries Frequencyresponseisaffectedbythefollowing: ●
●
●
●
When sound waves strike a boundary, pressure doubling occurs at the boundary surface, but does not occur outside the boundary, so there is a pressure difference at the edge of the boundary. This pressure difference createssoundwaves. Thesesoundwavesgeneratedattheedgeoftheboundarytraveltothe microphoneinthecenteroftheboundary.Atlowfrequencies,theseedge wavesareoppositeinpolaritytotheincomingsoundwaves.Consequently, theedgewavescancelthepressure-doublingeffect. At low frequencies, pressure doubling does not occur, but at mid to high frequencies, pressure doubling does occur. The net effect is a mid- to high-frequencyboost,whichcouldbeconsideredalow-frequencylossorshelf. Incomingwaveshavingwavelengthsaboutsixtimestheboundarydimensions are canceled by edge effects while waves much smaller than the boundarydimensionsarenotcanceledbyedgeeffects. Waves having wavelengths on the order of the boundary dimensions are subjecttovaryinginterferenceversusfrequency,i.e.,peaksanddipsinthe frequencyresponse.
Section | 1
●
●
●
Microphones
69
Atthefrequencywherethewavelengthequalstheboundarydimension,the edge wave is in phase with the incoming wave. Consequently, there is a responserise(about10dBabovethelow-frequencyshelf)atthatfrequency. Above that frequency, there is a series of peaks and dips that decrease in amplitudewithfrequency. Theedge-waveinterferencedecreasesiftheincomingsoundwavesapproach theboundaryatanangle. Interferencealsoisreducedbyplacingthemicrophonecapsuleoff-center. Thisrandomizesthedistancesfromtheedgestothemicrophonecapsule, resultinginasmootherresponse.
1.6.2 Lavalier Microphones Lavalier microphonesaremadeeithertowearonalavalieraroundtheneckor toclipontoatie,shirt,orotherpieceofclothing.Theolderheavystylelavalier microphone, Fig.1.72, which actually laid on the chest, had a frequency responsethatwasshapedtoreducetheboomingnessofthechestcavity,and thelossofhigh-frequencyresponsecausedbybeing90°offaxistothesignal, Fig.1.73.Thesemicrophonesshouldnotbeusedforanythingexceptasalavaliermicrophone. Lavalier microphones may be dynamic, condenser (capacitor), pressurezone,electret,orhigh-impedanceceramic. Thenewerlossmassclip-onlavaliermicrophones,Figs.1.74and1.75,do notrequirefrequencyresponsecorrectionbecausethereisnocouplingtothe chestcavityandthesmalldiameterofthediaphragmdoesnotcreatepressure build-upathighfrequencies,creatingdirectionality. Mostlavaliermicrophonesareomnidirectional;however,moremanufacturersareproducingdirectionallavaliermicrophones.TheSennheiserMKE104 clip-onlavaliermicrophone,Fig.1.75,hasacardioidpickuppattern,Fig.1.76. Thisreducesfeedback,backgroundnoise,andcombfilteringcausedbythecancelingbetweenthedirectsoundwavesandsoundwavesthathitthemicrophone onareflectivepathfromthefloor,lectern,andsoforth.
FIGURE 1.72 Shure SM11 dynamic omnidirectional lavalier microphone. Courtesy Shure Incorporated.
70 Relative response–dB
PART | I
Electroacoustic Devices
�10 0 �10 �20
20
50 100
1K Frequency–Hz
10K 20K
FIGURE 1.73
Typicalfrequencyresponseofaheavy-styledynamiclavaliermicrophone.
FIGURE 1.74 Incorporated.
Shure SM183 omnidirectional condenser lavalier microphone. Courtesy Shure
FIGURE 1.75 SennheiserMKE104clip-onlavalierdirectionalmicrophone.CourtesySennheiser ElectronicCorporation.
OneofthesmallestmicrophonesistheCountrymanB6,Fig.1.77.TheB6 microphonehasadiameterof0.1inchesandhasreplaceableprotectivecaps. Because of its small size, it can be hidden even when it’s in plain sight. By choosingacolorcaptomatchtheenvironment,themicrophonecanbepushed throughabuttonholeorplacedinthehair. Lavaliermicrophonesarenormallyusedtogivethetalkerfreedomofmovement.This causes problems associated with motion—for instance, noise being transmitted through the microphone cable. To reduce this noise, soft, flexible microphonecablewithgoodfilltoreducewiremovementshouldbeused.The cable,orpowersupplyforelectret/condensermicrophones,shouldbeclippedto
71
Section | 1
Microphones
FIGURE 1.76 Corporation.
Polar response of the microphone in Fig.1.75. Courtesy Sennheiser Electronic
FIGURE 1.77 Countryman B6 miniature lavalier microphone. Courtesy of Countryman Associates,Inc.
theuser’sbeltorpantstoreducecablenoisetoonlythatcreatedbetweentheclip andthemicrophone,about2ft(0.6m).Clippingtothewaistalsohastheadvantageofactingasastrainreliefwhenthecordisaccidentallypulledorsteppedon. Asecondimportantcharacteristicofthemicrophonecableissize.Thecable shouldbeassmallaspossibletomakeitunobtrusiveandlightenoughsoitwill notpullonthemicrophoneandclothing.
72
PART | I
Electroacoustic Devices
Becausethemicrophoneisnormally10inches(25cm)fromthemouthofthe talkerandoutofthesignalpath,themicrophoneoutputislessthanamicrophone onastandinfrontofthetalker.Unlessthetorsoisbetweenthemicrophoneand loudspeaker,thelavaliermicrophoneisoftenaprimecandidateforfeedback. Forthisreason,themicrophoneresponseshouldbeassmoothaspossible. Asinanymicrophonesituation,thefartherthemicrophoneisawayfromthe source,themorefreedomofmovementbetweenmicrophoneandsourcewithoutadverseeffects.Ifthemicrophoneiswornclosetotheneckforincreased gain,theoutputlevelwillbegreatlyaffectedbytheraisingandloweringand turningofthetalker’shead.Itisimportantthatthemicrophonebewornchest highandfreefromclothing,etc.thatmightcoverthecapsule,reducinghighfrequencyresponse.
1.6.3 Head-Worn Microphones Head-worn microphones such as the Shure Model SM10A, Fig.1.78, and Shure Beta 53, Fig.1.79, are low-impedance, unidirectional, dynamic microphones,designedforsportsandnewsannouncing,forinterviewingandintercommunicationssystems,andforspecial-eventremotebroadcasting.TheShure SM10AisaunidirectionalmicrophonewhiletheBeta53isanomnidirectional microphone. Head-worn microphones offer convenient, hands-free operation without userfatigue.Asclose-talkingunits,theymaybeusedundernoisyconditions
FIGURE 1.78 Incorporated.
Shure SM10A dynamic unidirectional head-worn microphone. Courtesy Shure
Section | 1
73
Microphones
�20
dB
�10 0
�10 �20
20
50
100 200
500 1k Hz
2k
5k 10k
20k
Standard Cap Filtered Cap
FIGURE 1.79 ShureBeta53omnidirectionalmicrophoneanditsfrequencyresponse.Courtesy ShureIncorporated.
withoutlosingormaskingvoicesignals.Theyaresmall,lightweight,rugged, andreliableunitsthatnormallymounttoacushionedheadband.Apivotpermitsthemicrophoneboomtobemoved20°inanydirectionandthedistance betweenthemicrophoneandpivottobechanged9cm(31⁄2in). Another head-worn microphone, the Countryman Isomax E6 Directional EarSetmicrophone,isextremelysmall.Themicrophoneclipsaroundtheear ratherthanaroundthehead.Theunitscomeindifferentcolorstoblendinwith the background. The ultra-miniature condenser element is held close to the mouthbyathinboomandcomfortableearclip.Theentireassemblyweighs lessthanone-tenthofanounceandalmostdisappearsagainsttheskin,soperformerscanforgetit’sthereandaudiencesbarelyseeit,Fig.1.80. Themicrophonerequireschangeableendcapsthatcreateacardioidpickup patternforeaseofplacement,orahypercardioidpatternwhenmoreisolationis needed.TheC(cardioid)andH(hypercardioid)endcapsmodifytheEarSet’s directionality,Fig.1.81. TheEarSetseriesshouldalwayshaveaprotectivecapinplacetokeepsweat, makeup,andotherforeignmaterialoutofthemicrophone. Thehypercardioidcapprovidesthebestisolationfromalldirections,with anulltowardthefloorwherewedgemonitorsareoftenplaced.Thehypercardioidisslightlymoresensitivetoairmovementandhandlingnoiseandshould alwaysbeusedwithawindscreen.
74
PART | I
Electroacoustic Devices
FIGURE 1.80 Countryman Isomax E6 Directional EarSet microphone. Courtesy Countryman Associates,Inc. 0.25 in 1 in 24 in
Front
24 in
Null
A. Frequency response.
Cardioid cap
Hypercardioid cap
B. 1 kHz polar response. FIGURE 1.81 FrequencyresponseandpolarresponseoftheCountrymanE6EarSetmicrophone inFig.1.80.CourtesyCountrymanAssociates,Inc.
Thecardioidcapisslightlylessdirectional,withanullroughlytowardthe performer’s back. It’s useful for trade show presenters or others who have a monitorloudspeakerovertheirshouldersorbehindthem.
Section | 1
Microphones
75
The microphone can be connected to a sound board or wireless microphonetransmitterwiththestandard2mmcableoranextrasmall1mmcable, Fig.1.80.
1.6.4 Base Station Microphones Base station power microphones are designed specifically for citizens band transceivers,amateurradio,andtwo-wayradioapplications.Forclearertransmission and improved reliability, transistorized microphones can be used to replaceceramicordynamic,high-orlow-impedancemicrophonessuppliedas originalequipment. TheShureModel450SeriesII,Fig.1.82,isahighoutputdynamicmicrophonedesignedforpaginganddispatchingapplications.Themicrophonehas anomnidirectionalpickuppatternandafrequencyresponsetailoredforoptimumspeechintelligibility,Fig.1.83.Itincludesanoutputimpedanceselection switchforhigh,30,000Ω,andlow,225Ω,andalockingpress-to-talkswitch. The press-to-talk switch can be converted to a monitor/transmit switch withtheShureRK199SSplit-BarConversionKit.Whentheoptionalsplit-bar Transmit/MonitorSwitchConversionKitisinstalled,themonitorbarmustbe depressedbeforethetransmitswitchcanbedepressed,requiringtheoperator
FIGURE 1.82
Shure450SeriesIIbasestationmicrophone.CourtesyShureIncorporated.
76 Relative response—dB
PART | I
Electroacoustic Devices
�10 0 �10
20
50
100
200
500 1000
5000 10,000
Frequency—Hz
FIGURE 1.83 Frequency response of the Shure 450 Series II microphone shown in Fig.1.82. CourtesyShureIncorporated.
toverifythatthechannelisopenbeforetransmitting.Themonitorbarcanbe lockedintheonposition.Thetransmitbarismomentaryandcannotbelocked.
1.6.5 Differential Noise-Canceling Microphones Differential noise-canceling microphones, Fig.1.84, are essentially designed foruseinautomobiles,aircraft,boats,tanks,public-addresssystems,industrial plants,oranyservicewheretheambientnoiselevelis80dBorgreaterandthe microphoneishandheld.Discriminationisaffordedagainstallsoundsoriginating morethan1⁄4inch(6.4mm)fromthefrontofthemicrophone.Thenoise-cancelingcharacteristicisachievedthroughtheuseofabalancedportopening,which directs the unwanted sound to the rear of the dynamic unit diaphragm out of phasewiththesoundarrivingatthefrontofthemicrophone.Thenoisecanceling ismosteffectiveforfrequenciesabove2000Hz.Onlyspeechoriginatingwithin 1 ⁄4inch(6.4mm)oftheapertureisfullyreproduced.Theaveragediscrimination betweenspeechandnoiseis20dBwithafrequencyresponseof200–5000Hz.
FIGURE 1.84
Shure577Bdynamicnoise-cancelingmicrophone.CourtesyShureIncorporated.
Section | 1
Microphones
77
1.6.6 Controlled-Reluctance Microphones The controlled-reluctance microphoneoperatesontheprinciplethatanelectricalcurrentisinducedinacoil,locatedinachangingmagneticfield.Amagnetic armatureisattachedtoadiaphragmsuspendedinsideacoil.Thediaphragm, whendisturbedbyasoundwave,movesthearmatureandinducesacorrespondingvaryingvoltageinthecoil.Highoutputwithfairlygoodfrequencyresponse istypicalofthistypeofmicrophone.
1.6.7 Handheld Entertainer Microphones The handheld entertainer microphone is most often used by a performer on stage and, therefore, requires a special frequency response that will increase articulationandpresence.Themicrophonesareoftensubjectedtoroughhandling,extremeshock,andvibration.Forliveperformances,theproximityeffect canbeusefultoproducealowbasssound. Probably the most famous entertainer’s microphone is the Shure SM58, Fig.1.85.Themicrophonehasahighlyeffectivesphericalwindscreenthatalso reducesbreathpopnoise.Thecardioidpickuppatternhelpsreducefeedback. Thefrequencyresponse,Fig.1.86,istailoredforvocalswithbrightenedmidrangeandbassroll-off.Table1.2givesthesuggestedmicrophoneplacementfor besttonequality. Toovercomeroughhandlingandhandlingnoise,specialconstructiontechniquesareusedtoreducewind,popnoise,andmechanicalnoiseandtoensure thatthemicrophonewillwithstandsuddencollisionswiththefloor.TheSennheiser MD431, Fig.1.87, is an example of a high-quality, rugged, and lowmechanical-noisemicrophone.Toeliminatefeedback,theMD431incorporates asupercardioiddirectionalcharacteristic,reducingsidepickupto12%orless thanhalfthatofconventionalcardioids. Anotherproblem,particularlywithpowerfulsoundreinforcementsystems, ismechanical(handling)noise.Asidefromdisturbingtheaudience,itcanactuallydamageequipment.Ascanbeseeninthecutaway,theMD431isactuallya microphonewithinamicrophone.Thedynamictransducerelementismounted withinaninnercapsule,isolatedfromtheouterhousingbymeansofashock absorber.Thisprotectsitfromhandlingnoiseaswellasothermechanicalvibrationsnormallyencounteredinliveperformances. Toscreenoutnoisestillfurther,aninternalelectricalhighpassfilternetworkisincorporatedtoinsurethatlow-frequencydisturbanceswillnotaffect the audio signal.A built-in mesh filter in front of the diaphragm reduces the poppingandexcessivesibilanceoftenproducedbyclosemicing. The microphone case is a heavy-duty cast outer housing with a stainless steelfrontgrilleandreed-typeon-offswitch.Ahumbuckingcoilismounted behindthetransducertocanceloutanystraymagneticfields.
78
PART | I
ShureSM58vocalmicrophone.CourtesyShureIncorporated.
Relative response–dB
FIGURE 1.85
Electroacoustic Devices
�10 0 �10 20
60 100
1k Frequency–Hz
10 k 20 k
FIGURE 1.86 Frequency response of the Shure SM58 vocal microphone. Courtesy Shure Incorporated.
TABLE 1.2 Suggested Placement for the SM58 Microphone Application
Suggested Microphone Placement
Tone Quality
Lead and backup vocals
Lips less than 150 mm (6 in) away or touching the windscreen, on axis to microphone
Robust sound, emphasized bass, maximum isolation from other sources
Speech
from mouth, just above nose height
Natural sound, reduced bass
200 mm (8 in) to 0.6 m (2 ft) away from mouth, slightly off to one side
Natural sound, reduced bass, and minimal “s” sounds
1 m (3 ft) to 2 m (6 ft) away
Thinner; distant sound; ambience
Section | 1
79
Microphones
Three-layer stainless-steel gauze Dynamic transducer capsule
Hum compensation coil Metal springs
Outer switch housing Magnet Switch button
Switch locking device Rim for quick-release stand clip Connector pins
Suspension system Rigid outer stainless-steel
Rear sound inlet
Filter components
Magnetic shield
Handle
Shock absorber
FIGURE 1.87 Cutaway view of a Sennheiser MD431 handheld entertainment microphone. CourtesySennheiserElectronicCorporation.
1.6.8 Pressure-Gradient Condenser Microphones One of the most popular studio microphones is the Neumann U-87 multidirectionalcondensermicrophone,Fig.1.88,anditscousin,theNeumannU-89, Fig.1.89. This microphone is used for close micing where high SPLs are
FIGURE 1.88
NeumannU-87microphone.CourtesyNeumannUSA.
80
FIGURE 1.89
PART | I
Electroacoustic Devices
NeumannU-89microphone.CourtesyNeumannUSA.
commonlyencountered.Theresponsebelow30Hzisrolledofftopreventlowfrequencyblockingandcanbeswitchedto200Hztoallowcompensationfor thebassrisecommontoalldirectionalmicrophonesatcloserange. Thefigure8characteristicisproducedbymeansoftwocloselyspacedor assembledcardioidcharacteristiccapsules,whoseprincipalaxesarepointedin oppositedirectionsandareelectricallyconnectedinantiphase. Thesemicrophonesareusuallymadewithbackplatesequippedwithholes, slots,andchambersformingdelayelementswhoseperforationsactaspartfrictionresistancesandpartenergystorage(acousticinductancesandcapacitances), givingthebackplatethecharacterofanacousticlow-passnetwork.Inthecutoff rangeofthislow-passnetwork,abovethetransitionfrequencyft,themembrane isimpingedupononlyfromthefront,andthemicrophonecapsulechangestoa pressureorinterferencetransducer. Theoutputvoltagee(t)ofacondensermicrophoneusingdcpolarizationis proportionaltotheapplieddcvoltageEoand,forsmalldisplacementamplitudes
Section | 1
81
Microphones
of the diaphragm, to the relative variation in capacity c(t)/Co caused by the soundpressure c (t) Co
e ( t) = E o
(1.25)
where, Eoistheapplieddcvoltage, c(t)isthevariablecomponentofcapsulecapacity, Coisthecapsulecapacityintheabsenceofsoundpressure, tisthetime. The dependence of output voltage e(t) on Eo is utilized in some types of microphones to control the directional characteristic. Two capsules with cardioidcharacteristicsasshowninFig.1.90areplacedbacktoback.Theycan alsobeassembledasaunitwithacommonbackplate.Theaudio(ac)signals providedbythetwodiaphragmsareconnectedinparallelthroughacapacitor C. The intensity and phase relationship of the outputs from the two capsule halvescanbeaffectedbyvaryingthedcvoltageappliedtooneofthem(theleft cartridgeinFig.1.90).Thiscanbeaccomplishedthroughaswitch,orapotentiometer.Thedirectionalcharacteristicofthemicrophonemaybechangedby remotecontrolvialongextensioncables. IftheswitchisinitscenterpositionC,thentheleftcapsule-halfdoesnot contribute any voltage, and the microphone has the cardioid characteristic of therightcapsule-half.InswitchpositionA,thetwoacvoltagesareinparallel, resultinginanomnidirectionalpattern.InpositionEthetwohalvesareconnectedinantiphase,andtheresultisafigure8directionalresponsepattern. ThelettersA to EgivenfortheswitchpositionsinFig.1.90producethepatternsgiventhesamelettersinFig.1.91. C
D �60V
�60 V to �60 V 1G� 0
8
G 1G�
0V
1G�
S
A E D C B
�
� �60 V
0V
�60 V
FIGURE 1.90 CircuitoftheNeumannU-87condensermicrophonewithelectricallyswitchable directioncharacteristic.CourtesyNeumannUSA.
82
PART | I
Electroacoustic Devices
FIGURE 1.91 ByusingamicrophoneasshowninFig.1.89andsuperimposingtwocardioidpatterns (toprow),directionalresponsepatterns(bottomrow)canbeobtained.CourtesyNeumannUSA.
1.6.9 Interference Tube Microphones The interference tube microphone8 as described by Olson in 1938 is often called a shotgun microphone because of its physical shape and directional characteristics. Importantcharacteristicsofanymicrophoneareitssensitivityanddirectional qualities.Assumingaconstantsoundpressuresource,increasingthemicrophone tothesourcedistancerequiresanincreaseinthegainoftheamplifyingsystem toproducethesameoutputlevel.ThisisaccompaniedbyadecreaseinSNRand an increase in environmental noises including reverberation and background noisetowheretheindirectsoundmayequalthedirectsound.Thewantedsignal thendeterioratestowhereitisunusable.Distancelimitationscanbeovercome byincreasingthesensitivityofthemicrophone,andtheeffectofreverberation andnoisepickupcanbelessenedbyincreasingthedirectivityofthepattern.The interferencetubemicrophonehasthesetwodesirablequalities. The DPA 4017 is a supercardioid shotgun microphone. It is 8.3 inches (210mm)longandweighs2.5oz(71g),makingitusefulonbooms,Fig.1.92. ThepolarpatternisshowninFig.1.93. Thedifferencebetweeninterferencetubemicrophonesandstandardmicrophonesliesinthemethodofpickup. Aninterferencetubeismountedoverthediaphragmandisschematically drawninFig.1.94. Themicrophoneconsistsoffourpartsasshownintheschematic: 1. Interference tube with one frontal and several lateral inlets covered with fabricorotherdampingmaterial. 2. Capsulewiththediaphragmandcounterelectrode(s). 3. Rearinlet. 4. Electroniccircuit. Thedirectionalcharacteristicsarebasedontwodifferentprinciples: 1. Inthelow-frequencyrange,tubemicrophonesbehaveasfirst-orderdirectional receivers.The tube in front of the capsule can be considered as an acousticelementwithacomplianceduetotheenclosedairvolumeanda
83
Section | 1
Microphones
FIGURE 1.92 phones,Inc.
AsupercardioidinterferencetubeDPA4017microphone.CourtesyDPAMicro-
A. Directional characteristics.
B. Frequency response of switching filters.
C. On and off axis response measured at 60 cm (23.6 in). FIGURE 1.93 Directional characteristics and frequency response of a DPA 4017 microphone. CourtesyDPAMicrophones,Inc.
FIGURE 1.94
Schematicofaninterferencetubemicrophone.
84
PART | I
Electroacoustic Devices
resistancedeterminedbythelateralholesorslitsofthetube.Therearinlet isdesignedasanacousticlow-passfiltertoachievethephaseshiftforthe desiredpolarpattern(normallycardioidorsupercardioid). 2. Inthehigh-frequencyrange,theacousticalpropertiesoftheinterferencetube determinethepolarpatterns.Thetransitionfrequencybetweenthetwodifferent directionalcharacteristicsdependsonthelengthofthetubeandisgivenby fo =
c 2L
(1.26)
where, foisthetransitionfrequency, cisthevelocityofsoundinairinfeetpersecondormeterspersecond, Listhelengthoftheinterferencetubeinfeetormeters. ReferringtoFig.1.94,ifthetubeisexposedtoaplanarsoundwave,every lateralinletisthestartingpointofanewwavetravelinginsidethetubetoward thecapsuleaswellastowardthefrontalinlet.Apartfromthecaseoffrontal soundincidence,everyparticularwavecoversadifferentdistancetothecapsule and,therefore,arrivesatadifferenttime.Figure1.94showsthedelaytimesof wavesbandccomparedtowavea.Notethattheyincreasewiththeangleof soundincidence. The resulting pressure at the capsule can be calculated by the sum of all particular waves generated over the tube’s length, all with equal amplitudes but different phase shifts. The frequency and phase response curves can be describedby Pθ = P (θ = 0c)
πL ´ (1 − cos θ) F λ πL ´ (1 − cos θ) λ
sin
4.39 dB
6 dB
49,869 Time–ms A. An ETC made in a listening room with a GenRAd 1/2 inch microphone (LD 2 LR 5 0.24). 0
Dir/Rev Energy 5.54 dB Reverb(S) 0.16 s Alcons 2.26% RASTI 0.798 execellent SNR >–8.89 dB
6 dB
0
49,869 Time–ms B. The identical position measured with the ITE technique (LD 2 LR 5 0.24).
FIGURE 1.140 AnETCcomparisonofameasurementmicrophoneandtheITEtechniqueatthe samepositionintheroom.CourtesySyn-Aud-Con.
126
PART | I
Electroacoustic Devices
ITErecordingandPARplaybackallowagivenlistenertohearagivenspeech intelligibilityenvironmentasperceivedbyanotherperson’sheadandouterear configurationrightuptotheeardrum. RecordingsmadeusingITEmicrophonesintwodifferentpeople’searsof thesameperformanceinthesameseatingareasounddifferent.Playbackover loudspeakerswherethesystemisproperlybalancedforperfectgeometryfor onepersonmayrequireasmuchas10dBdifferentfront-to-backbalancefor anotherpersontohearperfectgeometryduringplayback. SinceITErecordingsaretotallycompatiblewithnormalstereophonicreproductionsystemsandcanprovidesuperiorfidelityinmanycases,thepractical useofITEmicrophonywouldappeartobeunlimited.
1.9 USB MICROPHONES Thecomputerhasbecomeanimportantpartofsoundsystems.Manyconsoles aredigitalandmicrophonesareconnecteddirectlytothem.Microphonesare alsoconnectedtocomputersthroughtheUSBinput. Theaudio-technicaAT2020USBcardioidcondensermicrophone,Fig.1.141, isdesignedforcomputerbasedrecording.ItincludesaUSB(UniversalSerial Bus) digital output that isWindows and Mac compatible.The sample rate is 44.1kHzwithabitdepthof16bits.Themicrophoneispowereddirectlyfrom the5VdcUSBoutput.
FIGURE 1.141 Audio-technicaAT2020USBmicrophone.CourtesyAudio-TechnicaU.S.,Inc.
Section | 1
Microphones
127
FIGURE 1.142 MXL.006USBmicrophone.CourtesyMarshallElectronics,Inc.
TheMXL.006USBisacardioidcondensermicrophonewithaUSBoutput thatconnectsdirectlytoacomputerwithouttheneedforexternalmicpreamps throughUSB1.1and2.0,Fig.1.142. TheanalogsectionoftheMXL.006microphonefeaturesa20Hz–20kHz frequencyresponse,alargegolddiaphragm,pressure-gradientcondensercapsule,andathree-position,switchableattenuationpadwithsettingsforHi(0dB), Medium(–5dB),andLo(–10dB).Thedigitalsectionfeaturesa16bitDelta SigmaA/Dconverterwithasamplingrateof44.1kHzand48kHz.Protecting theinstrument’scapsuleisaheavy-dutywiremeshgrillwithanintegratedpop filter. TheMXL.006includesaredLEDbehindtheprotectivegrilltoinform theuserthatthemicrophoneisactiveandcorrectlyoriented.TheMXL.006 ships with a travel case, a desktop microphone stand, a 10ft USB cable, windscreen,anapplicationsguide,andfreedownloadablerecordingsoftware forPCsandMac.
1.10 WIRELESS COMMUNICATION SYSTEMS Wireless communication systems are wireless microphones (radio microphones),Fig.1.143,andarelatedconcept,wirelessintercoms.Oftenthesame end user buys both the microphones and intercoms for use in television and radiobroadcastproduction,filmproduction,andrelatedentertainment-oriented applications. Wireless microphone systems can be used with any of the preceding microphones discussed. Some wireless microphone systems include a fixed microphone cartridge while others allow the use of cartridges by various manufacturers.
128
PART | I
Electroacoustic Devices
AblockdiagramofawirelessmicrophonesystemisshowninFig.1.144. Thesendingendofawirelessmicrophonesystemhasadynamic,condenser, electret,orpressurezonemicrophoneconnectedtoapreamplifier,compressor, andasmalltransmitter/modulatorandantenna. The receiving end of the system is an antenna, receiver/discriminator, expander,andpreamplifier,whichisconnectedtoexternalaudioequipment. Inastandardintercomsystem,eachpersonhasaheadsetandbeltpack(or equivalent),allinterconnectedbywires.Wirelessintercomsareessentiallyidenticalinoperation,onlytheyusenocablebetweenoperators.Instead,eachbelt packincludesaradiotransmitterandreceiver.Thewirelessintercomusertypicallywearsaheadset(aboommicrophonewithoneortwoearpieces)andcan simultaneouslytransmitononefrequencyandreceiveonanother.Thewireless intercom transmitter is virtually identical to a wireless microphone transmitter,butthereceiverisminiaturizedsothatit,too,canbeconvenientlycarried aroundandoperatedwithminimumbatterydrain. Wirelessmicrophonesarewidelyusedintelevisionproduction.Handheld models(integralmicrophonecapsuleandtransmitter)areusedbyperformers “oncamera,”wheretheynotonlyfreetheperformertowalkaroundandgesture spontaneously,theyalsoavoidtheneedforstagepersonneltofeedwiresaround
FIGURE 1.143 ShureUHF-Rwirelessmicrophonesystem.CourtesyShureIncorporated.
Antenna Microphone Preamp
Compression
A. Transmitter audio section.
Transient modulator B. Radio-frequency section.
Antenna Receive discriminator
Expander
Audio amp Output
C. Receiver audio section. FIGURE 1.144 Wirelessmicrophonetransmittersectionwithbuilt-inpreamplifier,compressor, andtransmitter,andthereceiverwithabuilt-indiscriminatorexpander.
Section | 1
Microphones
129
cameras,props,etc.Lavaliermodels(smallpocket-sizedtransmittersthatwork withlavalierorminiature“hidden”microphones)areusedingameshows,soap operas, dance routines, etc., where they eliminate the need for boom microphonesandfurtheravoidvisualclutter. For location film production, as well as electronic news gathering (ENG) andelectronicfieldproduction(EFP),wirelessmicrophonesmakeitpossibleto obtainusable“firsttake”soundtracksinsituationswherepreviously,postproductiondialogueloopingwasnecessary. In theatrical productions, wireless microphones free actors to speak and/ or sing at normal levels through a properly designed sound-reinforcement system. Inconcerts,handheldwirelessmicrophonespermitvocaliststomovearound withoutrestriction,andwithoutshockhazardevenintherain.Somelavalier modelshavehigh-impedancelineinputsthatacceptelectricguitarcordstocreatewirelessguitars. In all of the above applications where wireless microphones are used, in the studio or on location, a wireless intercom also is an invaluable communication aid between directors, stage managers, camera, lighting and sound crews,andsecuritypersonnel.Forcueingoftalentandcrews(ormonitoring intercomconversations),economicalreceive-onlyunitscanbeused.Insports production,wirelessintercomsareusedbycoaches,spotters,players,productioncrews,andreporters.Amajoradvantageiszerosetuptime.Incriticalstunt coordination,awirelessintercomcanmakethedifferencebetweenasafeevent ornoneatall.
1.10.1 Criteria for Selecting a Wireless Microphone There are a number of criteria that must be considered in obtaining a wirelessmicrophonesystemsuitableforprofessionaluse.19,20 Ideally,suchasystemmustworkperfectlyandreliablyinavarietyoftoughenvironmentswith goodintelligibilityandmustbeusablenearstrongRFfields,lightingdimmers, andsourcesofelectromagneticinterference.Thisrelatesdirectlytothetypeof modulation(standardfrequencymodulationornarrow-bandfrequencymodulation), the operating frequency, high-frequency (HF), very high-frequency (VHF),ultrahigh-frequency(UHF),thereceiverselectivity,andsoforth.The systemmustbeveryreliableandcapableofoperatingatleastfivehoursonone setofdisposablebatteries(orononerechargeifNi-Cadsareused).
1.10.1.1 Frequency Bands of Operation Based on the FCC’s reallocation of frequencies and the uncertainty of currentandfutureallocations,somewirelessmanufacturersareofferingsystems thatavoidtheVHFandUHFbandscompletely.TheISM(industrial,science, and medicine) bands provide a unique alternative to theTV bands. By international agreement, all devices are low powered so there will never be any
130
PART | I
Electroacoustic Devices
grosslyhigh-poweredRFinterferencepotential.The2.4GHzbandprovidesa viablealternativetotraditionalUHFbands,andaslongaslineofsightbetween transmittersandreceiversismonitoreduserscaneasilygeta100meterrange. Anotherbenefitof2.4GHzisthatitcansimplifywirelessinventoryfortravelingshows.Thesamewirelessfrequenciesareacceptedworldwide,sothereis noneedtoadheretothecountry-specificfrequencyrulesthatseverelycomplicate the situation for international tours.The same applies within the United States—thesamefrequenciesworkinallareas. Currently wireless microphones are licensed on several frequencies, the mostcommonbeing: VHFlowband(AMandFM) FMbroadcast(FM) VHFhighband(FM) UHF(FM)
25to50MHz 72to76MHz 88to108MHz 150to216MHz 470to746MHz 902to952MHz
TheVHF bands are seldom used anymore and can only be found on old equipment. The low band is in the noisiest radio spectrum and, because the wavelengthisabout20ft(6m)itrequiresalongantenna(5ftor1.5m).The VHFlowbandissusceptibletoskip,whichcanbedefinedasexternalsignals fromalongdistanceawaybouncingofftheionospherebacktoearth,creating interference. TheVHFhighbandismorefavorablethanthelowband.The1⁄4-wavelength antennaisonlyabout17inches(43cm)longandrequireslittlespace.TheVHF bandhassomepenetrationthroughbuildingsthatcanbeadvantageousanddisadvantageous.Itisadvantageousinbeingabletocommunicatebetweenrooms andaroundsurfaces.Itisdisadvantageousinthattransmissionisnotcontrolled (security),andoutsidenoisesourcescanreachthereceiver. Most often the frequencies between 174MHz and 216MHz are used in theVHF band, corresponding to television channels 7 to 13. TheVHF high band is free of citizens band and business radio interference, and any commercialbroadcaststationsthatmightcauseinterferencearescheduledsoyou knowwheretheyareandcanavoidthem.Inherentimmunitytonoiseisbuiltin becauseitusesFMmodulation.BetterVHFhigh-bandreceivershaveadequate selectivitytorejectnearbycommercialtelevisionorFMbroadcastsignals.If operating the microphone or intercom on an unused television channel—for instanceChannel7—protectionmightberequiredagainstalocaltelevisionstationonChannel8.AnotherproblemcouldbecausedbyanFMradiostation.If amulti–thousandwattFMstationisbroadcastingneara50mWwirelessmicrophone,evenawell-suppressedsecondharmoniccanhaveanRFfieldstrength comparabletothemicrophoneorintercomsignalbecausethesecondharmonic of FM 88MHz is 176MHz, which is in the middle of television Channel 7. ThesecondharmonicofFM107MHzis214MHz,whichisinthemiddleof
Section | 1
Microphones
131
Channel13.Thus,ifaVHFwirelesssystemistobeutilizedfully,especially with several microphones or intercoms on adjacent frequencies, the wireless receivermusthaveaveryselectivefrontend. Onetelevisionchanneloccupiesa6MHzwidesegmentoftheVHFband. Channel7,forexample,coversfrom174–180MHz.Awirelessintercomoccupiesabout0.2MHz(200kHz).ByFCCPart74allocation,upto24discrete VHFhigh-bandmicrophonesand/orintercomscanbeoperatedinthespace of a single television channel. In order to use multiple systems on adjacent frequencies,thewirelessmicrophone/intercomreceiversmustbeveryselectiveandhaveanexcellentcaptureratio(seeSection1.10.1.3).Onapractical basis,thismeansusingnarrow-deviationFM(approximately12kHzmodulation).Wide-deviationsystems(75kHzmodulationormore)caneasilycause interferenceonadjacentmicrophone/intercomfrequencies;suchsystemsalso requirewidebandwidthreceiversthataremoreapttobeplaguedbyinterference from adjacent frequencies.The trade-off between narrowband FM and wideband FM favor wideband for better overall frequency response, lower distortion, and inherently better SNR versus maximum possible channels withinanunusedTVchannelforequalfreedomfrominterference(max.6). PoorlydesignedFMreceiversarealsosubjecttodesensing.Desensingmeans thewirelessmicrophone/intercomreceiverismutedbecauseanothermicrophone,intercom,televisionstation,orFMstation(secondharmonic)istransmittingincloseproximity;thislimitstheeffectiverangeofthemicrophone orintercom. TheUHFbandequipmentisthebandofchoiceandistheonlyoneusedby manufacturerstoday.Thewavelengthislessthan3ft(1m)sotheantennasare only9inches(23cm).TherangeisnotasgoodasVHF,becauseitcansneak throughsmallopeningsandcanreflectoffsurfacesmorereadily. AlloftheprofessionalsystemsnowareinthefollowingUHFbands: ● ● ●
Aband710–722MHz. Bbandof722–734MHz 728.125–740.500MHzband.
TheFCChasassignedmostoftheDTVchannelsbetweenchannel2and51, andonlyfourchannelsbetween64and69,whichiswheremostoftheprofessionalwirelessmicrophonesoperate.
1.10.1.2. Adjustment of the System’s Operating Frequency Manyoftheprofessionalwirelessmicrophonesarecapableofbeingtunedto manyfrequencies.Inthepastthesystemswerefixedfrequency,oftenbecause thatwastheonlywaytheycouldbemadestable.WithPLL-synthesizedchannels(PhaseLockLoop),itisnotuncommonforsystemstobeswitchtunable to100differentfrequenciesintheUHFbandandhaveafrequencystabilityof 0.005%.ThisisespeciallyimportantwithDTVcomingintothescene.
132
PART | I
Electroacoustic Devices
1.10.1.3 Capture Ratio and Muting Captureratioandmutingspecificationsofthereceiverareimportant.Thecaptureratioistheabilityofthereceivertodiscriminatebetweentwotransmitters transmittingonthesamefrequency.Whenthesignalisfrequencymodulated (FM),thestrongersignalcontrolswhatthereceiverreceives.Thecaptureratio is the difference in the signal strength between the capturing transmitter and thecapturedtransmitterthatisblanketed.Thelowerthenumber,thebetterthe receiverisatcapturingthesignal.Forinstance,areceiverwithacaptureratioof 2dBwillcaptureasignalthatisonly2dBstrongerthanthesecondsignal. MostsystemshaveamutingcircuitthatsquelchesthesystemifnoRFsignal ispresent.Toopenthecircuit,thetransmittersendsaspecialsignalonitscarrierthatbreaksthesquelchandpassestheaudiosignal. 1.10.1.4 RF Power Output and Receiver Sensitivity ThemaximumlegalRFpoweroutputofaVHFhigh-bandmicrophoneorintercomtransmitteris50mW;mostdeliverfrom25–50mW.Upto120mWispermissibleinthebusinessband(forwirelessintercoms)underFCCpart90.217, but even this represents less than 4dB more than 50mW.The FCC does not permit the use of high-gain transmitter antennas, and even if they did, such antennasarelargeanddirectionalsotheywouldnotbepracticalforsomeone whoismovingaround.Incidentally,high-gainreceivingantennasarealsoabad ideabecausethetransmitterisconstantlymovingaroundwiththeperformerso muchofthereceivedradiosignalisactuallycaughtonthebouncefromwalls, props,andsoon(seeSection1.9.2). Even if an offstage antenna is aimed at the performer, it probably would beaimingatthewrongtarget.Diversityreceivingantennasystems,wheretwo ormoreantennaspickupandcombinesignalstofeedthereceiver,willreduce dropoutsorfadesforfixedreceiverinstallations. Thereceivedsignallevelcan’tbeboosted,giventherestrictionsonantenna and transmitted power, so usable range relies heavily on receiver sensitivity andselectivity(i.e.,captureratioandSNR)aswellasontheaudiodynamic range.Inthepre-1980timeframe,mostwirelessmicrophonesandintercoms usedasimplecompressortoavoidtransmitterovermodulation.Today,systems includecompandorcircuitryfor15–30dBbetteraudioSNRwithoutchanging the RF SNR (see Section 1.10.3). This is achieved by building a full-range compressor into the microphone or intercom transmitter, and then providing complementary expansion of the audio signal at the receiver—much like the encoderofatapenoise-reductionsystem.Thecompressionkeepsloudsounds fromovermodulatingthetransmitterandkeepsquietsoundsabovethehissand static.Theexpanderrestorestheloudsoundsafterreceptionandfurtherreduces any low-level hiss or static. Companding the audio signal can provide from 80–85dBofdynamicrangecomparedtothe50–60dBofastraightnoncompandedtransmit/receivesystemusingthesamedeviation.
Section | 1
Microphones
133
1.10.1.5 Frequency Response, Dynamic Range, and Distortion Nowirelessmicrophonewillprovideflatresponsefrom20Hz–20kHz,noris it really needed. Wireless or not, by the time the audience hears the broadcast,film,orconcert,thefrequencyresponsehasprobablybeenreducedtoa bandwidthfrom40Hz–15kHz.Probablythebestcriteriaforjudgingahandheldwirelessmicrophonesystemistocompareittothemicrophonecapsule’s nakedresponse.Ifthetransmit/receivebandwidthbasicallyincludesthecapsule’sbandwidth,itisenough.Generallyspeaking,agoodwirelessmicrophone shouldsoundthesameasahard-wiredmicrophonethatusesthesamecapsule. Wirelessintercomsystems,becausetheyareprimarilyforspeechcommunication,arelesscriticalwithregardtoaudiobandwidth;300Hz–3kHzistelephone quality,and50Hz–8kHzisexcellentforanintercom. Dynamic range is probably the most critical aspect of performance for natural sound.A good compandor system will provide 80–85dB of dynamic range,assumingthemicrophoneisadjustedto100%modulationontheloudest sounds.Leavingamarginofsafetybyturningdownthemicrophonemodulation level sacrifices SNR. Even with extra headroom and a working SNR of 75dB,themicrophonewillstillhaveabouttwicethedynamicrangeofatypical opticalfilmsoundtrackortelevisionshow. Thesystemshouldprovideatleast40–50dBSNRwitha10µVsignaland 70–80dBSNRwithan80µVsignal.Thisshowsupwhennoaudiosignalis beingtransmitted. When an electret condenser microphone is used, a major limitation in dynamicrangecanbethecapsuleitself,notthewirelesssystem.Typically,an electretpoweredbya1.5Vbatteryislimitedtoabout105dBSPL.Poweredby a9Vbattery,thesamemicrophonemaybeusableto120dBSPL.Thewirelessmicrophonesystemshouldbeabletoprovideahighenoughbiasvoltage toensureadequatedynamicrangefromthemicrophonecapsule.Althoughthe condensermaybehotterinoutputlevelthanadynamicmicrophone,itsbackgroundnoiselevelisdisproportionatelyhigher,sotheoverallSNRspecification maybelower. Wirelessintercomsystemsdonotneedthesamedynamicrangeasamicrophone.Theydonothavetoconveyanaturalmusicalperformance.However, natural dynamics are less fatiguing than highly compressed audio, especially givenalongworkshift.Soasidefromgreaterrange,thereareotherbenefitsto seekinggoodSNRintheintercom:40dBor50dBwouldbeusable,and60dB or70dBisexcellent.Anexceptionisinaveryhigh-noiseindustrialenvironment,whereacompressedloudintercomisnecessarytoovercomebackground noise.Agoodintercomheadsetshoulddoubleasahearingprotectorandexclude muchoftheambientnoise. Distortion is higher in a wireless system than in a hard-wired system—a radiolinkwillneverbeascleanasastraightpieceofwire.Still,totalharmonic distortion(THD)specificationsoflessthan1%overalldistortionareavailable
134
PART | I
Electroacoustic Devices
intoday’sbetterwirelessmicrophones.Inthesemicrophones,oneofthelargest contributorstoharmonicdistortionisthecompandor,sodistortionistradedfor SNR.ThewirelessintercomcantoleratemoreTHD,butlowerdistortionwill preventfatigueandimprovecommunication.
1.10.2 Receiving Antenna Systems RFsignaldropoutormultipathcancellationiscausedbytheRFsignalreflecting off a surface and reaching a single receiver antenna 180° out-of-phase withthedirectsignal,Fig.1.145.Thesignalcanbereflectedoffsurfacessuch as armored concrete walls, metal grids, vehicles, buildings, trees, and even people. Although you can often eliminate the problem by experimenting with receiverantennalocation,amorefoolproofapproachistouseaspacediversity systemwheretwoormoreantennaspickupthetransmittedsignal,asshownin Fig.1.146.Itishighlyunlikelythattheobstructionormultipathinterference willaffecttwoormorereceiverantennassimultaneously. There are three diversity schemes: switching diversity, true diversity, and antennacombination. ●
Switching Diversity.Intheswitchingdiversitysystem,theRFsignalsfrom twoantennasarecompared,andonlythestrongeroneisselectedandfedto onereceiver. Reflecting object
Transmitter
Receiver
Direct signal
Reflected signal
Combined (phasecanceled) signal at receiver antenna
FIGURE 1.145 Phasecancellationofradio-frequencysignalsduetoreflections.
Section | 1
●
135
Microphones
True Diversity.Thisreceivingtechniqueusestworeceiversandtwoantennassetupatdifferentpositions,Fig.1.147.Bothreceiversoperateonthe samefrequency.TheAFsignalistakenfromtheoutputofthereceiverthat atanygivenmomenthasthestrongersignalatitsantenna.Theprobability ofnosignalatbothantennasatthesametimeisextremelysmall.TheadvantagesofdiversitycomparedtoconventionalRFtransmissionareshownin
Reflecting object Phasecanceled signal A Diversity receiver
Transmitter
B No reflected signal: thus, no phase canceling
FIGURE 1.146 cancellation.
Diversity antenna system used to reduce multipath radio-frequency phase
Receiver 1
AF
AGC Voltage Comparator
AGC Voltage Receiver 2
AF
FIGURE 1.147 Functionaldiagramofatruediversityreceiver.
136
●
PART | I
Electroacoustic Devices
Fig.1.148. Only the receiving chain with the better input signal delivers audiooutput.Notonlydoesthissystemprovideredundancyofthereceiving end,butitalsocombinessignalstrength,polarity,andspacediversity. Antenna Combination Diversity.Theantennacombinationdiversitysystemisacompromiseoftheothermethods.Thissystemusestwoormore antennas,eachconnectedtoawidebandRFamplifiertoboostthereceived signal.Thesignalsfrombothreceivingantennasarethenactivelycombined andfedtoonestandardreceiverpermicrophone.Inthisway,thereceiver always gets the benefit of the signals present at all antennas.There is no switching noise, no change in background noise, and only requires one receiverforeachchannel.Adrawbackisthepossibilityofcompletesignal cancellationwhenphaseandamplituderelationshipsduetomultipathprovidetheproperunfavorableconditions.
RF Input voltage–mV
1000
Antenna 2
100 25 dB 10
Antenna 1
1 0
1
2
3
4
Time–s FIGURE 1.148 Effectofswitch-overdiversityoperation.SolidlineRFlevelatantenna1andthe dottedlineRFlevelatantenna2.CourtesySennheiserElectronicCorporation.
1.10.2.1 Antenna Placement Itisoftencommontouseanearantennaandafarantenna.Thenearantenna, which is the one nearest the transmitter, produces the majority of the signal mostofthetime;infact,itmayevenbeamplifiedwithanin-lineamplifier.The far-fieldantennamaybeoneormoreantennasusuallyoffsetinelevationand position;therefore,thepossibilityofdropoutisgreatlyreduced.Becausethe antennasarecommontoallreceivers,manywirelessmicrophonescanbeused atthesametimeonthesameantennasystem.Thismeansthattherearefewer antennasandagreaterpossibilityofproperantennaplacement. Thefollowingwillgenerallypreventdeadspots: ● ●
Donotsetupantennasinnichesordoorways. Keep the antennas away from metal objects including armored concrete walls.Minimumdistance:3ft(1m).
Section | 1
●
●
●
Microphones
137
Position the antennas as close as possible to the point where the action takesplace. KeepantennacablesshorttokeepRFlossesataminimum.Itisbettertouse longerAFleadsinstead. Note:Iflongrunsofantennacableareused,besuretheyareofthelow-loss type. Makeawalkaroundtest,i.e.,operatethetransmitteratallpositionswhere it will be used later. Mark all points where field strength is weak.Try to improve reception from these points by changing the antenna position. Repeatthisprocedureuntiltheoptimumresultisachieved.
Interference is mainly caused by spurious signals arriving at the receiver input on the working frequency. These spurious signals may have various causes: ● ●
●
●
●
Twotransmittersoperatingonthesamefrequency(notpermissible). Intermodulationproductsofamultichannelsystemwhosefrequencieshave notbeenselectedcarefullyenough. Excessive spurious radiation from other radio installations—e.g., taxi, police,CB-radio,etc. Insufficientinterferencesuppressiononelectricmachinery,vehicleignition noise,etc. Spurious radiation from electronic equipment—e.g., light control equipment,digitaldisplays,synthesizers,digitaldelays,computers,etc.
1.10.3 Companding Two of the biggest problems with using wireless microphones are SNR and dynamicrange.Toovercometheseproblems,thesignaliscompressedatthe transmitterandexpandedatthereceiver.Figures1.144and1.149graphically illustratehowandwhatthiscanaccomplishwithrespecttoimprovingtheSNR andreducingthesusceptibilitytolow-levelincidentalFMmodulation,suchas buzzzones. Asthetypicalinputlevelchangesbyafactorof80dB,theaudiooutputto themodulatorundergoesacontouredcompression,soachangeininputaudio levelistranslatedintoapseudologarithmicoutput.Thisincreasestheaverage modulation level, which reduces all forms of interference encountered in the transmissionmedium. Byemployingstandardnarrowbandtechniquesatthereceiver,therecovered audioisvirtuallyfreeofadjacentchannelandspuriousresponseinterference.In addition,uptotentimesthenumberofsystemscanbeoperatedsimultaneously withoutcross-channelinterference.Theabilityofthereceivertorejectallforms ofinterferenceisimperativewhenutilizingexpansionandcompressiontechniquesbecausethereceivermustcomplementarilyexpandtheaudiocomponent torestoretheoriginalsignalintegrity.
138
PART | I
10
Dynamic range without companding
Electroacoustic Devices
Dynamic range with companding 10
0
0
210
210
220
220
230
230
240
240
250
250
260
260
270 280 290
270
System noise plus modulation demodulating noise
2100 Compression Transmission at the transmitter path
280 290 2100 Expansion at the receiver
FIGURE 1.149 Compressionandexpansionoftheaudiosignal.Noticethe−80dBsignalisnot altered,andthe−20dBsignalisalteredsignificantly.
1.10.4 Waterproof Wireless Microphone Systems Wirelessmicrophonesthatarewornareveryusefulforcoachingallformsof athleticsincludingswimmingandaquaticaerobics.Iftheinstructoralwaysstays onthepooldeck,aweatherproofsystemmightbeadequate.Iftheinstructorisin thewater,acompletelysubmersibleandwaterproofsystemwillberequired. Hydrophonicsassemblesacompletelywaterproofandsubmersiblewireless microphone system.Assembled with Telex components, the system includes aheadsetmicrophonewithaspecialwaterproofconnectorandaTelexVB12 waterproofbeltpacktransmitter.Thetransmittercanoperateona9Valkaline battery or a 9V NiMH rechargeable battery. The rechargeable battery is recommendedasitdoesnotrequireremovingthebatteryfromthetransmitterfor rechargingandthereforereducesthechanceofwaterleakingintothetransmitter housing.ThereceiverisaTelexVR12forout-of-pooloperation,andcanbeconnectedtoanysoundsystemthesamewayasanyotherwirelessmicrophone. Aninterestingthingaboutthissystemisyoucandiveintothewaterwhile wearingthesystemandcomeupandimmediatelytalkasthewaterdrainsout ofthewindscreenrapidly. The DPAType 8011 hydrophone, Fig.1.150, is a 48V phantom powered waterproof microphone specially designed to handle the high sound pressure levelsandthehighstaticambientpressureinwaterandotherfluids.Thehydrophoneusesapiezoelectricsensingelement,whichisfrequencycompensatedto matchthespecialacousticconditionsunderwater.A10mhigh-qualityaudio
Section | 1
Microphones
139
FIGURE 1.150 DPA8011hydrophone.CourtesyDPAMicrophonesA/S.
cableisvulcanizedtothebodyofthehydrophoneandfittedwithastandard three-pinXLRconnector.Theoutputiselectronicallybalancedandoffersmore than100dBdynamicrange.The8011hydrophoneisagoodchoiceforprofessionalsoundrecordingsinwaterorunderotherextremeconditionswhereconventionalmicrophoneswouldbeadverselyaffected.
1.11 MULTICHANNEL WIRELESS MICROPHONE AND MONITORING SYSTEMS By Joe Ciaudelli and Volker Schmitt
1.11.1 Introduction The use of wireless microphones and monitoring systems has proliferated in thepastfewyears.Thisisduetoadvancementsintechnology,atrendtowards greatermobilityonstage,andthedesiretocontrolvolumeandequalizationof individualperformers.Consequently,installationsinwhichanumberofwirelessmicrophones,referredtoaschannels,arebeingusedsimultaneously,have increaseddramatically.Nowtheatresandstudioswithlargemultichannelsystems,greaterthanthirtychannels,arecommon.Systemsofthismagnitudeare adifficultengineeringchallenge.Carefulplanning,installation,operation,and maintenancearerequired. Wirelesssystemsrequireatransmitter,transmitantenna,andreceivertoprocesssoundviaradio-frequency(RF)transmission.First,thetransmitterprocesses thesignalandsuperimposesitonacarrierthroughaprocesscalledmodulation. Thetransmitantennathenactsasalaunchpadforthemodulatedcarrierandbroadcaststhesignaloverthetransfermedium:air.Thesignalmustthentravelacertain space or distance to reach the pickup element, which is the receiving antenna.
140
PART | I
Electroacoustic Devices
Finishinguptheprocess,thereceiver—whichselectsthedesiredcarrier—strips off the signal through demodulation, processes it, and finally reconstitutes the originalsignal.Eachwirelesschannelneedstooperateonauniquefrequency.
1.11.2 Frequencies Manufacturersgenerallyproducewirelessmicrophonesonultrahighfrequencies (UHF)withintheTVbandwithspecificationsoutlinedbygovernmentagenciessuchastheFederalCommunicationsCommission(FCC).Thewavelength is inversely proportional to the frequency. Higher frequencies have shorter wavelengths.UHFfrequencies(450–960MHz)haveawavelengthoflessthan 1meter.Theyhaveexcellentreflectivecharacteristics.Theycantravelthrougha longcorridor,bouncingoffthewalls,losingverylittleenergy.Theyalsorequire lesspowertotransmitthesamedistancecomparedtomuchhigherfrequencies, suchasmicrowaves.Theseexcellentwavepropagationcharacteristicsandlow powerrequirementsmakeUHFidealforperformanceapplications.
1.11.3 Spacing Inordertohaveadefinedchannel,withoutcrosstalk,aminimumspacingof300 KHzbetweencarrierfrequenciesshouldbeemployed.Awiderspacingiseven morepreferablesincemanyreceiversoftenexhibitdesensitizedinputstagesin thepresenceofcloselyspacedsignals.However,cautionshouldbeusedwhen linkingreceiverswithwidelyspacedfrequenciestoacommonsetofantennas. Thefrequenciesneedtobewithinthebandwidthoftheantennas.
1.11.4 Frequency Deviation ThemodulationofthecarrierfrequencyinanFMsystemgreatlyinfluencesits audioquality.Thegreaterthedeviation,thebetterthehigh-frequencyresponse andthedynamicrange.Thetrade-offisthatfewerchannelscanbeusedwithin a frequency range. However, since audio quality is usually the priority, wide deviationismostdesirable.
1.11.5 Frequency Coordination Multichannelwirelessmicrophonesystemscanbeespeciallydifficulttooperate, as they present several special conditions. Multiple transmitters moving aroundastagewillresultinwidevariationsoffieldstrengthseenatthereceiver antenna system. This makes frequency selection to avoid interference from intermodulation (IM) products highly critical.This is even more challenging inatouringapplicationsincetheRFconditionsvaryfromvenuetovenue.In thiscase,themixoffrequenciesisconstantlychanging.Thedauntingtaskto coaxeachofthesevariablestoexecuteclearaudiotransmissioncanbeachieved throughcarefulfrequencycoordination.
Section | 1
141
Microphones
Intermodulationistheresultoftwoormoresignalsmixingtogether,producingharmonicdistortion.Itisacommonmisconceptionthatintermodulation isproducedbythecarrierfrequenciesmixingwithintheair.Intermodulation occurs within active components, such as transistors, exposed to strong RF inputsignals.Whentwoormoresignalsexceedacertainthreshold,theydrive the active component into a nonlinear operating mode and IM products are generated.ThisusuallyhappensintheRFsectionofthereceiver,inantenna amplifiers,ortheoutputamplifierofatransmitter.Inmultichanneloperation, whenseveralRFinputsignalsexceedacertainleveltheIMproductsgrowvery quickly.Therearedifferentlevelsofintermodulationsdefinedbythenumber ofadditionterms. Inanywirelesssystemwiththreeormorefrequenciesoperatinginthesame range,frequencycoordinationisstronglyadvised. ItisnecessarytoconsiderpossibleIMfrequenciesthatmightcauseproblemsfortheaudiotransmission.The3rdand5thharmonics,inparticular,might raiseinterferenceissues. Thefollowingsignalsmaybepresentattheoutputofanonlinearstage: Fundamentals: SecondOrder: ThirdOrder: FourthOrder: FifthOrder: Additionalhigherorders
F1andF2 2F1,2F2,F1±F2,F2−F1 3F1,3F2,2F1±F2,2F2±F1 4F1,4F2,2F1±2F2,2F2±2F1 5F1,5F2,3F1±2F2,3F2±2F1
Asaresult,theintermodulationfrequenciesshouldnotbeused,asthose frequencies are virtual transmitters. The fundamental rule never use two transmitters on the same frequencyisvalidinthiscase.However,even-order products are far removed from the fundamental frequencies and, for simplicity, are therefore omitted from further considerations. Signal amplitude rapidly diminishes with higher-order IM products, and with contemporary equipmentdesign,considerationofIMproductscanbelimitedto3rdand5th orderonly. FormultichannelapplicationssuchasthoseonBroadway(i.e.,30+channels),theIMproductscanincreasesignificantlyandthecalculationofintermodulation-freefrequenciescanbedonebyspecialsoftware.Bylookingonly atthethirdharmonicdistortioninamultichannelsystem,thenumberofthirdorderIMproductsgeneratedbymultiplechannelsis: ● ● ● ● ● ●
2channelsresultin2. 3channelsresultin9. 4channelsresultin24. 5channelsresultin50. 6channelsresultin90. 7channelsresultin147.
142
● ● ●
PART | I
Electroacoustic Devices
8channelsresultin225. …. 32channelsresultin15,872third-orderIMproducts.
Addingmorewirelesslinkstothesystemwillincreasethenumberofpossiblecombinationswithinterferencepotentiallogarithmically:nchannelswill resultin(n3− n2)/2third-orderIMproducts.Equalfrequencyspacingbetween RFcarrierfrequenciesinevitablyresultsintwo-andthree-signalIMproducts andmustbeavoided! TheRFlevelandtheproximitydefinetheleveloftheIMproduct.Iftwo transmittersareclose,thepossibilityofintermodulationwillincreasesignificantly.Assoonasthedistancebetweentwotransmittersisincreased,theresulting IM product decreases significantly. By taking this into consideration, the physicaldistancebetweentwoormoretransmittersisimportant.Ifaperformer needstoweartwobodypacktransmitters,itisrecommendedtousetwodifferentfrequencyrangesandtowearonewiththeantennapointingupandtheother withitpointingdown. If the number of wireless channels increases, the required RF bandwidth increasessignificantly,Fig.1.151. External disturbing sources such as TV transmitters, taxi services, police services,digitalequipment,etc.,alsohavetobetakenintoconsideration.Fortunately,thescreeningeffectofbuildingsisratherhigh(30–40dB).Forindoor applications,thiseffectkeepsstrongoutsidesignalsatlowlevels.Asignificant problemcanoccurwhenpoorlyscreeneddigitalequipmentisworkinginthe sameroom.Thesewidebanddisturbingsourcesareabletointerferewithwirelessaudioequipment.Theonlysolutiontothisproblemistoreplacethepoorly screenedpieceofequipmentwithabetterone. OtherRFsystemsthathavetobeconsideredforcompatibilityare: 1. TVstations“On-Air.” 2. Wirelessintercoms.
Bandwidth–MHz
30 25 20 15 10 5 2 45
10 15 Number of channels
FIGURE 1.151 Bandwidthrequiredformultichannelsystems.
20
Section | 1
Microphones
143
3. IFBs. 4. Wirelessmonitorsystems. 5. Otherwirelesssystems. Compatibilitybetweencomponentsofasystemisachievedifthefollowing requirements are met: each link in a multichannel wireless system functions equallywellwithallotherlinksactiveandnosinglelink—oranycombination ofmultiplelinks—causesanyinterference. If the transmitter of a wireless mic channel is switched off, its complementaryreceivershouldalsobeswitchedofformutedatthemixingconsole. Areceiverthatdoesnotseeitstransmitterwilltrytolatchontoanearbysignal. ThatsignalmaybeanIMproduct.Thereceiverwillthentrytodemodulatethis signalandapplyittothespeakersystem. Equipment can be designed to minimize intermodulation.A specification knownasintermodulation rejectionorsuppressionisameasureoftheRFinput threshold before intermodulation occurs. For a well-designed receiver, this specification will be 60dB or greater.An intermodulation rejection of 60dB meansthatIMproductsaregeneratedatinputlevelsofapproximately1mV. Thehighestqualitymultichannelreceiverscurrentlyavailablefeatureanintermodulationrejectionof>80dB.Ifhigh-qualitycomponentsareused,havingan intermodulationsuppressionof60dBorgreater,onlythethird-orderproducts needtobeconsidered.
1.11.6 Transmitter Considerations Transmittersarewidelyavailableasportabledevices,suchashandheldmicrophones, body packs, and plug-on transmitters and are produced in stationary form as stereo monitors. When transmitting signals for most wireless applications via air, FM modulation is generally used; in doing so, one must also improvethesoundqualityinavarietyofways. AnRFtransmitterworkslikeaminiatureFMradiostation.First,theaudio signal of a microphone is subjected to some processing. Then the processed signalmodulatesanoscillator,fromwhichthecarrierfrequencyisderived.The modulatedcarrierisradiatedviathetransmitter’santenna.Thissignalispicked upbyacomplementaryreceiverviaitsantennasystemandisdemodulatedand processedbacktotheoriginalaudiosignal.
1.11.6.1 Range and RF Power TransmitterpowerisaratingofitspotentialRFsignalstrength.Thisspecification is measured at the antenna output.The range of a wireless transmission depends on several factors. RF power, the operating frequency, the setup of thetransmitterandreceiverantennas,environmentalconditions,andhowthe transmitterisheldorwornareallaspectsthatdeterminetheoverallcoverageof
144
PART | I
Electroacoustic Devices
thesystem.Therefore,powerspecificationsareofonlylimiteduseinassessing atransmitter’srange,consideringthesevariableconditions.Also,batterylifeis associatedwithRFoutputpower.Increasedpowerwillreducebatterylifewith onlyamoderateincreaseinrange. Using RF wireless microphone transmitters with the right amount of RF output power is important to ensure total system reliability. There is a commonmisconceptionthathigherpowerisbetter.However,inmanyapplications highpowercanaggravateintermodulation(IM)distortion,resultinginaudible noises. Firstofall,theappliedRFoutputpowermustfallwithinthelimitallowed by each country’s legislation. In the United States, the maximum RF output powerforwirelessmicrophonesislimitedto250mW.Inmostofthecountries inEuropethisfigureis50mW,whileinJapanitisonly10mW.Despitethe 10mW limitation, many multichannel wireless microphones are operating in Japan. This is achieved by careful attention to factors like antenna position, use of low loss RF cables and RF gain structure of the antenna distribution system. ThereareindeedsomeapplicationsinwhichmoreRFoutputpowerisan appropriate measure; a perfect example would be a golf tournament, as the wirelesssystemneedstocoverawidearea.Thereareusuallyonlyafewwireless microphones in use at this type of function, and those microphones are generallynotincloseproximitytoeachother. IftransmitterswithhighRFpowerareclosetogether,intermodulationusually occurs.At the same time, the RF noise floor in the performance area is increased.Asamatteroffact,atransmitterincloseproximitytoanothertransmitterwillnotonlytransmititsownsignal,butitwillalsoreceivethesignaland addthistotheRFamplifierstage.
1.11.6.2 Dc-to-Dc Converter Transmitters should be designed to provide constant RF output power and frequency deviation throughout the event being staged.This can be achieved throughtheuseofadc-to-dcconvertercircuit.Suchacircuittakesthedecaying batteryvoltageasitsinputandregulatesittohaveaconstantvoltageoutput. Once the voltage of the batteries drops below a minimum level, the dc-to-dc converter shuts off, almost instantaneously.The result is a transmitter that is essentially either off or on. While it is on, the RF output power, frequency deviation,andotherrelevantspecificationsremainthesame.Transmitterswithoutregulationcircuits,oncethebatteryvoltagebeginstodrop,willexperience reducedrangeandaudioquality. 1.11.6.3 Audio Processing To improve the audio quality, several measures are necessary because of the inherentnoiseoftheRFlink.
Section | 1
Microphones
145
1.11.6.3.1 Pre- and De-Emphasis ThismethodisastaticmeasurethatisusedinmostoftheFMtransmissions.By increasingthelevelofthehigheraudiofrequenciesonthetransmitterside,the signal-to-noiseratioisimprovedbecausethedesiredsignalisabovetheinherentnoiseflooroftheRFlink. 1.11.6.3.2 Companding The compander is a synonym for compressor on the transmitter side and for expanderonthereceivingend.Thecompressorraiseslowaudiolevelabovethe RFnoisefloor.Theexpanderdoesthemirroroppositeandrestorestheaudio signal.Thismeasureincreasesthesignal-to-noiseratiotoCDqualitylevel. 1.11.6.3.3 Spurious Emissions Apart from the wanted carrier frequency, transmitters can also radiate some unwanted frequencies known as spurious emissions. For large multichannel systemsthesespuriousfrequenciescannotbeignored.Theycanbesignificantly reducedthroughelaboratefilteringandcontainedbyusingawell-constructed, RFtightmetalhousingforthetransmitter.Also,anRFtighttransmitterisless susceptibletooutsideinterference. Ametalhousingisimportantnotonlyforitsshieldingproperties,butalso itsdurability.Thesedevicesusuallyexperiencemuchmoreabusebyactorsand othertalentthananyoneeverpredicts.
1.11.6.4 Transmitter Antenna Every wireless transmitter is equipped with an antenna, which is critically importanttotheperformanceofthewirelesssystem.Ifthistransmitterantenna comes in contact with the human body, the transmitted wireless energy is reducedandmaycauseaudiblenoisesknownasdropouts.Thiseffectofdetuningtheantennaoncontactiscalledbodyabsorption. Forthisreason,talentshouldnottouchtheantennawhileusinghandheld microphones.Unfortunately,thereisnoguaranteethattheywillfollowthisrecommendation.Takingthisintoaccount,optimizedantennasetupatthereceiver sideandtheoverallRFgainstructureofthesystembecomescritical. Thissameeffectcanoccurwhenusingbodypacktransmitters,especially if the talent is sweating.A sweaty shirt can act as a good conductive materialtotheskin.Ifthetransmitterantennatouchesit,reducedpowerandthus poorsignalqualitymayresult.Inthiscase,apossibleapproachistowearthe bodypackupsidedownnearorattachedtothebelt,withtheantennapointing down.Sometimesthismeasuredoesnotworkbecausethetalentwillsitonthe antenna.Inthiscase,apossiblesolutioniskeepingthetransmitterinthenormal positionandfittingathick-walledplastictubeovertheantenna,suchasthetype thatisusedforaquariumfilters.
146
PART | I
Electroacoustic Devices
1.11.7 Receiver Considerations Thereceiverisacrucialcomponentofwirelessaudiosystems,asitisusedto pickthedesiredsignalandtransferitselectricalinformationintoanaudiosignal.Understandingbasicreceiverdesign,audioprocessing,squelch,anddiversityoperationcanhelpensureoptimumperformanceofthesystem. Virtuallyallmodernreceiversfeaturesuperheterodynearchitecture,inwhich thedesiredcarrierisfilteredoutfromthemultitudeofsignalspickedupbythe antenna,thenamplifiedandmixedwithalocaloscillatorfrequencytogenerate thedifference:intermediate frequency (IF).ThisIFundergoesmorecontrolled discriminationandamplificationbeforethesignalisdemodulatedandprocessed torestoretheoutputwithallthecharacteristicsandqualitiesoftheoriginal. Audio signal processing of a receiver is the mirror opposite of the transmitter.Processingdoneinthetransmittersoftenincludespre-emphasis(boostinghighaudiofrequencies)aswellascompression.Thesearereversedinthe receiverbythede-emphasisandtheexpandercircuit. AninherentRFnoisefloorexistsintheair.Thesquelchsettingshouldbeset abovethisnoiselevel.Thisactsasanoisegatethatmutestheaudiooutputifthe wantedRFsignalfallsbelowathresholdlevel.Thispreventsablastofwhite noisethroughthePAiftheRFsignaliscompletelylost.Ifthesquelchsetting istoolow,thereceivermightpickthenoisefloorandthisnoisecanbeheard.If thesquelchsettingistoohightherangeofthewirelessmicrophoneisreduced.
1.11.7.1 RF Signal Level VaryingRFsignalstrengthismainlyduetomultipathpropagation,absorption, andshadowing.Thesearefamiliardifficultiesalsoexperiencedwithcarradios incities. AudibleeffectsduetolowRFsignals,knownasdropouts,canoccurevenat closerangetothereceiverduetomultipathpropagation.Someofthetransmittedwavesfindadirectpathtothereceiverantennaandothersaredeflectedoffa wallorotherobject.Theantennadetectsthevectorsum,magnitude,andphase ofdirectanddeflectedwavesitreceivesatanyparticularinstant.Adeflected wavecandiminishadirectwaveifithasdifferentphase,resultinginanoverall lowsignal.Thisdifferenceinphaseisduetothelongerpathadeflectedwave travels between the transmitter and receiver antennae and any phase reversal occurring when it hits an object.This phenomenon needs to be addressed in anindoorapplicationsincethefieldstrengthvariationinsideabuildingwith reflectingwallsis40dBormore.Itislesscriticaloutside. RFenergycanbeabsorbedbynonmetallicobjectsresultinginlowsignal strength.Asstatedpreviously,thehumanbodyabsorbsRFenergyquitewell.It isimportanttoplaceantennascorrectlytominimizethiseffect. Shadowingoccurswhenawaveisblockedbyalargeobstaclebetweenthetransmitterandreceiverantennas.Thiseffectcanbeminimizedbykeepingtheantennas highanddistanceof1⁄2wavelengthawayfromanylargeormetallicobjects.
Section | 1
Microphones
147
These problems are addressed by a diversity receiver.A diversity system isrecommendedevenifonlyonechannelisinoperation.Largemultichannel systemsareonlypossiblewithdiversityoperation. Therearedifferentkindsofdiversityconceptsavailable.Antennaswitching diversity uses two antennas and a single receiving circuit. If the level at one antenna falls below a certain threshold it switches to the other antenna.This isaneconomicalarchitecturebutitleavesthechancethatthesecondantenna could be experiencing an even lower signal than the one that falls below the thresholdlevel.Anotherapproachistheswitchingoftheaudiosignaloftwo independent receiver units where each receiver unit is connected to its own antenna.Thisisknownastrue diversity.Thistechniqueimprovestheeffective RFreceivinglevelbygreaterthan20dB.Dependingonthediversityconcept, activeswitchingbetweenthetwoantennasisadesiredresult. Theminimumdistancebetweenthetwodiversityantennasisveryoftenan issueofdebate.Aminimumof1⁄4ofawavelengthofthefrequencywaveseems tobeagoodapproach.Dependingonthefrequency,5–6inchesistheminimum distance.Ingeneral,agreaterdistanceispreferred.
1.11.8 Antennas Thepositionoftheantennaandthecorrectuseofitsrelatedcomponents—such astheRFcable,antennaboosters,antennaattenuators,andantennadistribution systems—arethekeytotrouble-freewirelesstransmission.Theantennasactas theeyesofthereceiver,sothebestresultscanbeachievedbyformingadirect lineofsightbetweenthetransmitterantennaandreceiverantennaofthesystem. Receiving and transmitting antennas are available as omnidirectional and directionalvariants. Forreceiving,omnidirectionalantennasareoftenrecommendedforindoor usebecausetheRFsignalisreflectedoffofthewallsandceiling.Whenworkingoutside,oneshouldchooseadirectionalantennasincethereareusuallylittle tonoreflectionsoutdoors,andthisdirectivitywillhelptostabilizethesignal.In general,itiswisetokeepanantennatoolboxthatcontainsbothomnidirectional anddirectionalantennasforuseincriticalRFsituations,sincetheytransmitand receivesignalsdifferently. Omnidirectionalantennastransmitorreceivethesignalbyprovidinguniformradiationorresponseonlyinonereferenceplane,whichisusuallythehorizontaloneparalleltotheearth’ssurface.Theomnidirectionalantennahasno preferreddirectionandcannotdifferentiatebetweenawantedandanunwanted signal. Ifadirectionalantennaisused,itwilltransmitorreceivethesignalinthe pathitispointingtoward.Themostcommontypesaretheyagiantennaandthe log-periodicantenna,whichareoftenwide-rangefrequencyantennascovering thewholeUHFrange.Inanoutdoorvenue,thedesiredsignalcanbereceived andtheunwantedsignalfromaTVstationcanberejectedtoacertaindegree
148
PART | I
Electroacoustic Devices
bychoosingthecorrectantennaposition.Adirectionalantennaalsotransmitsor receivesonlyinoneplane,likeanomnidirectionalantenna. Severaltypesofomnidirectionalanddirectionalantennasalsoexistforspecificconditions.Thetelescopicantennaisanomnidirectionalantennaandoften achievesawiderange(450–960MHz).Iftelescopicantennasareinusetheyshould beplacedwithinthelineofsightofthecounterpartantenna.Theyshouldnot,for example,bemountedinsideametalflightcasewithcloseddoorsasthiswillreduce theRFfieldstrengthfromthetransmitterandcompromisetheaudioquality. Systemperformancewillberaisedconsiderablywhenremoteantennasare used.Aremoteantennaisonethatisseparatedfromthereceiverortransmitter unit.Theseantennascanbeplacedonastandsuchasthatforamicrophone.This will improve the RF performance significantly. However, when using remote antennas,somebasicrulesneedtobeconsidered.Onceagain,aclearlineofsight shouldbeestablishedbetweenthetransmitterandreceiverantenna,Fig.1.152. Placing antennas above the talent increases the possibility the transmitter andreceiverremainwithinlineofsight,ensuringtrouble-freetransmission. Ifadirectionalantennaisused,thepositionoftheantennaandthedistance tothestageisimportant.Onecommonsetupispointingbothreceivingantennas towardthecenterofthestage.Onceagain,alineofsightbetweenthereceiver andtransmitterantennasisbestforoptimumtransmissionquality. Directionalandomnidirectionalantennasdohaveapreferredplane,whichis eitherthehorizontalorverticalplane.Ifthepolarizationbetweenthetransmitter andreceiverantennaisdifferent,thiswillcausesomesignificantlossoftheRF level.Unfortunately,itisnotpossibletohavethesamepolarizationoftheantennas allofthetime.Inatheatricalapplication,theantennaisinaverticalpositionwhen theactressoractorwalksonthestage.Thepolarizationofthetransmittermay changetothehorizontalpositionifascenerequiresthetalenttoliedownorcrawl acrossthestage.Inthiscase,circularpolarizedantennascanhelp.Thesekindsof antennascanreceivetheRFsignalinallplaneswiththesameefficiency. Becausethepolarizationoftheantennaiscriticalandtelescopicantennas areoftenused,itisnotrecommendedtousethereceiverantennasstrictlyina horizontalorverticalplane.Rather,angletheantennasslightlyasthiswillminimize the possibility that polarization would be completely opposite between transmitterandreceiver.
> 2 m Antenna height Stage FIGURE 1.152 Placing antennas above the talent increases the possibility the transmitter and receiverremainwithinlineofsight,ensuringtrouble-freetransmission.
Section | 1
149
Microphones
Onelastnote:Thepluralformforthetypeofantennadiscussedinthisarticleisantennas.Antennaearefoundoninsectsandaliens.
1.11.8.1 Antenna Cables and Related Systems Antenna cables are often an underestimated factor in the design of a wirelesssystem.Thedesignermustchoosethebestcableforpracticalapplication, depending on the cable run and the installation,Table 1.3.As the RF travels downthecableitsamplitudeisattenuated.Theamountofthislossisdependent onthequalityofthecable,itslength,andtheRFfrequency.Thelossincreases withlongercableandhigherfrequencies.Bothoftheseeffectsmustbeconsideredforthedesignofawirelessmicrophonesystem. RF cables with a better specification regarding RF loss are often thicker. Thesearehighlyrecommendedforfixedinstallations.Inatouringapplication, inwhichthecablemustbestoredawayeachday,theseheaviercablescanbe verycumbersome. AsanyRFcablehassomeRFattenuation,cablelengthshouldbeasshort aspossiblewithoutsignificantlyincreasingthedistancebetweenthetransmitter andreceiverantennas.Thisaspectisimportantforreceivingapplicationsbutis evenmorecriticalforthetransmissionofawirelessmonitorsignal.
TABLE 1.3 Different Types of RF Cables with Various Diameters and the Related Attenuation for Different Frequencies Cable Type
Frequency (MHz)
Attenuation (db/100’)
Attenuation (dB/100 m)
Cable Diameter (inches/mm)
RG-174/U
400 700
19.0 27.0
62.3 88.6
0.110 / 2.8
RG-58/U
400 700
9.1 12.8
29.9 42.0
0.195 / 4.95
RG-8X
400 700
6.6 9.1
21.7 29.9
0.242 / 6.1
RG-8/U
400 700
4.2 5.9
13.2 19.4
0.405 / 10.3
RG-213
400 700
4.5 6.5
14.8 21.8
0.405 / 10.3
Belden 9913
400 700
2.7 3.6
8.9 11.8
0.405 / 10.3
400 700
2.9 3.9
9.5 12.8
0.405 / 10.3
Belden 9913F 9914
Source: Belden Master Catalogue.
150
PART | I
Electroacoustic Devices
Inareceivingapplication,itisimportanttoconsiderlossesfromthecable aswellasfromanysplitterintheantennasystemduringthedesignandconcept stageofawirelessmicrophonesystem.Ifthelossesinthesystemaresmall, anantennaboostershouldnotbeused.Inthiscase,anydropoutisnotrelated to the RF loss in the antenna system; instead, it is more often related to the antennapositionandhowthetransmitterisusedandwornduringtheperformance.Anantennaboosterisrecommendedifthelossintheantennasystem isgreaterthan6dB. Ifanantennaboosterisnecessary,itshouldbeplacedascloseaspossible tothereceivingantenna.Antennaswithabuilt-inboosterareknownasactive antennas. Some of these have a built-in filter, only allowing the wanted frequencyrangetobeamplified.Thisisrecommendedbecauseitreducesthepossibilityofintermodulation. Twoantennaboostersshouldnotbeusedback-to-backwhentheRFcable runisverylong.Thesecondantennaboosterwouldbeoverloadedbytheoutput ofthefirstamplifierandwouldproduceintermodulation. Specialcaremustbetakenwhenusinganantennaboosterifthetransmitter comesclosetothereceiverantenna.Theresultingstrongsignalcoulddrivethe antennaboosterpastitslinearoperationrange,thusproducingintermodulation products.Itisrecommendedtodesignandinstallasystemsuchthatthetransmitterremainsatleast10feetfromthereceiverantennaatalltimes. Anotherimportantfactoristhefilterattheinputstageoftheantennabooster. TheapproachistoreducetheamountofunwantedsignalsintheRFdomainas much as possible. This is another measure to reduce the possibility of intermodulationofthisamplifier. Also, signals that come from a TV station—such as a digital television (DTV)signal—areunwantedsignalsandcanbethereasonforintermodulation productsinthefirstamplifier. IfthefreeTVchannelbetweentheDTVshouldbeusedforwirelessmicrophonetransmission,theDTVsignalsmightcausetheproblems.Toreducethe effectofDTVsignals,anarrowinputfilterwillhelptoovercomethepossible effectofintermodulation. Often a narrower filter at the input stage of a wireless receiver is preferable.This will often work for fixed installations because there are decreased possibilitiesthattheRFenvironmentwillchange.Thisisespeciallythecase whentheRFenvironmentisdifficultandalotofTVstationsorotherwireless systemsareoperating.
1.11.8.2 Splitter Systems Antenna splitters allow multiple receivers to operate from a single pair of antennas.Activesplittersshouldbeusedforsystemsgreaterthanfourchannelssothattheamplifierscancompensateforthesplitterloss.Securityfrom interference and intermodulation can be enhanced by filtering before any amplifier stage.As an example, a 32-channel system could be divided into
Section | 1
151
Microphones
Filter Antenna A AB 400
Amp Splitter
Antenna B AB 400
SAS 432 Diversity side A Diversity side B
Range 1 Range 2 Range 3 Range 4
Range 1 Range 2 Range 3 Range 4
FIGURE 1.153 Diversity antenna setup with filtered boosters, long antenna cables, and active splitterwithselectivefiltering.
foursubgroupsofeightchannels.Thesubgroupscanbeseparatedfromeach otherbyhighlyselectivefilters.Thesubgroupscanthenbeconsideredindependentofeachother.Inthisway,frequencycoordinationonlyneedstobe performedwithineachgroup.Itismucheasiertocoordinateeightfrequenciesfourtimesthantoattempttocoordinateasinglesetof32frequencies, Fig.1.153.
1.11.9 Wireless Monitor Systems Wireless monitor systems are essential for stage-bound musical productions. Perhapsthebiggestadvantageofawirelessmonitorsystemistheabilitytouse anindividualmonitormixforeachmusicianonstage.Furthermore,awireless monitorsystemsignificantlyreducestheamountof,oreveneliminates,monitor speakersintheperformancearea.Thisresultsinlowerriskoffeedbackanda morelightweight,compactmonitorsystem. Somespecialprecautionsmustbetakenbeforeusingwirelessmonitorsystems. In most cases, this signal is a stereo signal.This multiplexed signal is moresensitivetodropoutsandstaticandmultipathsituations.Forlongrange applications,monooperationcanimprovesystemperformance. Ifwirelessmicrophonesandwirelessmonitorsystemsareusedinparallel, thosesystemsshouldbeseparatedinawaythatthefrequenciesareatleast8MHz apartandthatthephysicaldistancebetweenthetransmitterandthereceiveris maximized.Thiswillreducetheriskofblocking—aneffectthatdesensitizes
152
PART | I
Electroacoustic Devices
a receiver and prevents the reception of the desired signal. Therefore, if a body pack wireless mic transmitter and a wireless monitor receiver are both attachedtothesametalent,thosedevicesshouldnotbemounteddirectlybeside eachother. When musicians use the same monitor mix, one transmitter can be used to provide the RF signal to more than one wireless monitor receiver. If individualmixesaredesired,eachmixrequiresitsowntransmitteroperatingona uniquefrequency.Toavoidintermodulationdisturbances,thewirelessmonitor transmittersshouldbecombined,andthecombinedsignalshouldthenbetransmitted via one antenna.Active combiners are highly recommended. Passive combinerssufferfromsignallossandhighcrosstalk.Anactivecombinerisolateseachtransmitterbyaround40dBfromtheotherandkeepstheRFlevelthe same(0dBgain),thusminimizingintermodulation.Again,intermodulationisa majorissuewithintheentirewirelessconcept.Whenusingstereotransmission, itisevenmorecritical. Whenconsideringanexternalantenna,oneimportantfactormustbetaken into consideration: the antenna cable should be as short as possible to avoid losses via the RF cable. A directional external antenna is recommended to reducemultipathsituationsfromreflections,anditwillhavesomeadditional passivegainthatwillincreasetherangeofthesystem. Ifremoteantennasareusedforthewirelessmonitortransmittersaswell as wireless mic receivers, those antennas should be separated by at least 10–15 feet. Blocking of the receivers, as discussed above, is then avoided. Furthermore,theantennasshouldnotcomeindirectcontactwiththemetalof thelightingrig.Thiswilldetunetheantennaandreducetheeffectiveradiated wirelesssignal.
1.11.10 System Planning for Multichannel Wireless Systems When putting together a multichannel wireless microphone system, several items are essential for trouble-free operation. First, you must understand the environmentinwhichthesystemwillbeused. Location. Thelocationofavenuecanbedeterminedbyusingmappingtools ontheInternet,suchasGoogleEarth.Ifyoufigureoutthecoordinatesofthe venue,simplyplugthisinformationintotheFCChomepage,http://www.fcc. gov/fcc-bin/audio/tvq.html.The result shows all transmitters licensed by the FCCinthisarea.ThisinformationwillallowthedesignerofthewirelesssystemtoplanwhichvacantTVchannelscanbeusedforwirelessaudiodevices. IfthereisaTVtransmitterclosetothelocationofthewirelessmicrophonesystem(10ms,thetimingofthebandwillbethrownoff.Furthermore,ifstreamingvideoisprojectedtoaccommodatealargeaudiencethe pictureandsoundwillbeoutofsync. New audio data compression algorithms show good performance with a verylowlatency.However,audiocompressionwouldintroducethepossibility ofaudibleartifacts(atleastwithawkwardsignals). Astechnologyimproves,therewillbesolutionstotheobstaclesdescribed aboveanddigitalwillbecomeavailableforwirelesstransmission. Thekeyquestionsforadigitalsystematthistimeare: ● ●
● ●
Isdatacompressionused? What RF spectrum is necessary and how will this impact multichannel systems? Whatisthelatencyofthesystem? Whatisthebatterylifetime?
1.11.12 Conclusion Largemultichannelwirelesssystemsdemandexcellentplanning,especiallyin theinitialphase,andgoodtechnicalsupport.Observingalltheabove-mentioned items, perfect operation of a system can be guaranteed, even under difficult conditions.
156
PART | I
Electroacoustic Devices
1.12 MICROPHONE ACCESSORIES 1.12.1 Inline Microphone Processors Theoverallsoundofamicrophonecanoftenbenefitfromsignalprocessing, and most mixers provide some basic equalization as a tool for customizing the sound of the microphone. Digital mixers provide an even greater set of tools, including parametric EQ, compression, gain management, and other automatedfunctions.Dedicatedsignalprocessingforeachmicrophoneina system provides a real advantage for the user, and some manufacturers are offering this sort of custom processing on a per-microphone basis via processors that plug inline with the microphone.These include automatic gain control, automatic feedback control, control for plosives and the proximity effect,andintegratedinfraredgatesthatturnthemicrophoneonandoffbased on the presence of a person near the microphone. These phantom-powered processors allow for targeted solutions to many problems caused by poor microphonetechnique.
1.12.1.1 Sabine Phantom Mic Rider Pro Series 3 Theseries3MicRiderincludesinfraredgatesthatturnthemicrophoneonand offbasedonthepresenceofaperson.Theheat-sensingIRmoduleismounted onthegooseneckorisbuiltinonthehandheldversion.TheIRsensorcanbe adjustedforbothtimetoturnoffof5–15sandfordistancesof3–9ft,Figs.1.156 and1.157.
Sabine Mic Rider
FIGURE 1.156 TheSabineInlineMicRider.CourtesySabine,Inc.
Section | 1
Microphones
157
FIGURE 1.157 TheSabinegooseneckMicRiderwithbuilt-inIRsensor.CourtesySabine,Inc.
1.12.1.2 Sabine Phantom Mic Rider Pro Series 2 Theseries2MicRiderincludestheadjustableIRGateplusthreeaudioprocessors:automaticgaincontrol,proximityeffectcontrolforcontrollingincreased bass due to the proximity effect, and plosive control for reducing pops and burstsfromcertainconsonantsoundsinspeech. 1.12.1.3 Sabine Phantom Mic Rider Series 1 The series 1 Mic Rider includes Sabine’s patented FBX Feedback Exterminator for maximizing gain before feedback plus the automatic gain control, proximityeffectcontrol,andplosivecontrol.AnonadjustableIRgateisalso included. The Phantom Mic Rider works with 48Vdc phantom power sources that conformtoindustrystandards(DINstandard45596orIECstandard268–15A). Devicesthatdonotconformcanbemodifiedtomeetthestandard,orexternal phantompowersuppliescanbeused. 1.12.1.4 Lectrosonics UH400A Plug-On Transmitter Thedesignofthistransmitterwasintroducedin1988,inaVHFversionaimed at broadcast ENG applications at a time when production crews were being downsized.Convertingthepopulardynamicmicrophonesofthedaytowirelessoperationeliminatedthecable,whichwasveryusefulforthetwo-person production crews that had evolved. During the 20 years that followed, the design continued to evolve to address an ever-increasing variety of applications.Theadditionofselectablebiasvoltageallowedthetransmittertopower electretmicrophones.ThemovetoUHFfrequenciesandadual-bandcompandorincreasedoperatingrangeandaudioquality.Modificationstothedesign continued through the present day, leading to the current DSP-based model availableintwoversionsforusewithalltypesofmicrophonesandmodestline levelsignalsources.
158
PART | I
Electroacoustic Devices
TheUH400Amodelhasa12dB/octavelow-frequencyroll-offdown3dB at70Hz.TheUH400TMmodeloffersanextendedlowfrequencyresponse down 3dB at 35Hz, Fig.1.158. Figure 1.159 is the block diagram of the transmitter. Themostcommonapplicationsofthistransmitterareeliminatingthecable between a microphone and a sound or recording system.A prime example
FIGURE 1.158 LectrosonicsUH400Aplug-ontransmitter.CourtesyLectrosonics,Inc.
FIGURE 1.159 BlockdiagramoftheLectrosonicsUH400Atransmitter.CourtesyLectrosonics,Inc.
Section | 1
Microphones
159
is acoustic analysis in a large auditorium or stadium where measurements must be made at multiple locations around the sound system coverage area and extremely long cable runs are not practical. In this case, the wireless not only speeds up the process of making measurements, but it also allows more measurements to be taken, which can improve the final sound system performance. Digital Hybrid Wireless™ is a patented process that combines a 24 bit digitalaudiostreamwithwidedeviationFM(U.S.Patent7,225,135).Theprocesseliminatesacompandortoincreaseaudioqualityandexpandtheapplicationstotestandmeasurementandmusicalinstrumentapplications. Audioissampledat88.2kHzandconvertedtoa24bitdigitalstream.The DSP applies an encoding algorithm that creates what might be likened to an instruction set that is sent to the receiver via an FM carrier.The DSP in the receiverthenappliesaninverseoftheencodingalgorithmandregeneratesthe 24bitdigitalaudiostream. AnadditionalbenefitoftheFMradiolinkistheabilityoftheDSPtoemulateacompandorforcompatibilitywithanalogreceiversfromLectrosonicsand twoothermanufacturers. Inthenativehybridmode,theFMdeviationis±75kHztoprovideawide dynamicrange.Thiswidedeviationcombinedwith100mWofoutputpower providesasignificantimprovementintheaudioSNRandthesuppressionofRF noiseandinterference. Usedwithamicrophone,theantennaisadipoleformedbetweenthetransmitterhousingandthemicrophonebody.Whenpluggedintoaconsoleormixer output,thehousingofthetransmitterissimilartotheradiatorofagroundplane antenna,withtheconsoleormixerchassisfunctioningastheground. Phantompowercanbesetto5,15,or48Vorturnedoff.Thetransmittercan provideupto15mAofcurrentin5and15Vsettings,andupto7mAinthe48V setting,allowingittobeusedwithanytypeofmicrophone,includinghigh-end studiocondensermodels. Thetransmitterisavailableon9differentfrequencyblocksintheUHFband between 470 and 692MHz. Each block provides 256 frequencies in 100kHz steps.
1.12.1.5 MXL Mic Mate™ USB Adapter TheMXLMicMate™,Fig.1.160,isaUSBadapterusedtoconnectamicrophonetoaMacintoshorPCcomputer.Itusesa16bitDeltaSigmaA/Dconverter withTHD + N =0.01%atsamplingratesof44.1and48.0kHzandincludesa three-positionanaloggaincontrol.Itincludesastudio-qualityUSBmicrophone preampwithabalancedlownoiseanaloginput,supplies48Vdcphantompower tothemicrophone,andincludesMXLUSBRecorderSoftwarefortwotrack recording.TherearethreedifferentMicMates,oneforcondensermicrophones, onefordynamicmicrophones,andonefornewslinefeeds,videocameras,etc.
160
PART | I
Electroacoustic Devices
FIGURE 1.160 MXLMicMate™USBadapter.CourtesyMarshallElectronics.
1.12.2 Windscreens and Pop Filters A windscreen is a device placed over the exterior of a microphone for the purposeofreducingtheeffectsofbreathnoiseandwindnoisewhenrecording outofdoorsorwhenpanningorgunningamicrophone.Awindscreen’seffectivenessincreaseswithitssurfaceareaandthesurfacecharacteristics.Bycreatinginnumerableminiatureturbulencesandaveragingthemoveralargearea, thesumapproacheszerodisturbance.Itfollowsthatnogainisderivedfrom placingasmallfoamscreeninsidealargerblimptype,Fig.1.161A,whereas afurrycovercanbring20dBimprovement,Fig.1.161B.Mostmicrophones made today have an integral windscreen/pop filter built in. In very windy conditions,thesemaynotbeenough;therefore,anexternalwindscreenmust beused. With a properly designed windscreen, a reduction of 20–30dB in wind noisecanbeexpected,dependingontheSPLatthetime,windvelocity,and thefrequencyofthesoundpickup.Windscreensmaybeusedwithanytypeof microphone because they vary in their size and shape. Figure 1.162 shows a
A. Blimp-type windscreen.
B. Furry cover to surround the windscreen in A. FIGURE 1.161 Blimp-typewindscreenforaninterferencetubemicrophone.CourtesySennheiser ElectronicCorporation.
Section | 1
161
Microphones
windscreenproducedbyShureemployingaspecialtypeofpolyurethanefoam. Thismaterialhaslittleeffectonthehigh-frequencyresponseofthemicrophone becauseofitsporousnature.Standardstyrofoamisnotsatisfactoryforwindscreenconstructionbecauseofitshomogeneousnature. A cross-sectional view of a windscreen employing a wire frame covered withnyloncrepeformountingona1inchdiametermicrophoneisshownin Fig.1.163.TheeffectivenessofthisscreenasmeasuredbyDr.V.BrüelofBrüel andKjaerisgiveninFig.1.164.
1.12.2.1 Wind Noise Reduction Figures for Rycote Windshielding Devices Rycote has developed its own technique for measuring wind noise that uses realwindandarealtimedifferentialcomparison.Thetechniquecomparesthe behavioroftwomicrophonesunderidenticalconditions,onewithaparticular
FIGURE 1.162 ShureA81Gpolyurethanewindscreen.CourtesyShureIncorporated.
Two layers nylon crepe 2/60.25 meshes PSI
Spherical wire frame 120-mm radius B&K–UA 0082 B&K model 4131 microphone (1 inch diameter)
FIGURE 1.163 Review.
Typicalsilk-coveredwindscreenandmicrophone.CourtesyBandKTechnical
162
PART | I
Electroacoustic Devices
120 km/h 75 mi/h
Decibels
Decibels
120 km/h 75 mi/h
30 km/h 19 mi/h
30 km/h 19 mi/h 1/3 Octaves B. Wind noise as a function of frequency with the wind direction at right angles to the membrane.
1/3 Octaves A. Wind noise as a function of frequency with the wind direction parallel to the capsule.
With windscreen dia. � 120 mm (4.7 in) D � 240 mm (9.5 in)
Decibels
Decibels
Without windscreen
(A) (B)
D � 480 mm (19 in) (C) (D) km/h C. Wind noise measured with different sizes of windscreens.
(A) 4131 with standard protection grid when parallel to diaphragm (B) As (A) but with wind at right angle to diaphragm (C) As (B) with windscreen (D) As (A) with windscreen
km/h D. Wind noise as a function of wind speed in the range 20 Hz to 20 kHz.
FIGURE 1.164 The effectiveness of the windscreen shown in Fig.1.163. Courtesy B and K TechnicalReview.
windnoisereductiondevicefittedandtheotherwithout,andproducesastatisticalcurveoftheresultcorrectedforresponseandgainvariations. Figure1.165 is a Sennheiser MKH60 microphone—a representative short rifle microphone—without any low-frequency attenuation in a wideband (20Hz–20kHz)testrig. Whenawindnoisereductiondeviceisfitted,itseffectontheaudioresponse isaconstantfactor—ifitcausessomelossofhigh-frequency,itwilldoitat all times. However, the amount it reduces wind noise depends on how hard the wind is blowing. If there is a flat calm it will have no beneficial effect and the result will be a degradation of the audio performance of the microphone.However,inastronggaleasmalldeviationfromaperfectflatresponse maybeinsignificantfora>30dBreductioninwindnoise.Windnoiseisinthe low-frequencyspectrum.ForanakedSennheiserMKH60,thewind-produced energyisalmostentirelybelow800Hz,risingtoapeakof40dBatabout45Hz. Itistheeffectofashieldattheselowerfrequenciesthatismostimportant.Cavitywindshieldsinevitablyproduceaslightdecreaseinlow-frequencyresponse indirectionalmicrophonesbutthisisnotusuallynoticeable.Baskettypeshave
Section | 1
163
Microphones
Attenuation–dB
Windshield
A B Windshield and Windjammer
Softy C
FIGURE 1.165 WindnoisereductionoptionsforaSennheiserMKH60microphoneunderreal windconditions.CourtesyRycoteMicrophoneWindshieldsLTD.
verylittleeffectonhigh-frequency.Furcoverings,whilehavingamajoreffect inreducinglow-frequencynoise,willalsoattenuatesomehigh-frequency. Addingthelow-frequencyattenuationavailableonmanymicrophonesormixers(whichisusuallynecessarytopreventinfrasonicoverloadandhandlingnoise whenhandholdingorboomingamicrophone)maygiveextrawindnoisereductionimprovementsof>10dBatthecostofsomelow-frequencysignalloss. Thestandard(basket)windshieldshowsupto25dBwindnoiseattenuation at35Hzwhilegivingalmostnosignalattenuation,Fig.1.165. The Softy Windshield is a slip-on open cell foam with an integral fitted furcover.TheSoftyreduceswindnoiseandprotectsthemicrophone.Itisthe standard worldwide in TV. A Softy attenuates the wind noise about 24dB, Fig.1.165B. AddingaWindjammer(furrycover)tothebasketwindshieldwillgivean improvement of about 10dB at low-frequency to −35dB, Fig.1.165C. The attenuation of the Windjammer is approximately 5dB at frequencies above 6kHzalthoughthiswillincreaseifitisdamporthefurisallowedtogetmatted. Overallthiscombinationgivesthebestperformanceofwidebandwindnoise reduction against signal attenuation.To determine the correct windscreen for microphonesofvariousmanufacturers,gotowww.microphone-data.com. Popprotectionisbestappreciatedwhenclose-talkingandexplosivebreath sounds are particularly bothersome. These explosive breath sounds are commonly produced when saying words involving P and T sounds. The phrase explosive breath soundissomewhatofamisnomersincethesesounds,without amplification,arenormallyinaudibletoalistener.15 Theelectricaloutputfromthemicrophoneisactuallythetransientmicrophoneresponsetothislow-velocity,high-pressure,pulse-typewavefront.The PandTsoundsareprojectedindifferentdirectionsandcanbeshownbysaying thelettersPandTwhileholdingyourhandabout3inches(7.6cm)infrontof yourmouth.NotethattheTsoundisfeltataconsiderabledistancebelowthe Psound.
164
PART | I
Electroacoustic Devices
Formostmicrophones,popoutputvarieswithdistancebetweenthesource and microphone, reaching a peak at about 3 inches (7.6cm).Also the worst angleofincidenceformostmicrophonesisabout45°tothemicrophoneand foraglancingcontactjustattheedgeofthemicrophonealongapathparallel tothelongitudinalaxis. sE Dual Pro Pop Filter. AninterestingpopfilterisshowninFig.1.166.The sEDualProPoppopscreenisatwo-filterdevicetosuitvocalperformances. Thedevicehasastronggooseneckwithbothastandardfabricmembraneanda prometalpopshieldonahingemechanism.Theycanbeusedseparatelyorboth simultaneouslydependingontheapplication. In an emergency, pop filters can be as simple as two wire-mesh screens treatedwithflockingmaterialtocreateanacousticresistance.
FIGURE 1.166 sEDualProPoppopscreen.CourtesysEElectronics.
1.12.2.2 Reflexion Filter TheReflexionFilterbysEElectonicsisusedtoisolateamicrophonefromroom noiseshittingitfromunwanteddirections,Fig.1.167. Thereflexionfilterhassixmainlayers.Thefirstlayerispunchedaluminum, whichdiffusesthesoundwavesastheypassthroughittoalayerofabsorptive wool.Thesoundwavesnexthitalayerofaluminumfoil,whichhelpsdissipate energyandbreakupthelowerfrequencywaveforms.Fromtheretheyhitanair spacekeptopenbyrodspassingthroughthevariouslayers. Next the waves hit an air space that acts as an acoustic barrier. The sound wavespasstoanotherlayerofwoolandthenthroughanouter,punched,aluminum wallthatfurtherservestoabsorbandthendiffusetheremainingacousticenergy. Thevariouslayersbothabsorbanddiffusethesoundwaveshittingthem, soprogressivelylessoftheoriginalsourceacousticenergypassesthrougheach
Section | 1
Microphones
165
FIGURE 1.167 ReflexionFilter.CourtesysEElectronics.
layer, reducing the amount of energy hitting surfaces so less of the original sourceisreflectedbackasunwantedroomambiencetothemicrophone.The ReflexionFilteralsoreducesreflectedsoundfromreachingthebackandsides ofthemicrophone.Thesystemonlychangesthemicrophoneoutputbyamaximumof1dB,mostlybelow500Hz. Thestandassemblycomprisesamicstandclampfitting,whichattachesto boththeReflexionFilterandanystandardfittingshockmount.
1.12.3 Shock Mounts Shock mountsareusedtoeliminatenoisefrombeingtransmittedtothemicrophone,usuallyfromthefloorortable.
166
PART | I
Electroacoustic Devices
Microphones are very much like an accelerometer in detecting vibrations hittingthemicrophonecase.Shockmountsuspensionsallowamicrophoneto staystillwhilethesupportmoves. Suspensionsalluseaspringyarrangementthatallowsthemicrophonetobe displacedandthenexertsarestoringforcetoreturnittotherestpoint.Itwill inevitablyovershootandbouncearound,butthesystemshouldbedampedto minimizethis. Asfrequencylowers,thedisplacementwavelengthincreasessothesuspensionhastomovefarthertodothejob.Foranyparticularmassofmicrophoneand compliance(wobbliness)ofsuspension,thereisafrequencyatwhichresonance occurs.Atthispointthesuspensionamplifiesmovementratherthansuppressesit. Thesystemstartstoisolateproperlyataboutthreetimestheresonantfrequency. ThemicrophonediaphragmisthemostsensitivealongtheZ-axistodisturbances.ThereforetheidealsuspensionsaremostcompliantalongtheZ-axis, butshouldgivefirmercontrolonthehorizontal(X)andvertical(Y)axestostop themicsloppingaround,Fig.1.170. Suspension Compliance. Diaphragm and so-called donut suspensions can workwell,buttendtohaveacousticallysolidstructuresthataffectthemicrophone’s polar response. Silicone rubber bands, shock-cord cat’s cradles, and metal springs are thinner and more acoustically transparent but struggle to maintainalowtension,whichcreatesalowresonantfrequency,whileatthe same time providing good X-Y control and reliable damping.The restraining force also rises very steeply with displacement, which limits low-frequency performance. ShockmountsmaybethetypeshowninFig.1.168.Thismicrophoneshock mount,aShureA53M,mountsonastandard 5⁄8inch–27threadandreduces
FIGURE 1.168 ShureA53Mshockmount.CourtesyShureIncorporated.
Section | 1
167
Microphones
mechanicalandvibrationnoisesbymorethan20dB.Becauseofitsdesign,this shockmountcanbeusedonafloorortablestand,hungfromaboom,orused asalow-profilestandtoplacethemicrophonecartridgeclosetoasurfacesuch asafloor.TheshockmountinFig.1.169isdesignedtobeusedwiththeShure A89SMshotgunmicrophone. Shock mounts are designed to resonate at a frequency at least 21⁄2 times lowerthanthelowestfrequencyofthemicrophone.21Thegoalissimplebut therearepracticallimitations.Theresonantfrequency(fn)ofamechanicalsystemcanbecomputedfrom fn =
1 2π
Kg w
(1.32)
where, Kisthespringrateoftheisolator, gistheaccelerationduetogravity, wistheload. A microphone shock-mount load is almost completely determined by the weightofthemicrophone.Toobtainalow-resonantfrequency,thespringrate orstiffnessmustbeaslowaspossible;however,itmustbeabletosupportthe microphonewithouttoomuchsagandbeeffectiveinanypositionthemicrophonemaybeused. TheRycotelyrewebsrelyprimarilyontheirshapetogivedifferentperformance on each axis.Typically, a 100g force will barely move a microphone 1mmalongthe(upanddown)Y-axis,whereasitwillmoveaboutfourtimes that on the (sideways) X-axis. In the critical Z-axis, it will move almost ten timesasfar,Fig.1.170.
FIGURE 1.169 Incorporated.
Shure A89SM shock mount for a shotgun microphone. Courtesy Shure
168
PART | I
Electroacoustic Devices
FIGURE 1.170 A Rycote lyre-type microphone suspension (shock mount) system. Courtesy RycoteMicrophoneWindshieldsLTD.
Withaverylowinherenttensiontheresonantfrequencycanbeverylow, and the Z displacement can be vast. Even with small-mass compact microphones,aresonanceof 2 jω 2π R e m ´r S ω W SS 1 + Q ω − ω 2 WW t 0 0 T X
(2.23)
where, Emistheamplitudeoftheappliedvoltage, Reisthedcresistanceofthevoicecoil, Bisthefluxdensityinthemagnetgap, listhelengthofthevoicecoilconductorinthegap,and risthedistancefromthesourcetotheobservationpoint. Qtisgivenby Qt =
k ´m ´ β2 l2 Rm = Re
(2.24)
where, Rmisthewoofermechanicalresistance(damping). Notetheseparationoftherightsideoftheequationforpressureresponseintotwo parts.Thefirstcontainsamplitudeinformationresultingfromthedrivingvoltage, wooferparameters,anddistancefromthesource,andthesecondprovidesfrequency responseinformation.AvoltageexcitationoftheformofE = Eme jωtisassumed.
Section | 2
267
Loudspeakers
Fromthefirsttermintheequation,wecanseeseveralwaysinwhichthe system’soutputcanbeincreasedforagivendistanceanddrivingvoltage: 1. Increasethefluxdensity,B.Increasingmagnetsizewillaccomplishthisup tothepointatwhichthepolepieceissaturated. 2. Increasethelengthloftheconductorinthegap.ThiswillincreaseRe,however,ifallwedoistoaddturnstothevoicecoil. 3. Increasethediaphragmsurfacearea,Sd.Doingsowithoutchangingthedensityofthematerialwillalsoincreasem,however. Changinganyoftheabovewillpotentiallyhaveaneffectonthevalueof Qt.IfthetotalsystemQhasavalueof0.707,theresponseofthesystemwillbe maximallyflat,alsoknownasaButterworthalignment.IftheQishigherthan this,therewillbeapeakintheresponsejustabovethecutofffrequency. IftotalQislowerthan0.707,theresponsewillfalloff,orsag,intheregion abovecutoff,Fig.2.59. 20 QT 510
Relative response– dB
10 0 210
4 2 1 0.5 0.2
220
0.1
230 240 0.1 0.2 0.4 0.6 0.1 0.2 4.0 6.0 10 f/f0 where f0 is the system resonance
FIGURE 2.59 Response of closed-box system versus Q and normalized frequency relative to systemresonance.
2.10.2.2 Vented Boxes Prior to the existence of analytical models for a vented-box woofer, it was understoodthatanopeningcouldbecutinalow-frequencyenclosure,creating aHelmholtzresonance.Theventitselffunctionsasanadditionalradiatorinthis case,anditsradiationcanaddconstructivelytothatofthewooferoveralimited rangeoffrequencies.A.N.Thieledevelopedtheoriginalpublishedanalytical modelforventedboxradiators,andhisworkwaslatersupplementedbythatof RichardSmall. Theeffectoftheenclosureonthespringconstantofthewooferisthesame asinasealedenclosure.Theventfunctionsasapassiveradiatorcoupledtothe
268
PART | I
Electroacoustic Devices
wooferconeviatheairintheenclosure.Inmodelingtheresponseofavented enclosurewoofer,wemustaccountforthemotion,andthereforetheacoustic radiation,ofthevent.Theairintheventisassumedtomoveasaunittoallow themathematicstoremainmanageable.Thefollowingexpressiongivesthefarfieldhalf-spaceacousticpressurefromaventedenclosureatlowfrequencies: R V ω 4 S W d n ω0 S W Em βlρ0 Sd H´S W p => 2 2 +ω 2 2π rRe m´d r 3 S ω ω j ω v v d ω ω W 1 ω 1 S 4 + dω n − e 2 o + Q f 2 p d ω n WW 2 S ω0 Qt 0 ω0 ω0 0 t ω0 T X (2.25) where, ωis2πf, ω0 is
(ωd ωv ),
ωv is
(k v /m v ) .
Thefirstportionoftherightsideisidenticaltoitscounterpartinthesealedboxequation.Thesecondpartdescribesafourth-orderhigh-passfilter.There arethreegeneralalignmentclassesforsuchfilters:Butterworth,ormaximally flat;Chebychev,orpeaked;andBessel,ormaximallyflatgroupdelay. Acomparisonoftheattributesofsealedandventedenclosuresisinorder.The sealedsystemhastheadvantageofanintrinsicexcursion-limitingmechanism— theadditiontothewoofer’sspringconstantduetotheairinthechamber—for frequenciesbelowthesystemcutoff.Theventedsystem,ontheotherhand,can allowexcessivewooferexcursionifexcitedwithout-of-bandsignals,soanelectricalhigh-passfilterisadesirableprotectiveelement.Thehigher-ordernature oftheventedsystemrendersitmoresusceptibletomisalignmentscausedbyproductionvariationsinwooferparametersandchangesinatmosphericconditions, butithastheadvantageofrequiringlesswooferexcursionforagivenacoustic outputinthelowestportionofitsusablebandwidth.Ingeneral,asealedsystem willhavemorehighlydampedlow-frequencytransientresponsewhencompared toaventedsystemwiththesamecutofffrequency.Thiseffectismostnoticeable forfrequenciesintheneighborhoodofthesystemlowercutofffrequency. Thedesignandmodelingofventedandsealedwooferenclosureshasbeen greatlysimplifiedinrecentyearsduetothereadyavailabilityofcomputersoftware developed specifically for the purpose. The following Thiele-Small (in honorofA.N.ThieleandRichardSmall)loudspeakerparametersarerequired asminimuminputtoanenclosuredesignprogram: 1. QTisthetotalloudspeakerQ. 2. FSisthefree-airconeresonanceoftheloudspeaker.
Section | 2
269
Loudspeakers
3. VASistheequivalentvolumecomplianceofthesuspension. 4. Xmaxisthemaximumlinearexcursion. 5. Pmaxisthemaximumthermalpowerhandling. Most manufacturers provide the above parameters as perAudio Engineering Society(AES)recommendedpracticeonloudspeakerspecifications. Loudspeakerenclosurescanhaveresonances,orstandingwaves,causedby internalreflections.Thecharacteristicfrequencies(fb)ofthesestandingwaves aregivenby: c fn = 2
f
nx lx
2
p +f
ny ly
2
p +f
nz lz
2
p
(2.26)
where, x, y,andzarethethreeboxdimensions, listhedesignateddimensionofthebox, ntakesonallpossibleintegervalues(n =0,1,2,3,…). Ifthelowestmodalfrequencyfoundbysettingn =0forallbutthelongest boxdimensionisabovethebandofuse,therewillbenostandingwavesinside theenclosure.Inthemorecommoncaseofawooferbeingusedabovethefirst modefrequency,acousticabsorptioncanbeaddedtodamptheseunwantedresonances.Itshouldbeunderstoodthattheadditionoflargeamountsofdamping materialcanhaveanadverseeffectonthebox/porttuning,soabalancemust oftenbestruckbetweencontrolofstandingwavesandoptimallow-frequency responsealignment.
2.10.2.3 Measurement of Thiele-Small Parameters ThemostaccuratewayofdeterminingfSisbyobservationofaLissajouspattern onanoscilloscopeofvoltageversuscurrentwiththespeakerinfreeair.When thepatterncollapsestoastraightdiagonalline,thephaseoftheimpedanceis zero,whichindicatesresonance.Oncethisconditionhasbeenachieved,thefrequencyshouldbemeasuredwithafrequencycounterthatisaccurateto0.1Hz orbetter. VAS may be determined with the loudspeaker suspended in free air as follows: 1. Find total moving mass (m) by attaching an extra mass (MX) to the cone (asclosetothevoicecoilaspossible)andobservingthenewresonantfrequency,fSX.MXcanbeputtyorclay;measuremXaccurately.Thetotalmass canthenbefoundbytheequation ml=
Mx
f
fS fSX
2
p −1
(2.27)
270
PART | I
Electroacoustic Devices
2. Thesuspensionspringconstantisthen k = _ 2π f s i 2 m ´
(2.28)
3. Fromtheeffectivediaphragmareaoftheloudspeaker,SD,VASisgivenby VAS =
ρ0 c 2 SD 2 k
(2.29)
One means for determining the effective diaphragm area is to excite the woofernearitsresonantfrequencyandobserveitsmotionusingastrobelight tuned almost, but not exactly, to the frequency of excitation. The resulting apparentslowmotionofthewooferwillallowanaccuratedeterminationofthe portionoftheconethatismoving. QTmaybefoundusingthefollowingprocedure: 1. Determinetheimpedanceofthewoofer,Zmax,atitsresonantfrequency.This issimplytheappliedvoltagedividedbythecurrentinthecoil. 2. Identifythefrequencies,f+andf−,aboveandbelowfs,respectively,atwhich themagnitudeoftheappliedvoltagedividedbythemagnitudeofthecurrent inthecoilisequalto. 3. MechanicalQisgivenby Qm = f
f0 f+ − f−
p
Z max Re
(2.30)
4. The total Q is then Q1 = Qm f
Re Z max
p
(2.31)
2.10.3 Horns Although there are numerous mathematical treatments of horns in the texts on acoustics, they all suffer from a common set of inadequacies: the models developedintheliteratureaccountforenergytransmissioninsidethehorn,but therearenoclosed-formsolutionstotheproblemofhorndirectivity—i.e.,the behaviorofahorn’sradiationoutsidetheboundariesofthehornwalls,where listenersarelocated.Modernhorndesignershavebeenfarlessconcernedwith optimizingacousticloadingthanwithcreatingdesirabledirectivitycharacteristics,andthedesignshavewithoutexceptionbeenderivedempiricallyrather thananalytically. Inanexponentialhorn,thecross-sectionalareaisgivenby S = S0 em x
(2.32)
Section | 2
271
Loudspeakers
where, S0isthecross-sectionalareaatthehorn’sthroat,orentry, mxiscalledtheflareconstant. Theradiationimpedanceofanexponentialhorn,assumedtobeinfinitely longforourpurposes,is Zr =
ρ0c S0
=
1−
m 2 c2 mc G . +j 2 2ω 4ω
(2.33)
Thefirstterminthebracketsistheradiationresistance,andthesecondisthe radiationreactance.Ofinterestisthefrequencyatwhichthevalueofexpression insidethesquarerootbecomeszero ωc =
mc 2
(2.34)
fc =
mc 4π
(2.35)
or
Thisisknownasthehorn cutoff frequency.Theabovetheorypredictsthat no sound will be transmitted through the horn below this frequency. Clearly, thisisnotthecasewithrealhorns,sothetheorycontainsoneormoreassumptionsthatarenotmetinpractice.Notealsothatthesecondterminthebrackets, theradiationreactance,goestozerointhehigh-frequencylimit.
2.11 LOUDSPEAKER TESTING AND MEASUREMENT Aswithmostotherdevicesthattransmitorprocessasignalcontaininginformation,measurementtechniqueshavebeendevelopedspecificallyforthetesting andevaluationofloudspeakers.Beforetheearly1980s,accurate,comprehensive testing of loudspeakers generally required expensive anechoic chambers or large outdoor spaces. Since that time, the advent of computer-based time- windowedmeasurementshasrevolutionizedthefieldofacousticinstrumentation,particularlyasregardsthetestingofloudspeakers.
2.11.1 Linear Transfer Function Oneobjectiveintestingaloudspeakeristodeterminethelinearportionofitscharacteristictransferfunction(or,equivalently,impulseresponse).Themostcommon meansforacquiringthisdataisaspectrumanalyzer.Aspectrumanalyzerappliesa signalwithknownspectralcontenttotheinputofasystemandprocessesthesignal thatappearsattheoutputofthedevicetoacquirethesystem’stransferfunction.
272
PART | I
Electroacoustic Devices
2.11.1.1 Spectrum Analysis Concepts Allspectrumanalysistechniquesaresubjecttoasetofgeneralconstraints imposedbythemathematicalrelationshipbetweentimeandfrequency.Itis usefultohaveafeelingfortheseconstraintswhengatheringorevaluating loudspeakerdata.Timeandfrequencyarethemathematicalinversesofeach other.Asignalthathasonlyonefrequencymustexistforalltimeand,conversely,asignalthatexistsforafiniteamountoftimemustcontainmultiple frequencies. A signal that exists only within a known time interval—i.e., at all times before time t0 the value of the signal is zero and at all times after time t1 the value is zero—can only contain frequencies given by the expression: f=N
1 ( t 1 − t 0)
(2.36)
where, Nisaninteger. The frequency corresponding to N = 1 gives the best (lowest) frequency resolution that is possible in a test conducted for that precise time interval. Allotherfrequencieswillbeintegermultiplesofthisfrequency.Inorderto have infinitesimally small frequency resolution (i.e., perfectly resolved frequencydata),atestwouldhavetobeconductedforaninfiniteamountoftime. It follows that all realizable response tests have a limit on their frequency resolution. Theeffectoffrequencyresolutiononatransferfunctionmeasurementisto smooththeappearanceofaplotoftheresults,therebypossiblyobscuringsome ofthedetailsofthetransferfunction.Thissmoothingispresenttosomedegree inalltransferfunctionmeasurements.Inthecaseofelectronicdevices,transfer functions are typically well behaved enough that the frequency resolution of a response test does not cause meaningful loss of detail.With loudspeakers, theoppositeisoftentrue:aloudspeaker’stransferfunctionoftenhassomuch finestructurethatapracticaltestwillnoticeablysmoothoutthepeaksanddips in the speaker’s response. The degree to which this fine structure is audibly significantisamatterofsomecontroversy.Asaresult,thereisnowidespread agreement in the industry on the minimum desirable frequency resolution in testingloudspeakers.
2.11.2 Chart Recorders Prior to the advent of computer-based measurement systems, the most commonlyemployedloudspeakermeasurementinstrumentationcomprisedastrip chartrecorderandasignalsweepgenerator.Thetwodevicesaresynchronized suchthat,foragivenfrequencyinthesweep,thepenontherecorderisinthe
Section | 2
Loudspeakers
273
appropriatex(frequency)positiononpreprintedgraphpaper.Thepen’syposition would correspond to the amplitude of the signal received from the test microphone,andtherefore,hopefully,totheamplituderesponseofthespeaker atthatfrequency. If the y amplifier is logarithmic, then the amplitude will be expressed in decibels.Ascommonasthestrip-chartmeasurementtechniquewaspriortothe 1980s,ithadseveralprominentdisadvantages: 1. Thereisnomeansofmeasuringaloudspeaker’sphaseresponsewiththis technique. 2. Themeasurementisincapableofdiscriminatingbetweendirectsoundfrom the device under test and sound that is reflected from surfaces in the test environment. This necessitated the construction of very costly anechoic chambers.Eveninsuchachamber,theinclusionofsomereflectedsoundin astrip-charttypemeasurementisunavoidable. 3. The measurement technique does not isolate the linear portion of a loudspeaker’s transfer function. Distortion products are simply added to the amplitude of the loudspeaker’s transfer function at the fundamental frequencythatexcitesthem. 4. Thereisnodirect,accuratewaytodetermineorcontrolthefrequencyresolutionofastripchartmeasurement.Reducingpenspeedand/orincreasing chart(andsweep)speedhavetheeffectofreducingfrequencyresolution, orsmoothing,thedata,butthedegreetowhichthishastakenplaceisnot alwaysapparent. 5. Datafromthisformofmeasurementisonlygeneratedinhardcopyformat. 6. This measurement technique provides no ready means to compensate for propagation delay: the time required for sound to travel from the loudspeakertothetestmicrophone. 7. Sincethereisnomeansfordistinguishingbetweentheoutputsignalfrom theloudspeakerandambientnoise,thetestenvironmentmustbequiet.
2.11.3 Real Time Analyzers Althoughinitiallydevelopedtomeasuretheresponseofsoundsystemsintheir operatingenvironments,realtimeanalysishasalsobeenusedtomeasureloudspeakerresponseincontrolledenvironments.Withthistestingtechnique,apink noisesignalisappliedtotheloudspeaker.Pinknoiseisarandomsignalthat contains equal energy for each unit of logarithmic frequency—e.g., for each octaveorfractionthereof.Thesignalfromthetestmicrophoneisappliedtoa seriesofbandpassfiltersofconstantpercent-octavebandwidth,eachofwhich istunedtoadifferentbandcenter,andtheaveragedoutputlevelofeachfilteris displayed,eitheronaCRT,LCD,orLEDdisplay.Thedisplayrepresents,within the limitations due to the measurement technique and the test environment, theamplituderesponseoftheloudspeaker.Becausethefrequencycontentofa
274
PART | I
Electroacoustic Devices
randomsignalhassmallfluctuationsovertime,thedisplaymaybeaveraged,or integrated,toproduceastablegraph.Realtimeanalysissuffersfromthesame generaldisadvantagesasthechartrecordermethodofmeasurement.
2.11.4 Time-Windowed Measurements Thedevelopmentofinexpensivecomputershasliterallyrevolutionizedthefield ofacousticinstrumentation.Thisislargelytheresultofthecomputer’sability toprocessandstorelargeamountsofsignaldata.Withacomputer-basedmeasurementsystem,processinganddisplayofthedatacanbeaccomplishedatany timeaftertherawdatahasbeentaken. The effects of time windowing are present in any signal measurement, sincethemeasurementmustbeinitiatedandcompletedinafiniteamountof time (window). In digital measurement systems, however, the exact size of thetimewindow,andthereforetheresultanttradeoffsinresolution,aremore directly controllable. There are two general approaches to acquiring a loudspeaker’stransferfunctionviatime-windowedmeasurements:measurementof thedevice’simpulseresponse(time-domainmeasurement)oracquisitionofthe device’stransferfunctioninthefrequencydomain.
2.11.4.1 Measurement of the Impulse Response Oneformofinputsignalthatishighlyusefulasanexcitationfortestandanalysispurposesisanimpulse.Mathematically,thesignalisdescribedbyaDirac deltafunction.Descriptively,animpulseisavoltage“spike”ofveryshortduration and relatively large amplitude.An interesting property of an impulse is that it contains all frequencies at the same level. The impulse response of a loudspeaker,viatheFouriertransform(orfastFouriertransform,FFT,asimplementedincomputer-basedinstruments),givesthespeaker’stransferfunction. Itisthisequivalencyviatransformoftheimpulseresponseandtransferfunctionthatallowsustofullycharacterizeatwo-portdevicethroughmeasurement takeninonlyonedomainortheother(timeorfrequency). If a loudspeaker is excited with an impulse, the signal from a suitably well-behavedtestmicrophoneplacedinfrontofthespeakerwillrepresentthe loudspeaker’simpulseresponseatthatpoint.Thissignalmaybedigitizedand postprocessedtoyieldthefrequency-domaintransferfunction,aswellasanumberofotherfunctions.Thesamplingprocesstakesplaceoverafixedamountof time(thetimewindow),anditsinitiationmaybedelayedtoremovetheeffect oftimerequiredforsoundtotravelfromtheloudspeakertothetestmicrophone (propagationdelay).Additionally,thelengthofthetimewindowmaybechosensoastorejectreflectionsfromtheroominwhichthemeasurementisbeing made. This is termed a quasianechoic measurement, and its availability has madeitpossibletoacquireaccuratedirect-fieldresponsedataonloudspeakers withoutananechoicchamber.
Section | 2
Loudspeakers
275
ThemathematicsofFourierseriesrequiresthatthesignalvaluebezeroat thebeginningandendofthetimewindow.Sincethisconditionisnotgenerally satisfied, a window function is applied to the sampled data to force the endpointsofthewindowtozero.Theeffectofthewindowfunctionistocreate inaccuraciesinthecalculatedtransferfunction,butthesearegenerallymuch less than the spectral inaccuracies that would result from unwindowed (truncated)data.Varioustypesofwindowfunctionsmaybeused,includingsquare (equivalenttounwindoweddata),Gaussian,Hamming,andHanning.Eachhas itsadvantagesanddrawbacks. AmongthedisadvantagesofimpulseexcitationisthatofSNR.Becausethe impulseisofshortduration,quiescentnoiseinthetestenvironmentcaneasily corruptthetestdata.Onemeansofreducingtheeffectofbackgroundnoiseis to average the results of multiple tests. If the noise is random, and therefore uncorrelatedwiththetestsignal,eachdoublingofthenumberofaverageshas thepotentialofreducingtherelativenoiselevelby3dB.Averagingincreases theamountoftimerequiredtoacquirethedata. Anotherdisadvantageofimpulseexcitationisthatitprovidesnomeansofidentifyingnonlinearities(distortion)intheloudspeaker.Distortionproductsappear nodifferentlytotheanalyzerthanthelinearportionofthedevice’sresponse.
2.11.4.2 Maximum Length Sequence Measurements A variation on the method of impulse excitation is called maximum length sequence, or MLS, testing. In this form of measurement, the excitation signalisaseriesofpulsesthatrepeatsitself.Theloudspeaker’simpulseresponse is derived by calculating the cross-correlation between the input and output signals.Thisexcitationsignalhastheadvantageofproducinghigheraverage signallevelsthananimpulse,thereforeimprovingthesignal/noiseperformance ofthetest. Additionally,theeffectsofcertaintypesofdistortioncanbereducedsubstantiallybyrunningaseriesoftestsemployingastrategicallychosensetofsignalsequencelengths.Tobeaccurate,anMLStestmustbeconfiguredsuchthat thedurationofthesignalsequenceexceedsthelengthoftheimpulseresponse ofthedeviceundertest.Inthecaseofloudspeakermeasurements,theimpulse responseoftheacousticenvironmentmustbeaccountedforinordertosatisfy therequirement. 2.11.4.3 Other FFT-Based Measurements YetanothervariationontheFFTtechniqueisatypeofmeasurementthatcan useanarbitrarysignalastheexcitationsignal.Thisformofmeasurementisa dual-channelFFTmeasurement,andthevariationsonthistechniquehaveseveralcommonelements.Thetechniqueinvolvessamplingthesignalatapointin thechainpriortotheinputoftheloudspeaker(input),aswellasthesignalfrom atestmicrophone(output).Theoutputmaybesampledatalatertimeinorder
276
PART | I
Electroacoustic Devices
toaccountforthetimerequiredforsoundtopropagatefromtheloudspeakerto thetestmicrophone.AswithMLStesting,cross-correlationbetweeninputand outputsignalswillyieldtheimpulseresponseoftheloudspeaker.ItisalsopossibletoperformanFFTonbothinputandoutputsignalsandobtainthetransfer functionoftheloudspeakerbycomplexdivision. Thedual-channelapproachhastheadvantageofallowingawiderangeof signals,includingmusic,tobeemployedasexcitation.Thecommerciallyavailableimplementationsofthistechniqueincorporateseveralrefinementsofthe basicproceduredescribedabove,andthesesystemsofferthepossibilityofmeasuringtheresponseofasoundsystemwhileitisinoperation. SNR is a possible issue with this form of measurement, so averaging is generally performed to improve the accuracy of the results.Additionally, the spectrumoftheinputsignalmaynotcontainsufficientenergyatallfrequencies tosufficientlyexcitethesystemundertest.Forthisreason,acoherencefunction isusedtoindicatethosefrequencyrangeswherethesignalenergyisinsufficient toyieldgoodresults.
2.11.5 Swept Sine Measurements Althoughthechartrecorderisasweptsinemeasurement,itfailstotakeadvantageofallthepossibilitiesofferedbytheuseofasweep(alsoknownasachirp) asatestsignal.DickHeyserdevelopedandpatentedatechniqueknownastime delayspectrometry,orTDS.InTDS,theanalyzer’sreceivingcircuitryemploys abandpassfilter,thecenterfrequencyofwhichissweptinsynchronicitywith thefrequencyofthesignalappliedtotheloudspeaker.Adelaymaybeapplied tothesweepofthebandpassfiltertoaccountfortheamountoftimerequired forsoundtopropagatefromthedeviceundertesttothemicrophone,hencethe nameofthetechnique. Thebandpassfilterwillrejectfrequenciesthataredisplacedbysomeamount from its center frequency. If the receive delay for the filter is chosen appropriately,theanalyzerwilladmitthedirectsignalfromtheloudspeaker,while simultaneously rejecting signals that have been reflected from environmental surfaces,therebytravelingalongerpathandarrivinglaterthanthedirectsignal. Theeffectofthisabilitytorejectunwantedreflectionsisthecreationofatime window,eventhoughthedataistakeninthefrequencydomain.Additionally, thebandpassfilterattenuatesbroadbandnoisebyamuchgreateramountthanit doesthedirectsignalfromtheloudspeaker. DuetotheinherentlyhighSNRofTDS,averagingofmultipletestsisusually unnecessary.Thenumberofsamplesanalyzedisalsonotafunctionofthetime window,asitiswithanFFT-basedanalyzer.Furthermore,thebandpassfilter removesdistortionproducts,soTDSisintrinsicallymorecapableofseparating thelineartransferfunctionfromdistortionproducts.Itisalsopossibletousethe techniquetotrackspecificharmonicswhilerejectingthefundamental.
Section | 2
Loudspeakers
277
Loudspeakertestinstrumentationismorepowerfulandlessexpensivenow thanatanyothertimeinthehistoryofloudspeakers.Whilethecurrentsituation makesitpossibletogatherevermoredetailedinformationaboutthebehaviorof loudspeakers,itisimportanttokeepinmindthebasicsofinstrumentationand spectrum analysis.This awareness will assist in identifying loudspeaker data thatissuspectorincomplete.
REFERENCE 1. R.H.Small,“Constant-VoltageCrossoverNetworkDesign,”Journal of the Audio Engineering Society,Vol.19,no.1,January1971.
BIBLIOGRAPHY A.A. Janszen, “An Electrostatic Loudspeaker Development,” J. Audio Eng. Soc.,Vol. 3, no. 2, April1955. A.H.Benade,Fundamentals of Musical AcousticsOxford,UniversityPress,1976. A.N.Theile,“LoudspeakersinVentedBoxes:PartI&II,”J. Audio Eng. Soc.,Vol.19,no.5,and 6,MayandJune1971. A.Wood,Acoustics,NewYork:DoverPublications,Inc.,1966. AllanD.Pierce,Acoustics An Introduction to Its Physical Principles and Applications,NewYork: McGraw-Hill,1981. B. N. Locanthi, “Application of Electric Circuit Analogies to Loudspeaker Design Problems,” J. Audio Eng. Soc.,Vol.19,no.9,October1971. C. A. Henricksen, “Heat Transfer Mechanisms in Loudspeakers: Analysis Measurement and Design,”J. Audio Eng. Soc.,Vol.35,no.10,October1987. C.A.Henricksen,“PhasePlugModelingandAnalysis:RadialversusCircumferentialTypes,”presentedatthe59thConventionoftheAES,preprint1328(F-5). C.R.HannaandJ.Slepian,“TheFunctionsandDesignofHornsforLoudspeakers,”J. Audio Eng. Soc.,Vol.25,no.9,pp.573–585,September1977. D.B.KeeleJr.,“Low-FrequencyLoudspeakerAssessmentbyNearieldSound-PressureMeasurements,”J. Audio Eng. Soc.,Vol.22,no.3,pp.154–162,April1974. D.B.KeeleJr.,“What’sSoSacredaboutExponentialHorns?”AESpreprint1038(F-3). D.B.KeeleJr.,“Low-FrequencyHornDesignUsingThiele/SmallDriverParameters,”presentedat the57thconventionoftheAES,preprintno.1250(K-7),May1977. D.B.KeeleJr.,“EffectivePerformanceofBesselArrays,”AESpreprint2846(H1)presentedatthe 87thConventionoftheAES,October1989. G.L.Augsburger,“ElectricalversusAcousticalParametersintheDesignofLoudspeakerCrossover Networks,”Journal of the Audio Engineering Society,Vol.19,no.6,June1971. H.F.Olson,Acoustical EngineeringPrinceton:VanNostrand,1976. H.F.Olson,Music,Physics and Engineering,NewYork:DoverPublications. H. Suzuki and J. Tichy, “Diffraction of Sound by a Convex or a Concave Dome in an Ininite Bafle,”Journal of the Acoustical Society of America,preprint0(5),November1981. J.L.Bernstein,Audio Systems,NewYork:JohnWiley&Sons,1966. J.M.Kates,“RadiationfromaDome,”J. Audio Eng. Soc.,Vol.24,no.9,November1976. J.R.AshleyandA.L.Kaminsky,“ActiveandPassiveFiltersasLoudspeakerCrossoverNetworks,” J. Audio Eng. Soc.,Vol.19,no.6,June1971.
278
PART | I
Electroacoustic Devices
J. R. Ashley and M. D. Swam, “Experimental Determination of Low-Frequency Loudspeaker Parameters,”J. Audio Eng. Soc.,Vol.17,no.5,October1969. J.R.Wright,“FundamentalsofDiffraction,”J. Audio Eng. Soc.,May1997. JohnBorwick(Editor),Loudspeaker and Headphone Handbook,London:Butterworths,1988. JohnVanderkooy,“ASimpleTheoryofCabinetEdgeDiffraction,”J. Audio Eng. Soc.,December 1991. K. O. Johnson, “Single-Ended Wide-Range Electrostatic Tweeters with High Eficiency and ImprovedDynamicRange,”J. Audio Eng. Soc.,July1964. L.E.KinslerandA.R.Frey,Fundamentals of Acoustics,NewYork:JohnWiley&Sons,1962. L.L.Beranek,Acoustics,NewYork:McGraw-Hill,1954. N.W.McLachlan,Loudspeakers; Theory, Performances, Testing, and Design,CorrectedEdition, NewYork:DoverPublications,1960. P.M.Morse,Vibration and Sound,NewYork:McGraw-Hill,1948. P.W.Klipsch,“ALow-FrequencyHornofSmallDimensions,”Journal of the Acoustical Society of America,October1941. P.W.Klipsch,“ModulationDistortioninLoudspeakers,”J. Audio Eng. Soc.,Vol.17,no.2,April 1969.(SeePartsIIandIII,J. Audio Eng. Soc.,Vol.18,no.1,February1970andVol.20,no.10, December1972.) P.W.Klipsch,“ModulationDistortioninLoudspeakersPartII,”J. Audio Eng. Soc.,Vol.18,no.1, February1970. R.F.Allison,“TheInluenceofRoomBoundariesonLoudspeakerPowerOutput,”J. Audio Eng. Soc.,Vol.22,no.5,June1974. R.G.Brown,Lines Waves and Antennas,NewYork:TheRonaldPressCo.,1961. R.H.Small,“Direct-RadiatorLoudspeakerSystemAnalysis,”J. Audio Eng. Soc.,Vol.20,no.5, pp.383–395,June1972. R. H. Small, “Closed Box Loudspeaker Systems; Part I:Analysis,” J. Audio Eng. Soc.,Vol. 20, no.10,pp.798–808,December1972. R.H.Small,“VentedBoxLoudspeakerSystems;PartI,II,III,andIV,”J. Audio Eng. Soc.,Vol.15, nos.5,6,7,and8;June,July,August,andSeptember1973. R.H.Small,“PhaseandDelayDistortioninMultiple-DriverLoudspeakerSystems,”J. Audio Eng. Soc.,Vol.19,no.1,January1971. R.Heyser,“LoudspeakerPhaseCharacteristicsandTimeDelayDistortionPart1,”J. Audio Eng. Soc.,Vol.17,no.1,p.30,January1969. R.Heyser,“LoudspeakerPhaseCharacteristicsandTimeDelayDistortionPart2,”J. Audio Eng. Soc.,Vol.17,no.2,p.130,April1969. S. H. Linkwitz, “Active Crossover Networks for Non-Coincident Drivers,” J. Audio Eng. Soc., Vol.24,no.1,p.2,January/February1976. S.IshiiandK.Takahashi,“DesignofLinearPhaseMulti-WayLoudspeakerSystem,”AES52nd Convention,October1975,preprint1059.
Section 3
Loudspeaker Cluster Design Ralph Heinz
3.1 Why Array? 3.2 Array Problems and Partial Solutions: A Condensed History 3.3 Conventional Array Shortcomings 3.4 Conventional Array Shortcoming Analysis 3.5 Coincident Acoustical Centers: A Practical Approach 3.5.1 TRAP Horns: A New Approach 3.5.2 TRAP Performance 3.6 Low-Frequency Arrays: Beneficial Interference 3.6.1 Horizontal Woofer Arrays: Maintaining Wide Dispersion 3.6.2 Vertical Woofer Arrays
3.7 Line Arrays and Digitally Steerable Loudspeaker Column Arrays 3.7.1 What Affects Intelligibility 3.7.2 Measuring Intelligibility 3.7.3 Architecture and Room Acoustics 3.7.4 Line Arrays 3.7.5 DSP-Driven Vertical Arrays 3.7.6 Multichannel DSP Can Control Array Height 3.7.7 Steerable Arrays May Look Like Columns But They Are Not
3.1 WHY ARRAY? Forthepurposesofthisdiscussionwecandefinealoudspeakerarrayas a group of two or more full-range loudspeaker systems, arranged so their enclosures are in contact.Systemdesignersusearraysofmultipleenclosureswhenasingle enclosurecannotproduceadequatesoundpressurelevels,whenasingleenclosurecannotcovertheentirelisteningarea,orboth.Theseproblemscanalsobe dealtwithbydistributingsingleloudspeakersystemsaroundthelisteningarea, butmostdesignersprefertousearrayswheneverpossiblebecauseitiseasierto maintainintelligibilityusingasoundsourcethatapproximatesapointsource thanbyusingmanywidelyseparatedsources. Electroacoustic Devices: Microphones and Loudspeakers Copyright©2009byFocalPress,Inc.Allrightsofreproductioninanyformreserved. Sections1,2,and3wereoriginallypublishedinHandbookforSoundEngineers,FourthEditionbyGlenBallou.
279
280
PART | I
Electroacoustic Devices
3.2 ARRAY PROBLEMS AND PARTIAL SOLUTIONS: A CONDENSED HISTORY First-generationportablesoundsystemsdesignedformusicusedaveryprimitiveformofarray:theysimplypileduplotsofrectangularfullrangespeaker systemstogether,withallsourcesaimedinthesamedirection,inordertoproduce the desired SPL. This type of array produced substantial interference, becauseeachlistenerheardtheoutputofseveralspeakers,eachatadifferent distance.Thedifferenceinarrivaltimesproducedpeaksandnullsintheacoustic pressurewaveateachlocation,andthesereinforcementsandcancellationsvariedinfrequencydependingonthedistancesinvolved.Soalthoughthesystem producedthedesiredSPL,thefrequencyresponsewasveryinconsistentacross thecoveragearea.Evenwhereadequatehigh-frequencyenergywasavailable, intelligibilitywascompromisedbymultiplearrivalsateachlisteninglocation. Second-generation systems incorporated compression drivers and hornloading techniques derived from cinema sound reinforcement and used for large-scalespeech-onlysystems(theoriginalmeaningofpublicaddress).When twoorthreeofthesehornswereincorporatedinasingleenclosurewithtrapezoidalsidesthatsplayedthehornsawayfromeachother,thefirstarrayablesystemswereintroducedtothemarketplace.Theseproductspromisedtoeliminate lobinganddeadspots(peaksandnulls)andtodrasticallyreducecombfiltering(interference).Theydidimproveperformanceoverthestackofrectangular enclosuresloadedmainlywithdirectradiatingcones.Butfrequencyresponse acrossthecoveragearearemainedinconsistent.Inadditiontothemidrangeand high-frequencyvariationsacrossthecoverageareaofthearray,low-frequency outputvariedfromthefronttotherearandsidetoside.Low-frequencyenergy wasfocusedalongthelongitudinalaxisofthearrayandclosetoit,producinga “poweralley”thatgavetheseatswiththebestviewstheworstsound,Fig.3.1.
3.3 CONVENTIONAL ARRAY SHORTCOMINGS Aswesaidinthefirstparagraph,theperformanceadvantagesofthearray(whether horizontalorvertical)derivefromitsabilitytoapproximateaperfectacoustical pointsource.Buteventhesmallestarraystypicallyincludethreeormoreloudspeakerenclosures,eachwithtwoorthreeseparateacousticcentersofitsown.It’s easytoappreciatethatgettingallthosediscretesourcestobehavelikeatheoretical pointsourceisdifficultinpractice.Signalprocessingsolutionsattempttocompensateforthedifferencebetweentheoryandrealitybysacrificingthecoherency oftheelectronicsignal.Theyapplyfrequencyshadingand/ormicro-delaystothe signalssenttodifferentenclosures,inordertoamelioratetheacousticproblems. Theseapproachesarecostly,complicated,andoftenmeetwithlimitedsuccess. A rigorous analysis of the acoustical physics can point the way toward a practical,physicalsolution.First,considerwhatisprobablythemostcommon arrayablesysteminusetoday:60°×40°hornsinenclosureswith15°trapezoidalsides,Fig.3.2.
Section | 3
Loudspeaker Cluster Design
281
FIGURE 3.1 Atypicalsecond-generationloudspeakercluster.Evenwhenasingleenclosureis designedtoresembleapointsource,multipleenclosureswillalwaysinterferewitheachotherwhen connectedtoacoherentaudiosignal.
Tight-packingthreeofthesesystemswiththeir15°sidestouchingproduces a30°splaybetweenthehorns,foratotalincludedangleof120°.Atfirstglance, this seems like an ideal alignment. But the EASE interference predictions in Fig.3.3showthefamiliarandclearlyaudibleproblemswiththisconfiguration: significant interference above 1kHz, with variations of 8dB–9dB depending ontheangle.Onaxis,thereisabout10dBofgainatfrequenciesbelow1kHz.
282
PART | I
Electroacoustic Devices
FIGURE 3.2 Averycommonarrayusesthree60°×40°hornsinenclosureswith15°trapezoidalsides;tight-packed,thisarrayproducessubstantialoverlapandinterferencebetweenadjacent horns.
FIGURE 3.3 Theinterferencepatternsshownabovewereproducedbytight-packingthreearrayable loudspeakers using 60° × 40° constant directivity horns in enclosures with 15° trapezoidal sides.Whilethisisanimprovementoverapileofdirectradiatingtransducers,itisfarfromthe idealpointsourcearray.
Section | 3
Loudspeaker Cluster Design
283
WheremaximumSPListhemainconsideration,thistypeofarraywilldeliver acceptableperformance.Whenthefront-of-housemixpositioncanbelocated ontheaxisofleftandrightarrays,theycanusuallybetweakedtodeliveracceptablereproductioninthislimitedarea.Otherareasofthehouse,includingthe highrollerseatsupfront,willsuffer. TheinterferencepatternsdisplayedinFig.3.3canbereducedbywidening thesplaybetweencabinetsto30°,asillustratedinFig.3.4.Thisarraywillnot lookasprettyasthefirst,butitdoeshavemuchmoreevenresponseacrossthe coveragearea,Fig.3.5.At2kHzand4kHz,theindividualhornsareclearlydiscernibleintheALS-1predictions.Alsonotethattheseamsbetweenthehorns becomedeeperwithincreasingfrequency.
FIGURE 3.4 Widening the splay between horns reduces interference and widens the coverage angleto180°,butreducesforwardgain.Asalways,energyisconserved.
FIGURE 3.5 ALS-1interferencepredictionsforawidersplayshowreducedinterference,butthe threehornsareclearlyapparentathigherfrequencies.
284
PART | I
Electroacoustic Devices
FIGURE 3.6 Theacousticpressurewaveexpandsasasphere,andmultiplesphericalsectionswill alwaysoverlapunlesstheyoriginatefromacommoncenter.
Figure3.6showswhytherewillalwaysbeinterferencewithconventional hornarrays(whethertheyareenclosedinarrayablecabinetswithtrapezoidal sidesormountedinfreeair).Asthewavefrontsradiatefrompointsoforigin that are separated in space, they will always create some interference at the coverageboundaries.
3.4 CONVENTIONAL ARRAY SHORTCOMING ANALYSIS Foranarrayinfarfield,dependenceonangleis SPL ] θ g = 10 log P0 2 dB
(3.1)
Foradistancetothelisteningareaverymuchlargerthanthearraydimensions,letthesoundpressurePbetherealpartof P ] θ g = Ai ] θ g j _ ωτ − kSi i
(3.2)
where, Pisthesoundpressure, ωistheangularfrequency, Ai(θ) is a function of the angle between the array longitudinal axis and the directionofthedistantlisteningpoint.Itgivestheratioofthesoundpressure duetothesourceasaratioofitson-axisvalueatthesamedistance. FortheithsourceshowninFig.3.7,assumingidenticalsources,thepressure contributionisgivenby: Pi = Ai ] θ g j (ωτ − kSi)
(3.3)
Section | 3
285
Loudspeaker Cluster Design
where, kis2π/λ =2λfc, λisthewavelength, fisthefrequency, cisthespeedofsound, Siisthedistancebywhichthepathlengthfromtheithsourcetothedistantpoint exceedsthedistancefromtheorigintothatpoint. Foranarrayofnsources,thetotalpressurePisgivenby: n
/ Ai qejωτ − kS i =1 = ε jωτ / A (θ) ε jωτkS
P(θ) =
i
(3.4)
i
i =1
Thesquareofthepressureamplitudeisgivenby: 2
P0 2 (θ) = = / Ai θkSi G + 7 Ai (qθ) sin (kSi ) A 2 n
i =1
(3.5)
where, Ai(θ)isAi(θ–αi). Foracirculararcarray,theadditionalpathlengthSiasshowninFig.3.7,for the ithsourceatradiusRandangleαisgivenby: –Si ] θ g = Ri cos _ θ − αi i
FIGURE 3.7
Foracirculararcarray,theadditionalpathlengthSiisasshown.
(3.6)
286
PART | I
Electroacoustic Devices
Therefore,thesmallerRiis,thesmallertheSidifferences,andthelessthe interferencebetweensources.Ideally,R =0forallsources.AsRapproaches 0,theinterferencewillbecomelessaudibleandfrequencyresponseacrossthe array’sintendedcoverageareawillbecomemoreuniform.
3.5 COINCIDENT ACOUSTICAL CENTERS: A PRACTICAL APPROACH Clearly, the ideal solution is to collocate all the acoustic points of origin, as showninFig.3.8.Wecouldachievethisbystackingthehornsvertically,but thiswouldsolvetheprobleminthehorizontalplanebycreatingaworsesituationinthevertical(fronttoback)direction.Figure3.9showsamorerealistic approximationthattakesintoaccountthephysicalconstraintsofloudspeaker design(thedimensionsofthetransducers,horns,enclosurewalls,etc.).Because theacousticsourcesarerealphysicalobjects,wecannotreduceRito0.Butwe cangetcloseenoughtomakemeasurable,audibleimprovementsintheperformanceofthemulti-enclosurearray.
FIGURE 3.8 Theacousticideal—colocatingtheacousticcentersofallhornsisnotapractical possibility.
3.5.1 TRAP Horns: A New Approach Figure 3.9 implies that the way to minimize Ri—and the resultant interference—istomovetheacousticcentersasfartotherearoftheenclosureaspossible.Wecanattempttominimizethesizeofthedrivers,forinstancebyusing high-outputmagneticmaterialssuchasneodymium.Butthebiggestobstacle tocoincidentacousticcentersisthehornitself.Thisisbecausetypicalconstant
Section | 3
Loudspeaker Cluster Design
287
FIGURE 3.9 Becausedriversandenclosuresarephysicalobjects,theacousticcentersofTRAP horns are not perfectly coincident but they are close enough to achieve measurable and audible reductionsininterference.
directivity horns exhibit astigmatism: their apparent points of origin are differentinthehorizontalandverticalplanes.Inordertocreateawidercoverage patterninthehorizontalplane,theapparentapexismovedforward,whilethe verticalapexisfarthertotherearbecauseitscoveragepatternisusuallynarrower. This is certainly the case with the most popular horn patterns in use today:60°×40°and90°×40°.Oneapproachtoasolution,then,istorotate thehornandusetheverticalapexofthehorninthehorizontalplane.Bydoing so,weareeffectivelymovingtheacousticcenterasfartotherearofthecabinet aspossible.Thistechniquewhencombinedwithcabinetdesignthatminimizes thespacebetweenadjacentdriversinanarray,whilematchingthetrapezoidal sideswiththeopeningangleofthehorn,createsasystemcapableofminimal interferenceinthefrequencyrangewherethehorniseffective.Thisformsthe basisforwhatIcalltheTrueArrayPrinciplebyRenkusHeinz. Subsequent refinements to the horn flare itself have been awarded U.S. Patent#5,750,943.ThisArrayguidetopologygoesevenfartherinlocatingthe apparentacousticorigintowardtherearoftheenclosure.Torepeat,moving the acoustic centers to the rear minimizes R, the distance between acoustic pointsoforiginwithinthearray,andtheresultinginterferencebetweenarray elements. Figure3.10showstheALS-1predictionsforthefirstgenerationofTRAP horns.Itisclearthatinterferencehasalmostdisappeared. Figure3.11showsmeasuredEASEdataforathree-widearrayofTRAP40 enclosures. Frequency response is consistent in both vertical and horizontal planeswithin±4dB.Thisisanoutoftheboxarray,usingnofrequencyshadingormicro-delaytoimproveperformance.Measuredresultsdon’ttrackthe predictions100%becausetheactualpatternofthehornsvariessomewhatwith frequency:first-generationTRAPhornsmaintainnominalcoverage±10°from 1kHzto4kHz.
288
PART | I
Electroacoustic Devices
FIGURE 3.10 TRAPdesignproducestrulyarrayablesystemswithminimaldestructiveinterferenceinthehorns’passband.
FIGURE 3.11 TheTRAParrayproducesalmostnomeasurableinterferencefromatight-packed three-wide cluster. This is because the three spherical wave-fronts produced by the three horns originatefromacommonacousticalcenter.Thereforetheybehaveasasingleacousticunit,without overlaporinterference.
Section | 3
Loudspeaker Cluster Design
289
3.5.2 TRAP Performance SystemsbasedontheTrueArrayPrinciplecanextendpatternbandwidth(the frequencyrangeoverwhichcoveragevarieslessthan±5°)downtothefrequency atwhichmutualcouplingbetweenadjacentcabinetsceases.TRAPsystemsare designedsothattheenclosuresprovideoptimumsplayanglesof40°between thehorns:thetrapezoidalsidesarethereforesteeperthanmanyotherdesignsat 20°perside.Thecombinationofsymmetricalhornsandsteepersidewallangles maintainscoincidentacousticcentersforalltheelementsinthearray. Note that moving the horizontal apex to the same location as the vertical resultsinasymmetrical40°×40°coveragepattern.Thisinturnrequirestheuse offourenclosurestocover160°withalmostnovariationinfrequencyresponse inthehorizontal(sidetoside)plane.With60°×40°cabinetswecoulddeliver soundto180°ofcoverage,albeitwithsomequiteaudiblevariations. There are other commercially available systems offering similar array performance to that described above. TheARC’s system from French loudspeakermanufacturerL-Acousticusesatypeofpathlengthequalizertoforce theemergingwavefronttoconformtotheopeningangleoftheirhornandalso putstheacousticcenterbehindthecabinet.InthecaseofARC,thecabinet’s trapezoidal side walls also serve as the waveguide for the high frequencies. Asthewaveguidesopeninganglematchesthatofthecabinet,thisiscertainly an elegant solution to creating minimum interference arrays at the frequency wherethehorniseffective. IntheKF900seriesfromEAW,simplephasehornsforthemidandhigh frequenciesputtheacousticcenterasclosetotherearofthecabinetaspossible,whiletheiropeninganglesalsomatchthetrapezoidalsidesoftheenclosure.TherelativelylargesizeoftheKF900seriesenclosuresandhornsbrings minimuminterferenceperformancetofrequencieslowerthanthosebasedon smallerwaveguides.Remember,thatthistechniqueforminimuminterference arrays,includingtheTrueArrayPrinciple,onlyholdstrueforthosefrequencies wherethehorniseffective.
3.6 LOW-FREQUENCY ARRAYS: BENEFICIAL INTERFERENCE Intheprecedingparagraphs,Ioutlinedtheparametersnecessarytominimize destructiveacousticinterferencebetweenadjacentcabinetsorhornsinanarray. Butthesetechniquesareonlybeneficialatthefrequencieswherethehornsare effective.Yettheseverysystemsorhornsareusedatfrequencieswellbelow theirdirectivitycutoffandlower,downtofrequencieswherethewooferspiston sizeoffersnodirectionalcontrolatall.
3.6.1 Horizontal Woofer Arrays: Maintaining Wide Dispersion For our first example, let’s look at the additional problems and opportunities we create when arraying small (12 inch woofer, 1 inch compression driver) full range loudspeaker enclosures as in Fig. 3.12. For a full range
290
PART | I
Electroacoustic Devices
FIGURE 3.12 TRAParrayscanbequitesmall;however,thesizeofthehornswilldeterminethe lowerfrequencylimitatwhichtheTrueArrayPrincipleceasestooperate.
array module, there are three frequency zones that exhibit different wavelengthrelatedbehavior.Atthelowestfrequencies,orlongestwavelengths, thesemodulesexhibitonlybeneficialinterferenceormutualcoupling.Each additionalmodulecreatesadditionalon-axisacousticoutput.Theopportunityhere,isthatlessequalizationisrequiredtomakethearray’sfrequency response flat down to these lower frequencies as compared to a single cabinet. Apotentialproblemiscreatedwhenthearraybecomestoowide,however. Fourorfiveelementarraysarewideenoughastobecomequitedirectionalin theforwardplaneatthoselowerfrequencies(20Hztoroughly500Hzormore, dependentonthemodule).Withoutasignalprocessingscheme,thisarraycannotbeequalizedtohavethesamefrequencyresponsethroughoutitsintended coverage.Itwillsoundboomyinthemiddleandthinatitscoverageextremes. Asolutionistotaperthelengthofthearrayinthehorizontalplaneinorderto maximizehorizontaldispersionofthelowerfrequencies.Theentirearraycan beusedforthelowestfrequenciesasthewavelengthsarelongest(20Hzupto about 200Hz), but at higher frequencies, as the wavelengths get shorter, the arraylengthmustalsogetshortertomaintainwidedispersion.Thisisachieved bylowpassingtheoutermostwoofersofthearray,suchthatonlytwoorthree maxwoofersareusedatfrequencieshigherthanthis. The second frequency zone that can be problematic in arrays based on full range modules, occurs at wavelengths where cabinet spacing no longersupportsmutualcoupling,andthehornhasyettoattainitsdirectivity cutoff. This typically applies to a small half octave range where adjacent cabinet spacing approaches a wavelength. Here we observe combinations
Section | 3
Loudspeaker Cluster Design
291
of destructive and constructive interference at various observation points aroundthearrayintendedcoverage,causingfrequencyresponsevariations greaterthan±6dB.Fortunatelythereisasignalprocessingtechniquethatcan minimize this effect. By simply notching this frequency range from every othercabinetwithacutequaltothegreatestamountofvariance(typically 6dB of attenuation), and width equal to the bandwidth of the aberrations (typicallyhalfanoctave),thefrequencyresponsevariationsthroughoutthe arrayscoveragecanbeminimized. The third frequency zone of wavelength related behavior for arrays basedonfullrangemodules,isthenatthefrequenciesabovewhichthehorn is effective. Let us assume that the horns depicted in Fig. 3.12 place the acousticcentertowardstherearofthecabinets,andthattheiropeningangle also matches that of the trapezoidal sides of the cabinet. Based on these assumptions,thearrayperformancewillexhibitminimuminterferencefor frequencies above 1–2kHz which happens to be the effective directivity cutoffofthehorn.Eachadditionalmodulesimplyaddsadditionalcoverage tothearray.
3.6.2 Vertical Woofer Arrays 3.6.2.1 Directivity at Frequencies Where Size Makes Horns Impractical Beneficial destructive interference sounds like an oxymoron, but there are severalcommerciallyavailablewooferarraysthattakeadvantageofthisvery technique. By applying the fundamental physics described by Harry Olson, directional woofer arrays are now available that outperform large woofer horns. Whentwopointsourcesaresuperimposedononeanother,theiroutputssimplyaddupinalldirections.Asthetwopointsourcesarespreadapart,theoutput diminishesalongtheplaneofseparationduetophasecancellation.Atexactly ½wavelength,apurenulloccurs,andweachievetheclassicfigure8,dipole polar pattern. The current commercially available systems take advantage of thisphenomena,directivitythroughoff-axisattenuation,byplacingwoofersin averticalarrayandspacingthemtocreatethisdipolarpatternatfrequencies belowwhichhornsbecometoolarge. Figure3.13isanexampleofonesucharray.TermedTri-Polarbyitsdesigner Vance Breshears, it uses the vertical spacing between the three woofers with appropriate signal processing to maintain consistent low-frequency pattern controlfrom400Hzdowntobelow100Hz.Oneofthefirstsystemsavailable wasdevelopedbyCraigJanssen,termedTuned Dipolar,whichusestwoseparate arrays.With drivers, spacing, and signal processing appropriate for their respective passbandsTuned Dipolar offers exceptional low-frequency pattern control over an extended bandwidth. Even subwoofers are now benefitting
292
PART | I
Electroacoustic Devices
FIGURE 3.13 ReferencePointArrayusingfour40°×40°mid-highenclosuresandsixlow-frequency modulesinTri-Polarconfigurationforverticalpatterncontrol,alongwithappropriatesmallfullrange systemsfordownfill.
from this type of technology. Meyer Sound is achieving cardioid patterns at lowestfrequenciesfromitsPSW-6,providingsignificantattenuationofthose frequenciesdirectlybehindtheenclosures.
3.7 LINE ARRAYS AND DIGITALLY STEERABLE LOUDSPEAKER COLUMN ARRAYS For the communication between a source and a listener to be effective, it is importantthatthelistenerreceiveandcomprehendthemessage.Inlargespaces where people gather, including auditoria, houses of worship, sports venues, transitterminals,andclassrooms,oftentheacousticrequirementsthatenable effectivespeechareinconflictwiththearchitecturalneedsofthespaces.When theacousticsofavenuecannotbealteredtoenableeffectivespeechcommunication,designingasoundreinforcementsystemtodosocanbeachallenge.
Section | 3
Loudspeaker Cluster Design
293
Recent advances in efficient amplification and digital signal processing have enabledanewclassofloudspeaker:thedigitally steerable columnorline array asit’softencalled.Theacousticalandarchitecturalbenefitsoftheseloudspeakers for sound reinforcement in highly reverberant or reflective environments willbeshown. Wewilldiscusseffectivecommunicationsanddefineintelligibilityandhow to measure it both subjectively and objectively. We will look at architecture andacousticsandatreverberationanditseffectonintelligibilityinlargepublic spaces.Finallywe’lllookatdigitallysteerablecolumnarrays,theirdesignconsiderations,andtheirperformanceandbenefitswhenusedinlargereverberant spaces. Someofthebasicprinciplesinvolvedinvoicecommunicationsare: ● ●
●
●
●
Invoicecommunicationsintelligibilityisthecapabilityofbeingunderstood. Itassumestheexistenceofacommunicationprocessbetweenatalkeranda listener,orbetweenasourceandalistener. For the conveyance of meaning, the English language is highly dependentupontheeffectivereceiptandcomprehensionofconsonants.Thisis howwedifferentiatewordsbasedonsimilarvowels.Forexample,Zoo, Two,New. Intermsoffrequencyresponse,speechrangesbetween100Hzand8kHz, withmaximumenergyaround250Hz. Inspeech,thefrequencyrangethatconveysthemostconsonantinformation istheoctavearound2kHz.
3.7.1 What Affects Intelligibility Majorinfluencesthataffectintelligibilityare: ●
●
●
●
●
●
Elocutionandpronunciationofthetalker.It’shardtounderstandsomeone whomumblesunderanycondition. Hearingacuityoflistener.Anoftenoverlookedinfluence,thosewithahearinglosshavetroubleunderstandingwhat’sbeingsaid. SNR.We’veallbeenplaceswhereitwassonoisywecouldn’tunderstand whatwasbeingsaid. Direct to reverberant ratio. The higher the reverberation level, the more difficultitistounderstandwhat’sbeingsaid. Directivity of the loudspeaker or loudspeakers. Highly directional loudspeakersdirectmoreofthesoundontotheaudienceandlessontothereflectivewallsandceilings. Thenumberofloudspeakers.Largernumbersofloudspeakerstranslateinto moreacousticenergybeingtransmittedintotheroomandhigherreverberationlevels.
294
●
●
PART | I
Electroacoustic Devices
Reverberation time. The longer the reverberation time, the more likely it willinterferewithintelligibility. Distanceofsourcetolistener.Thecloserthelisteneristotheloudspeaker, thelesslikelyreverberationwillinterfere.
Secondaryinfluencesare: ● ● ● ● ● ●
Genderoftalker. Microphonetechnique. Vocabularyandcontextofspeechinformation. Directionofmainsoundtolistenerand/ordirectionofreflectionsandechoes. Systemfidelity,equalization,anddistortion. Uniformityofcoverage.
3.7.2 Measuring Intelligibility 3.7.2.1 Subjectively Statisticaltestswithtrainedtalkersandlistenerscanbethemostreliablemetric fordeterminingtheintelligibilityofasystem.Toensurethatallspeechsounds arerepresentedinatest,PhonemicallyBalanced(PB)wordlistsarecommonly used.Thesewordlistscanbeaslongas1000words.Testsusingnonsensesyllablesorlogatoms,andModifiedRhymeTestsarealsoused.Thesetestsare verytimeconsumingandaredifficulttosetup. 3.7.2.2 Objectively Articulation Index. ArticulationIndexorAIwasoneofthefirstattemptsto quantifyintelligibilitywithmeasurements.AIisprimarilyconcernedwiththe affectofnoiseonspeech.Theindexrangesfrom0to1with0representingno intelligibility. %ALcons. %ALconsorthearticulationlossofconsonantswasdevelopedby PeutzinHollandduringthe1970’s.%ALconstakesbothnoiseandreverberationintoaccountandisbasedontheimportanceoftheoctavearound2000Hz inconveyingconsonantinformation.%ALconsusesascalerunningdownwards from0where0isperfectintelligibility,or0%articulationloss. AlthoughPeutzused2000Hzasthecenterfrequencyand2000Hzisstill theEuropeanstandard,manyacousticiansintheUSApreferusing1000Hz.As ageneralrule,%ALconscalculatedat1000Hzshowahigherarticulationloss thanonescalculatedat2000Hz. STI. STIorSpeechTransmissionIndexconsidersthesource/room/listeneras a transmission channel and measures the reduction in modulation depth of a specializedtestsignalwhichreplicatestheburstnatureofrealspeech.TheSTI scalerangesfrom0to1,where1representsperfectintelligibility.STIisconsideredthemostaccurateoftheintelligibilitymeasures.
Section | 3
295
Loudspeaker Cluster Design
Evaluation
STI
%ALcons
Bad
0.20 to 0.34
24.3 to 57
Poor
0.35 to 0.50
11.3 to 24.2
Fair
0.51 to 0.64
5.1 to 11.2
Good
0.65 to 0.86
1.6 to 5.0
Excellent
0.87 to 1.00
0.0 to 1.5
Copied from The Audio System Designer Technical Reference by Peter Mapp and published by Klark Teknik.
3.7.3 Architecture and Room Acoustics 3.7.3.1 Reverberation Reverberationisthepersistenceofsoundinaspaceaftertheoriginalsoundhas beenremoved. RT60isthemeasureforreverberation,anditisdefinedastheamountof timerequiredfortheaveragesoundenergydensityinaspacetodecreasefrom itsoriginalvalueby60dBaftertheoriginalsoundhasstopped. TheSabineequationrelatesRT60tothevolumeofaroomwithitssurface areaandtheabsorptioncoefficientsofthematerialsappliedtothesurfaces. As room volume increases relative to surface area and absorption coefficients,theRT60increases. As surface area and absorption increase relative to room volume, RT60 decreases.Itisthispersistenceofsoundthatinterfereswithourcomprehension ofconsonantsandcontributestowardsdegradingintelligibility.Table3.1shows theeffectofreverberationonintelligibility. TABLE 3.1 Intelligibility Comparison Chart RT60
1.5 s
Careful system design is required.
RT60
>1.7 s
Limit for good intelligibility in large spaces.
RT60
>2 s
Very directional loudspeakers are required, intelligibility can have limitations.
RT60
>2.5 s
Intelligibility will probably have limitations.
RT60
>4 s
Highly directional loudspeakers will be required to achieve acceptable intelligibility.
296
PART | I
Electroacoustic Devices
3.7.4 Line Arrays Figures3.14to3.16showthedirectsoundcoverageofvariousloudspeakersin asanctuary100ft×65ft.Thechanceladds20fttoitslength.Theroofpeaks at52ft.Theroomvolumeisroughly250,000ft3.Theroomhasplasterwalls, woodceiling,terrazzofloors,andemptywoodenpews.ThisproducesaRT60 ofabout3.5s. NoticethehighSPLlevelsonthewallsandceilingintheflown-hornarray simulation. The high-frequency beaming of the mechanically tilted column
FIGURE 3.14
Flownlargeformathornarray.
FIGURE 3.15 Mechanicallytilted4metercolumnarray.
FIGURE 3.16
Digitallysteeredcolumnarray.
Section | 3
Loudspeaker Cluster Design
297
array prevents good coverage of the front of the audience area.The digitally steerablecolumnarraycoversonlytheaudienceareaandhasverylittlecoverageonthewallsandnocoverageoftheceiling.Onlythesteeredcolumnarray hasacceptable(goodtofair)intelligibilitythroughouttheaudienceareas.Digitallysteerablecolumnarrayscanoffersuperiorcoverageandtheycanprovide improvedD/R.Theycanprovideimprovedintelligibilityinhighlyreverberant spaces,plustheyblendbetterwiththeirsurroundingarchitectureandarenearly invisibleinuse.
3.7.4.1 Digitally Steered Column Arrays When the room size and volume are fixed and adding absorption to reduce the RT times is not an option, digitally steerable column arrays offer a new solution: ● ●
● ●
●
Theyhavetheabilitytobemuchmoredirectionalthanthelargesthorns. Theideaisnotnew;theconceptsforthesecolumnarraysweredescribedby HarryOlsonin1957.Onlytheimplementationisnew. Thehardwarerequiredtoimplementtheseideasisnowavailable. DigitalSignalProcessingrequiredisnowamaturetechnology,verypowerfulandrelativelyinexpensive. Compact, highly efficient Class D amplifiers are capable of high-fidelity performance.
LineArraysarenotanewidea.HarryF.Olsondidthemathanddescribed thedirectionalcharacteristicsofacontinuouslinesourceinhisclassicAcoustical Engineering,firstpublishedin1940.Traditionalcolumnloudspeakershave alwaysmadeuseoflinesourcedirectivity. Simplelinearrays(columnarrays)arebasicallyanumberofdriversstacked closely together in a line, Fig. 3.17. Simple line arrays become increasingly directionalintheverticalplaneasthefrequencyincreases.Thespacingbetween driverscontrolsthehigh-frequencylimits.Theheight(length)ofthelinearray determinesthelow-frequencycontrollimit.Figure3.18showsthelinesource directivityasdescribedbyHarryOlsonin1957. Thedirectivityofalinearrayisafunctionofthelinelengthandthewavelength.Asthewavelengthapproachesthelinelength,thearraybecomesomnidirectional, Fig. 3.19. Figure 3.20 shows the vertical dispersion pattern of a typicallinearray
3.7.4.2 Controlling High-Frequency Beaming Simplelinearraysbecomeincreasinglydirectionalasthefrequencyincreases, infact,athigherfrequenciestheybecometoodirectional.Theverticaldirectivitycanbemademoreconsistentbymakingthearrayshorterasthefrequency increasesbyusingfewerdrivers.OneamplifierchannelandoneDSPchannel perdrivermakethispossible.
298
PART | I
Electroacoustic Devices
Driver spacing controls HF limit
Array height controls LF limit
FIGURE 3.17
Basiclinearraytheory.
FIGURE 3.18 Directionalcharacteristicsofalinesourceasafunctionofthelengthandthewavelength.Thepolargraphdepictsthesoundpressureatalargefixeddistance,asafunctionofangle. Thesoundpressurefortheangle0°isarbitrarilychosenasunity.Thedirectioncorrespondingtothe angle0°isperpendiculartotheline.Thedirectionalcharacteristicsin3Daresurfacesofrevolution aboutthelineasanaxis.(FromAcousticalEngineeringbyHarryOlson.)
3.7.4.3 Beam Steering Thebeamcanbesteeredupordownbydelayingthesignaltoadjacentdrivers. DSPcontrolalsoallowsustodevelopmultiplebeamsfromasinglelinearray andindividuallysteerthesebeams. DSP control also allows us to move each beam’s acoustic center up and downthecolumnallowingustocreatemultiplebeamsandalsosteerthebeam, Figs.3.21and3.22.
Section | 3
Loudspeaker Cluster Design
FIGURE 3.19
Simplelinesourcedirectivityasafunctionoflinelengthversusfrequency.
FIGURE 3.20
Typicallinearrayverticaldispersiondisplay.
FIGURE 3.21
Verticaldispersiondisplayshowingmultiplebeamcapability.
299
300
FIGURE 3.22
PART | I
Electroacoustic Devices
Graphicillustrationofbeamsteering.
3.7.5 DSP-Driven Vertical Arrays 3.7.5.1 Acoustical, Electronic, and Mechanical Considerations Practical examples are taken from the new Renkus Heinz IC Series Iconyx steerablecolumnarrays.Iconyxisasteerablecolumnarraythatcombinesvery highdirectivitywithaccuratereproductionofsourcematerialinacompactand architecturallypleasingpackage,Fig.3.23. Likeeveryloudspeakersystem,Iconyxisdesignedtomeetthechallengesof aspecificrangeofapplications.Manyofthecriticaldesignparametersare,of course,determinedbythenatureofthesetargetapplications.Tounderstandthe decisionsthathavebeenmadeduringthedesignprocesswemuststartwiththe particularproblemsposedbytheintendedapplications. ThefunctionofindividualdrivercontrolandDSPistomakemoreeffective useofthisphenomenon.Noamountofsiliconcangetaroundthelawsofacousticalphysics.Theacousticalpropertiesoffirst-generationcolumnloudspeakers aresetbytheacousticalcharacteristicsofthetransducersandthephysicalcharacteristicsofthepackage: 1. Theheightofthecolumndeterminesthelowestfrequencyatwhichitexerts anycontrolovertheverticaldispersion. 2. Theinter-driverspacingdeterminesthehighestfrequencyatwhichthearray actsasalinesourceratherthanacollectionofseparatesources. 3. Horizontal dispersion is fixed and is typically set when the drivers are selected,becausecolumnloudspeakersdonothavewaveguides. 4. Other driver characteristics such as bandwidth, power handling, and sensitivity will determine the equivalent performance characteristics of the system. Oneunfortunatecorollaryofthesecharacteristicsisthatthepowerresponse ofaconventionalcolumnloudspeakerisnotsmooth.Itwilldelivermuchmore low-frequencyenergyintotheroomandthisenergywilltendtohaveawider vertical dispersion. This can make the critical distance even shorter because the reverberant field contains more low-frequency energy, making it harder
301
Section | 3
Loudspeaker Cluster Design
FIGURE 3.23
TypicallinearrayandatypicalIconyxColumn.
for the listener to recognize higher-frequency sounds such as consonants or instrumentalattacktransients.
3.7.5.2 Point Source Interactions 3.7.5.2.1 Doublet Source Directivity Doubletsourcecancelseachother’soutputdirectlyaboveandbelow,because theyarespaced½wavelengthapartintheverticalplane.Inthehorizontalplane, bothsourcessum.TheoveralloutputlookslikeFig.3.24.
5
l 2
5
5
FIGURE 3.24 loudspeakers.
Output of a signal whose wavelength is ½ of the space between the two
302
PART | I
Electroacoustic Devices
Whentwosourcesare¼wavelengthapartorless,theybehavealmostlikea singlesource.Thereisveryslightnarrowingintheverticalplane,Fig.3.25a. Thereissignificantnarrowingintheverticalplaneat½wavelengthspacing, becausethewaveformscanceleachotherintheverticalplane,wheretheyare 180°outofphase,Fig.3.25b. Atonewavelengthspacingthetwosourcesreinforceeachotherinboththe verticalandhorizontaldirections.Thiscreatestwolobes,oneverticalandthe otherhorizontal,Fig.3.25c. Astheratioofwavelengthtointer-driverspacingincreases,sodothenumber oflobes.Withfixeddriversasusedinlinearrays,theratioincreasesasfrequency increases(λ = c/fwherefisthefrequencyandcisthespeedofsound),Fig.3.25d.
FIGURE 3.25a
λ/4(¼wavelength).
FIGURE 3.25c
λ(1wavelength).
FIGURE 3.25b
λ/2(½wavelength).
FIGURE 3.25d Increasedwavelengthto inter-driverspacing.
Section | 3
303
Loudspeaker Cluster Design
3.7.5.2.2 Array Height versus Wavelength (λ) Driver-to-driverspacingsetsthehighestfrequencyatwhichthearrayoperates asalinesource.Thetotalheightofthearraysetsthelowestfrequencyatwhich ithasanyverticaldirectivity. Figures3.26athrough3.26dshowtheeffectofarrayheightversuswavelength. At wavelengths of twice the array height, there is no pattern control, the outputisthatofasinglesourcewithveryhighpowerhandling,Fig.3.26a. Asthefrequencyrises,wavelengthapproachestheheightoftheline.Atthis pointthereissubstantialcontrolintheverticalplane,Fig.3.26b. Athigherfrequenciestheverticalbeamwidthcontinuestonarrow.Some sidelobesappearbuttheenergyradiatedinthisdirectionisnotsignificantcomparedtothefrontandbacklobes,Fig.3.26c. Withstillfurtherverticalnarrowing,sidelobesbecomemorecomplexand somewhatgreaterinenergy,Fig.3.26d.
FIGURE 3.26a Wavelengthistwicethe loudspeakerheight.
FIGURE 3.26b Wavelengthisthe loudspeakerheight.
FIGURE 3.26c Wavelengthisone-half theloudspeakerheight.
FIGURE 3.26d Wavelengthis one-fourththeloudspeakerheight.
304
PART | I
Electroacoustic Devices
3.7.5.2.3 Inter-Driver Spacing versus Wavelength (λ) The distinction between side lobes and grating lobes should to be maintained. Sidelobesareadjacenttoandradiateinthesamedirectionastheprimarylobe. Gratinglobesarethestrongsummationstangentialtotheprimarylobe.Sidelobes will be present in any realizable line array, grating lobes form when the interdriverspacingbecomeslessthan½wavelength.Itmightalsobegoodtopointout thatallofthegraphicsforthissectionaredoneusingtheoreticalpointsources. Figures3.27athrough3.27dshowtheeffectofinter-driverspacingversus wavelength. When the drivers are spaced no more than ½ wavelength apart, the array producesatightlydirectionalbeamwithminimalsidelobes,Fig.3.27a. Asthefrequencyrises,wavelengthapproachesthespacingbetweendrivers. Atthispoint,gratinglobesbecomesignificantinthemeasurement.Theymay notbeaproblem,ifmostoralloftheaudienceislocatedoutsidethesevertical lobes,Fig.3.27b. Atstillhigherfrequencies,lobesmultiplyanditbecomeshardertoisolate theaudiencefromthelobesortheirreflections,Fig.3.27c.
FIGURE 3.27a Interspacingis one-halfthewavelength.
FIGURE 3.27c Interspacingistwo timesthewavelength.
FIGURE 3.27b Interspacingisone timesthewavelength.
FIGURE 3.27d Interspacingisfour timesthewavelength.
Section | 3
Loudspeaker Cluster Design
305
As inter-driver spacing approaches four times the wavelength, the array isgeneratingsomanygratinglobesofsuchsignificantenergythatitsoutput closelyapproximatesasinglepointsource,Fig.3.27d.Wehavecomefullcircle to where the array’s radiated energy is about the same as it was when array heightwas½λ.AsshowninFig.3.26d,thisisthehigh-frequencylimitofline arraydirectivity. Asrealdriversareconsiderablymoredirectionalthanpointsourcesatthe frequencieswheregratinglobesaregenerated,thegratinglobesaremuchlower inlevelthantheprimarylobe,Figs.3.28and3.29.
FIGURE 3.28
3Dviewofasecond-generationIconyxarrayat4000Hz.
FIGURE 3.29 Sideviewofasecond-generationIconyxarrayat4000Hz.
306
PART | I
Electroacoustic Devices
3.7.6 Multichannel DSP Can Control Array Height The upper limit of a vertical array’s pattern control is always set by the inter-driverspacing.Thedesignchallengeistominimizethisdimensionwhile optimizingfrequencyresponseandmaximumoutputanddoitwithoutimposing excessive cost. Line arrays become increasingly directional as frequency increases, in fact, at high frequencies they are too directional to be acousticallyuseful.However,ifwehaveindividualDSPavailableforeachdriver,we canuseittomakethearrayacousticallyshorterasfrequencyincreases—this willkeeptheverticaldirectivitymoreconsistent.Thetechniqueisconceptually simple—use low-pass filters to attenuate drive level to the transducers at the topandbottomofthearray,withsteeperfilterslopesontheextremeendsand moregradualslopesasweprogresstothecenter.Asbasicasthistechniqueis, itispracticallyimpossiblewithoutdevotingoneamplifierchannelandoneDSP channeltoeachdriverinthearray. AsimplifiedschematicshowshowmultichannelDSPcanshortenthearray asfrequencyincreases.Forclarity,onlyhalftheprocessingchannelsareshown anddelaysarenotdiagrammed,Fig.3.30.
FIGURE 3.30
MultichannelDSPshortenstheloudspeakerlength.
3.7.7 Steerable Arrays May Look Like Columns But They Are Not Simplecolumnloudspeakersprovideverticaldirectivity,buttheheightofthe beamchangeswithfrequency.TheoverallQoftheseloudspeakersistherefore lowerthanrequired.Manyearlydesignsusedsmall-conefullrangetransducers, andthepoorhigh-frequencyresponseofthesedriverscertainlydidnothingto enhancetheirreputation.
3.7.7.1 Beam-Steering: Further Proof that Everything Old is New Again AsDonDavisfamouslyquotesVernKnudsen,“Theancientskeepstealingour ideas.”HereisanotherillustrationfromHarryF.Olson’sAcoustical Engineering. This one shows how digital delay, applied to a line of individual sound
Section | 3
Loudspeaker Cluster Design
307
FIGURE 3.31 A delay system for tilting the directional characteristic of a line sound source. (FromAcousticalEngineeringbyHarryOlson.)
sources,canproducethesameeffectastiltingthelinesource.Itwouldbelong after1957beforethecostofthisrelativelystraightforwardsystembecamelow enoughforcommerciallyviablesolutionstocometomarket,Fig.3.31.
3.7.7.2 DSP-Driven Arrays Solve Both Acoustical and Architectural Problems 3.7.7.2.1 Variable Q DSP-drivenlinearrayshavevariableQbecausewecanusecontrolledinterferencetochangetheopeningangleoftheverticalbeam.TheRenkusHeinzIC Seriescanproduce5°,10°,15°,or20°openinganglesifthearrayissufficiently tall(anIC24istheminimumrequiredfora5°verticalbeam).Thisvertically narrowbeamminimizesexcitationofthereverberantfieldbecauseverylittle energyisreflectedofftheceilingandfloor. 3.7.7.2.2 Consistent Q with Frequency By controlling each driver individually with DSP and independent amplifier channels,wecanusesignalprocessingtokeepdirectivityconstantoverawide operatingband.Thisnotonlyminimizesthereverberantenergyintheroom, butdeliversconstantpowerresponse.ThecombinationofvariableQ,whichis muchhigherthanthatofanunprocessedverticalarray,withconsistentQovera relativelywideoperatingband,isthereasonthatDSP-drivenIconyxarraysgive acousticalresultsthataresomuchmoreuseful. 3.7.7.2.3 Ability to Steer the Acoustic Beam Independently of the Enclosure Mounting Angle Althoughbeam-steeringisrelativelytrivialfromasignal-processingpointof view,itisimportantforthearchitecturalcomponentofthesolution.Acolumn mountedflushtothewallcanbemadenearlyinvisible,butadown-tiltedcolumn isanintrusiononthearchitecturaldesign.AnyDSP-drivenarraycanbesteered.
308
PART | I
Electroacoustic Devices
Iconyxalsohastheabilitytochangetheacousticcenterofthearrayintheverticalplanewhichcanbeveryusefulattimes. 3.7.7.2.4 Design Criteria: Meeting Application Challenges Thepreviousfiguresmakeitclearthatanylinesource,evenwithverysophisticatedDSP,cancontrolonlyalimitedrangeoffrequencies.However,byusing full range coaxial drivers as the line source elements could make the overall soundofthesystemmoreaccurateandnaturalwithoutseriouslycompromising the benefits of beam-shaping and steering. In typical program material, most of the energy is within the range of controllable frequencies. Earlier designs radiate only slightly above and below the frequencies that are controllable. Thusmuchoftheprogramsourceissacrificed,withoutasignificantincrease inintelligibility. Tomaximizetheeffectivenessofadigitallycontrolledlinesource,it’snot enoughtostartwithhigh-qualitytransducers.TheRenkusHeinzIconyxloudspeakersystemusesacompactmultichannelamplifierwithintegralDSPcapability.TheD2audiomodulehastherequiredoutput,fullDSPcontrol,andthe addedadvantageofapurelydigitalsignalpathoption.WhenPCMdataisdeliveredtothechannelviaanAES/EBUorCobraNetinput,theD2audioprocessor/ amplifierconvertsitdirectlyintoPWMdatathatcandrivetheoutputstage. 3.7.7.2.5 Horizontal Directivity is Determined by the Array Elements Verticalarrays,includingIconyx,canbesteeredonlyintheverticalplane.Horizontalcoverageisfixedandisdeterminedbythechoiceofarrayelements.The transducersusedinIconyxmoduleshaveahorizontaldispersionthatisconsistentoverawideoperatingband,varyingbetween140°and150°from100Hz to16kHz. 3.7.7.2.6 Steering is Simple—Just Progressively Delay Drivers Ifwetiltanarray,wemovethedriversintimeaswellasinspace.Considera linearrayofdriversthatishingedatthetopandtilteddownward.Tiltingmoves thebottomdriversfurtherawayfromthelistenerintimeaswellasinspace.We canproducethesameacousticaleffectbyapplyingprogressivelylongerdelays toeachdriveraswemovefromtoptobottomofthearray. Again,steeringisnotanewidea.Itisdifferentfrommechanicalaiming— frontandrearlobessteerthesamedirection. 3.7.7.2.7 BeamWare: The Software That Controls Iconyx Linear Array Systems Aseriesoflow-passfilterscanmaintainconstantbeamwidthoverthewidest possiblefrequencyrange.Theideasaresimple,butforthemostbasicIconyx array,theIC16,wemustcalculateandapply16setsofFIRfilters,and16separatedelaytimes.Ifweintendtotakeadvantageofconstantinter-driverspacing
Section | 3
Loudspeaker Cluster Design
309
to move the acoustical center of the main lobe above or below the physical center of the array, we must calculate and apply a different set of filters and delays.Theoreticalmodelsarenecessary,butthebehaviorofrealtransducersis morecomplexthanthemodel.Eachofthecomplexcalculationsunderlyingthe Iconyxbeam-shapingfiltersweresimulated,thenverifiedbymeasuringactual arraysinourrobotictestandmeasurementfacility.Fortunately,thecurrentgenerationoflaptopanddesktopCPUsareuptothetask.BeamWaretakesuser inputingraphicform(sidesectionoftheaudiencearea,locationandmounting angleofthephysicalarray)andprovidesbothasimulationofthearrayoutput thatcanbeimportedintoEASEv4.0orhigher,andasetofFIRfiltersthatcan bedownloadedtotheIconyxsystemviaRS422serialcontrol.Theresultisa graphicaluserinterfacethatdeliversprecise,predictable,andrepeatableresults inreal-worldacousticalenvironments.
This page intentionally left blank
Index
A
Accelerometer,166 Acousticlenses,230–231 Acousticpower,262 Acoustictransferfunctionofhumanexternal ear,120–122 Acousticalnoise,32 Acousticalresonantdevices,28 Activecrossoversversuspassivecrossovers, 248–249 AI.seeArticulationindex %ALcons,294 Alnicomagnetwoofer,200 Aluminumdiaphragm,27 Ampliierpowering,methodsof,39 Amplituderesponsemeasurements,259 AMS.see Automaticmicrophonesystems Antennas,134–137,147–151 cablesandrelatedsystems,149–150 combinationdiversity,136 placement,136–137 splitters,150–151 switchingdiversity,134 transmitter,145 truediversity,135–136 ARCsystem,289 Arrayheight multichannelDSPcontrolling,306 versuswavelength(λ),303 Arrays,279 circulararc,285 conventional,280–284 analysis,284–286 digitallysteeredcolumn,297 DSP-drivenvertical.seeDSP-driven verticalarrays line.seeLinearrays lowfrequency horizontalwoofer,289–291 verticalwoofer,291–292 ofmultipleenclosure,281 problemsandpartialsolutions,280 ofsingleenclosure,280,281 steerable.seeSteerablearrays Articulationindex(AI),294
Astigmatism,287 Asymmetricdirectivityhorns,227–230 JBL4660,229 Attenuation,20,149,163 inversesquarelawequationfor,54 Attenuators,170 Automaticmicrophonesystems(AMS),94–95 ShureAMS22,94–95
B
Balancedmicrophones,59 Bandpassilter,276 Basestationpowermicrophones,75–76 Shure450SeriesII,75 frequencyresponseof,76 Basshorns,234 Beam-steering,298,306,307 Bidirectionalmicrophones,9–10,175 Binauralrecording,microphonesfor, 120–126 artiicialheadsystems,120–124 In-the-Ear(ITE™)recordingmicrophones, 124–126 Binauralsignalprocessing,120 Biradialhorns,226,227 Booms,169–170 AtlasBB-44microphoneboom,170 Butterworthalignment,267
C
Cables,microphone,56,60 Capacitormicrophones,33–35,174,182,183, 188 acousticwaveandelectricalsignals,35 advantagesof,35 conventionaltransducer,44 distortioncharacteristicsof,34,35 electret,48–49,133 noisereduction,45–48 noisesources,45 omnidirectionalpressure,46 SennheiserMKH20,46,47 phantompowerfor,37–40 artiicialcentertapconnectedmethod, 40
311
312 directcentertapconnectionmethod,39 droppingresistorvaluechartfor,39 preampliier,38 radio-frequency(RF).seeRadio-frequency capacitormicrophones symmetricaltransducer,44 unidirectionalpressure/pressure-gradient,46 voltagedivider-type,34,36–37 Capacitor-typeloudspeaker,208 Carbonmicrophones,23–24 Cardioidmicrophones,11,12,100,103,104 acousticaldelay,12,13 frequencyresponseof,13,14 high-frequencyrearresponseof,16 multiple-entry,22 Electro-VoiceRE20variable-D,22,23 proximityeffects,14 single-entry,17–20 Electro-VoiceDS35,17–19 ShureSM-81,18,19 three-entry,20–21 SennheiserMD441,20,21 two-diaphragm,12 two-way,23 AKGD-222EB,23,24 typesof,17,18 Ceramicmicrophones,24–26 Cerwin-VegaEhorn,233 Circulararcarray,285 CLF.see Commonloudspeakerformat Coaxialloudspeaker,239 Cocktailpartyeffect,120 Coincidentacousticcenters,286–289 ofTRAPhorns,286–288 Coincidentmicrophonetechnique,98–100 Comb-iltereffect,61 Commonloudspeakerformat(CLF),255,256 Companding,137,145 Compressiondriver,203–206 advantageof,203 Altec802/808alnico1inchthroat,204,205 ceramic-magnet,203,204 JBL24402inchalnicothroat,203,204 M4midrange,205 Condensermicrophone.see Capacitor microphones Conedriver,198 suspensionof,201 Coneradiator,221–222 Controlleddirectivityhorns,225–226 Electro-VoiceHR9040,225–226 Controlled-reluctancemicrophone,77
Index Conventionalhornarray,280–284 analysis,284–286 inenclosureswithtrapezoidalsides,280,282 interferencepatterns,282–284 wideningsplaybetween,283 Crossoverfrequency,243 Crossovernetwork,243–249 ilterslopesusedin,244 Crown®SASS-PMKII.see SASS microphones Crystalmicrophones,24–26
D
Dc-to-Dcconverter,144 Diaphragms,166 aluminum,27 Mylar™,27,28 types,199–200 Differencefrequencymethod,42,43 Differentialnoise-cancelingmicrophones,76 Shure577Bdynamic,76 Diffraction,7,9,16,249 Diffuse-ieldequalizer,123 Digitalaudiosignal,154 DigitalHybridWireless™,159 Digitalmixers,156 Digitalwirelesstransmission,future considerations,154–155 Digitallysteeredcolumnarrays,297 Directradiator,221 enclosuredesign,265–270 sealed-boxsystems,265–267 ventedboxes,267–269 Directionalmicrophones,175–176 Distancefactor,92 Dolby®Pro-LogicIIdecoder,120 Dolby®Pro-LogicIIencoder,119 Domeradiator,202,222 advantagesof,202 Dometweeter,on-axisresponsemeasurements on,250,251 Drummicingtechniques,176–180 aboveversusunderneathmicing,178 bassdrum,178–179 closedrum,176–177 cymbals,179 distantdrum,180 tom-tommicing,178 DSP-drivenlinearrays,307 DSP-drivenverticalarrays,300–305 acoustical,electronicandmechanical considerations,300–301
313
Index arrayheightversuswavelength(λ),303 doubletsourcedirectivity,301 inter-driverspacingversuswavelength(λ), 304–305 Dummyheadbinauralstereomicrophone system,123 Dynamicmicrophone,26–32 omnidirectionaldiaphragmandvoice-coil assembly,27,28 ShureBeta57,21 transientconditioneffectson,30,31 varyingsoundpressureeffectson, 29,30
E
EIAstandard.see ElectronicIndustries Associationstandard Electretcapacitormicrophone,48–49,133 ShureSM81,48 Electrodynamictransducers,198–199 Electromagneticloudspeaker displacementlimitof,215–217 heatconductionin,217 mechanicalfailure,219 outputlimitationsof,215–219 thermalfailure,219 thermallimitof,216 thermaltimeconstantof,218 Electronicieldproduction(EFP),129 ElectronicIndustriesAssociation(EIA) standard microphonesensitivity,51–52 RS-297,58 RS-221.A,57 SE-105,51 Electronicnewsgathering(ENG),129 Electrostaticloudspeaker couplingcircuitandhigh-voltagepower supplyfor,210 diaphragmof,207 maximumpoweroutputof,209 Electrostatictransducers,206–210.see also Electrodynamictransducers Equalizers,170 diffuse-ield,123 Explosivebreathsound,163
F
FastFouriertransfer(FFT)measurement, 275–276 Ferrite-magnetdrivers,symmetricield geometry,201,202
Ferroluids,218 FFTmeasurement.see FastFouriertransfer measurement First-ordercrossover,impulseresponseof, 246 First-ordergradientmicrophone.see Velocity microphone Focused-ieldgeometry,201 Foldedhorns,231–234 advantageof,234 Fourth-orderLinkwitz-Reillyilterpair, impulseresponseof,247 Frequencyresponse ofcardioidmicrophones,13,14 ofCountrymanIsomaxE6Directional EarSet,74 versusdistanceforElectro-VoiceDS35, 18,19 ofDPA4017microphone,83 ofheavy-styledynamiclavalier microphone,70 ofinterferencetubemicrophone,84,85 oflavaliermicrophones,70 ofloudspeakers,253 ofomnidirectionalmicrophones,8 ofpressurezonemicrophones,65–68 ofShure450SeriesIIbasestation microphones,76 ofShureSM58vocalmicrophone,78 ofunidirectionalmicrophones,15–17 Frequency-modulatedmicrophone,40
G
Groundingofmicrophones,56–57
H
Handheldentertainermicrophones,77–79 SennheiserMD431,77,79 ShureSM58vocal,77 frequencyresponseof,78 placementfor,78 Handheldwirelessmicrophones,128,129,133 Headsystems,artiicial,120–124 Head-wornmicrophones,72–75 CountrymanIsomaxE6DirectionalEarSet, 73,74 frequencyresponseandpolarresponse of,74 ShureBeta53,72,73 ShureSM10A,72 High-frequencyhorn,IMAX®PPS,228 High-impedancemicrophones,59
314 High-powerwoofers,heattransferdesignsfor, 219–221 Holophone®H4SuperMINIsurroundsound system,119–120 Holophone®H2-PROsurroundsoundsystem, 117–118 Horizontalwooferarrays,maintainingwide dispersion,289–291 Horns,223–224,270–271 asymmetricdirectivity,227–230 biradial,226,227 controlleddirectivity,225–226 cutofffrequency,271 folded,231–234 low-frequency,234 micing,188–189 multicell,224–225 radial,224 radiationimpedanceof,271 TRAP.seeTRAPhorns voicewarning,226–227 Hypercardioidmicrophones,11
I
Iconyxsteerablecolumnarrays,300,308 IMAX®PPShigh-frequencyhorn,228 Impedance ofloudspeakers,255–257 ofmicrophones,59–60 Impulseresponse ofirst-ordercrossover,246 offourth-orderLinkwitz-Reillyilterpair, 247 ofloudspeakers,274–275 ofsecond-ordercrossover,246 Independentofdirection(ID)equalization,123 Inlinemicrophoneprocessors,156–159 Intelligibility comparisonchart,295 factorsaffecting,293–294 measuring,294–295 Intercoms,wireless,128,133 Interferencetubemicrophone,82–89 DPA4017,82 directionalcharacteristicsof,82–84 frequencyresponseof,83 frequencyandphaseresponseof,84,85 polarpatternsof,85 SennheiserMKH60P48,polarpatternof, 86 ShureSM89,86,87 supercardioid,83,86
Index Intermodulation(IM),141 Intermodulationrejection,143 In-the-Ear(ITE™)recordingmicrophones, 124–126 ISM(industrial,science,andmedicine)bands, 129 ITErecordings.see In-the-Ear(ITE™) recordingmicrophones
J
JamesBulloughLansing(JBL),ferrite-magnet drivers,201
K
KF900seriesenclosuresandhorns,289 Klipschorn,233
L
Lavaliermicrophones,69–72 CountrymanB6,70,71 frequencyresponseof,70 polarresponseof,71 SennheiserMKE104clip-on,69,70 ShureSM11dynamicomnidirectional,69 ShureSM183omnidirectionalcondenser, 70 Lavaliermodels,129 Leadzirconatetitanate,piezoelectric transducers,210 Leaftweeter,202,203 Lecternmicrophones,186 LectrosonicsUH400Aplug-ontransmitter, 157–159 applications,158 blockdiagramof,158 Linearrays,241–243,296–300 beamsteering,298 controllinghighfrequencybeaming,297 DSP-driven,307 Iconyx,308 theory,298 Linearmotor,213 Lineartransferfunction,271–272 Lossyladdernetwork,20 Loudspeakerarray.see Array Loudspeakercluster irst-generation,280 second-generation,280,281 Loudspeakers,234–251 acousticboundariesof,249–251 acousticalandarchitecturalbeneitsof,295 amplituderesponsemeasurementsfor,259
Index characterizationfordesignpurposes, 257–258 characterizationforuser,258–261 communication,196 components,197 coniguration,235–236 decouplingcenterconein,200 diffractionin,249 directsoundcoverageof,296 effectonmaximumoutputof,244 effectonresponseof,244–246 eficiencyof,252 FFTmeasurement,275–276 frequencybandsin,236 frequencyresponse,253 impulseresponsemeasurement,274–275 lineartransferfunctionof,271–272 MLSmeasurements,275 networktransferfunctionof,253 nonlineardistortion,257 on-axisamplituderesponsefor,259 performance,characterizationof,252–261 performanceissuesinmultiway,237–241 piezoelectric,210–212 simpleharmonicdistortion,257 soundproduction,197 soundreinforcement,196 soundreproduction,197 testingandmeasurementof,271–277 transferfunction,253–255 two-way,236–237 two-waycoaxial,239 typesof,236–237 useof,196–197 Lowfrequencyarrays beneicialinterference,289–292 horizontalwoofer,289–291 verticalwoofer,291–292 Low-frequencyhorns,234.see also Highfrequencyhorn Low-impedancemicrophone,59–60
M
M4midrangecompressiondriver,205 Magneplanar®loudspeaker,202 Magneticconeloudspeaker,213 Maximumlengthsequence(MLS) measurements,275 Maximumpoweroutputsensitivity,50–51 Micingtechniques,176–192 drums.seeDrummicingtechniques forelectricinstruments,190–191
315 horns,188–189 musicians,176 forpercussioninstrument,191–192 piano,179–182 singers,182–184 stereo,173–174 strings,186–188 vocals,182,183 woodwinds,189–190 Microphones accessories,156–171 forbinauralrecording.seebinaural recording characteristicsof,174–176 choosing,174 classiications,5 performancecharacteristicsof,6 pickuppatterns bidirectional,9–10 omnidirectional,5–9 unidirectional,10–23 forstereophonictelevisionbroadcasting,111 techniques,171–192 Mid/side(M/S)cartridge.see M/S MLSmeasurements.see Maximumlength sequencemeasurements Modulation,139,140,155 MotorolaKSN1001Aloudspeakers,211 Moving-coilmicrophone.see Dynamic microphone M/Smicrophones AKGC-422,104,105,107 ShureVP88,104,108 polarresponseof,109 stereopickuppatternof,110–112 transformersumandmatrixfor,106 M/Sstereotechnique,104–111 Multicellhorns,224–225 AltecLansing311-60,225 AltecLansing1.4-inchthroat,225 MultichannelDSP,controllingarrayheight, 306 Multichannelwirelessmicrophones,139–155 antennas,147–151 cablesandrelatedsystems,149–150 splittersystems,150–151 digitalwirelesstransmission,154–155 frequencies,140 frequencycoordination,140–143 frequencydeviation,140 receiverconsiderations,RFsignallevel, 146–147
316 spacing,140 systemplanningfor,152–153 location,152 quantityandfrequencycoordination, 153 tuningcomponents,153 transmitterconsiderations,143–145 audioprocessing,144–145 Dc-to-Dcconverter,144 rangeandRFpower,143–144 transmitterantenna,145 wirelessmonitorsystems,151–152 Multiwayloudspeakersystems functionaldiagramof,237,238 performanceissuesin,237–241 MXLMicMate™USBadapter,159,160 Mylar™diaphragms,27,28
N
Noise acoustical,32 pink,16,273 random,16,50,88 thermal,53–54 Noisesourcesofcapacitormicrophones,45–48
O
Octavebandisobar,260–261 OficedeRadiodiffusionTélévisionFrançaise technique.see ORTFtechnique Omnidirectionalmicrophones,5–9,29,97 high-frequencydirectivityof,8 proximityeffects,8 switchablediffuseieldcorrection,46 vibrationsensitivityof,32 On-axisamplituderesponse,259 Open-circuitvoltagesensitivity,49–50 ORTFtechnique,103
P
Panelradiators,223 PARplayback.see PinnaAcousticResponse (PAR™)playback Parabolicmicrophones,90–91 polarpatternfor,91 Passivecrossoversversusactivecrossovers, 248–249 PCC.see Phasecoherentcardioidmicrophone Phasecoherentcardioidmicrophone(PCC),63 CrownPCC®-160,63 horizontalplanepolarresponseof,64 CrownPCC®-200,64
Index Pianomicing,179–182 Piezoelectricceramicmaterials,25,210 Piezoelectricloudspeaker,210–212. see also Electrostatictransducers LubellLabsunderwater,212 MotorolaKSN1001A,211 PZT,210 Piezoelectricity,25,210 Pinknoise,16,273 PinnaAcousticResponse(PAR™)playback, 124 loudspeakerarrangementusedfor,124 Piston directivity,263–265 directivityfunction,264 inininitebafle,261–263 resistancefunction,262 Placements,microphone abovethesource,56 behindobjects,56 directversusreverberantield,56 distancefromlargesurfaces,55 microphone-to-sourcedistance,54–55 forShureSM58vocalmicrophone,78 PolarFlex™microphonesystem,95–98 Schoeps,97 Polarityofmicrophone,57–59 forthree-pinandive-pinconnectors,58 Popilter,164–165,183 Powerampliiers,50,248,249,252 Pre-andde-emphasis,145 Pressureelectricity.see Piezoelectricity Pressuremicrophone,5 Pressurezonemicrophones(PZM),61–63, 180 CrownPZM-30seriesmicrophones,63 directivityof,63–65 direct-to-reverberantratioeffects,65 frequency-responseof,65–68 frequency-responseanomaliescausedby boundaries,68–69 receivingdirectandrelectedsound,62 sensitivityeffects,64–65 Pressure-gradientcondensermicrophones, 79–82 NeumannU-87,79 withelectricallyswitchabledirection characteristic,81 NeumannU-89,79,80 Proximityeffects cardioidmicrophones,13–14 omnidirectionalmicrophones,8
Index Publicaddressdriver,University7110XC3/4 inchthroat,206 PZM.see Pressurezonemicrophones PZT.see Leadzirconatetitanate
Q
Quasianechoicmeasurement,274
R
Radialhorns,224 Radiationresistance,261–263 Radiator acousticsof,261–265 cone,221–222 direct,221 dome,202 panel,223 ring,222–223 Radiomicrophones.see Wirelessmicrophones Radio-frequency(RF)capacitormicrophones, 40,41–42 SchoepsCMT26U,40,41 Sennheisermodel105,405,and805,42 Randomnoise,16,50,88 Realtimeanalyzers,274–275 Relexionilter,164–165 RenkusHeinzIconyxloudspeakersystem, 300,308 Reverberation,82,88,91,295 Ribbonloudspeaker,213 Ribbonmicrophones,5,175 Rilemicrophone,89–90 RCAMI-100006Avaridirectional,89–90 Ringradiators,222–223 RT60,295 Rycotelyre-typemicrophonesuspension system,167,168
S
SabinePhantomMicRider series1,157 series2,157 series3,156 SASSmicrophones,113–115 Schoeps5.1surroundsystem,115–117 SchoepsDSP-4Pmicrophoneprocessor,98 sEdualpropopilter,164 Sealed-boxsystems,265–267 springconstantof,265 Second-ordercrossover,impulseresponseof, 246 Sensitivity,microphone,49,53 conversionchart,52
317 EIAoutput,51–52 maximumpoweroutput,50–51 open-circuitvoltage,49–50 SPLformeasuring,49 Sensitivityofloudspeakers,252–253 Servo-driveloudspeakers,214 beltdrivesystemof,214 Shockmounts,165–169 ShureA53M,166 ShureA89SM,167 suspensioncomplianceof,166–169 Shotgunmicrophone.see Interferencetube microphone Signal-to-noiseratio(SNR),8,36,60,276 SLM.see Soundlevelmeter SNR.see Signal-to-noiseratio Softiewindshield,163 Sound directradiationof,261–271 production,197 reinforcement,196 reproduction,197 Soundcolumn,241 Soundlevelmeter(SLM),50 Soundpressurelevel(SPL),34,43,180, 252 formeasuringmicrophonesensitivity, 49 Spectrumanalysistechniques,272 Speechtransmissionindex(STI),294 SpheremicrophoneKFM360,115–117 DSP-4KFM360processor,115,117 Spiders,201 SPL.see Soundpressurelevel Springconstant,265 Spuriousemissions,145 Stands,169–170 Electro-Voicetablemicrophone,170 Steerablearrays,306–309 beam-steering,307 BeamWare,308 consistentQwithfrequency,307 designcriteria,308 horizontaldirectivity,308 variableQ,307 StereoAmbientSamplingSystem™.see SASS microphones Stereoeffect,99 Stereomicingtechniques,173–174 Stereomicrophones,98–120 boommicrophone,111–113 coincidentmicrophone,120 dummyheadbinauralsystem,123
318 M/Sstereotechnique,104–111 XYstereotechnique,100–102 Stereosignals,106,108 STI.see Speechtransmissionindex Stringmicing,186–188 Stripchartrecordermeasurement,272–273 Studiomicrophones,79 Supercardioidmicrophones,11,100 Surroundsoundmicrophonesystem,115–120 Holophone®H4SuperMINI,119–120 Holophone®H2-PRO,117–118 Schoeps5.1,115–117 Sweptsinemeasurements,276–277 Symmetricalpush-pulltransducermicrophone, 42–43 causeofnonlinearity,44 distortioncharacteristicsof,44 solvinglinearityproblem,43–44
T
THD.see Totalharmonicdistortion Thermalnoise,microphone,53–54 Thermalresistance,217 Thiele-Smallloudspeakerparameters,268–269 measurementof,269–270 Throatcompressiondriver Altec802/808alnico1inch,204,205 JBL24402inchalnico,203,204 Time-windowedmeasurements,274–276 Totalharmonicdistortion(THD),133,134 Transducers resistivedampingof,45 typesof,23–49,197–203 Transmitters,143–145 TRAPhorns,286–288,290 acousticcentersof,286–288 ALS-1interferencepredictionsfor,287,288 performance,289 Truearrayprinciple,287,289 Two-waycoaxialloudspeaker,239 Two-waycrossovers,246–248
U
UH400Amodel,158 Unbalancedmicrophones,59 Unidirectionalmicrophones,10–23,32–33, 175 advantages,10 cardioid.seeCardioidmicrophones differentialpressureatlowfrequencies on,33 distortioncharacteristicsof,42,43 frequencyresponseof,15–17
Index hypercardioid,11 polarresponsecurvesof,15–17 supercardioid,11 vectordiagramof,14,15 vibrationsensitivityof,32 USBmicrophones,126–127 Audio-technicaAT2020,126 MXL.006,127
V
Variable-directivitymicrophone.see Zoom microphone Vari-intensedevices,228 Velocitymicrophone,5 Ventedboxes,267–269 Verticalwooferarrays,291–292 Tri-Polar,291,292 TunedDipolar,291 Vocalmicing,182–184 group,184–185 Voicecoils resistance,equationfor,216 thermalriseof,217,218 Voicecommunications,principlesin,293 Voicewarninghorns,226–227 Voltagedivider-typecapacitormicrophones, 34,36–37 AKGC-460B,38 SennheiserK6series,36,37
W
Whorn,233 Waterproofwirelessmicrophonesystems, 138–139 DPAtype8011hydrophone,138 WindnoisereductionforRycotewindshielding devices,161–164 Windjammer,163 Windscreens,160–164 ShureA81Gpolyurethane,161 Wirelessintercoms,128,133 Wirelessmicrophones,127–139 blockdiagramof,128 companding,137–138 criteriaforselecting,129–134 adjustmentofsystem’soperating frequency,131 captureratioandmuting,132 frequencybandsofoperation,129–131 frequencyresponse,dynamicrange,and distortion,133–134 RFpoweroutputandreceiver sensitivity,132
319
Index multichannel.seeMultichannelwireless microphones receivingantennasystems,134–137 ShureUHF-R,128 waterproof,138–139 Wirelessmonitorsystems,151–152 Wire-meshgrille,107 Woodwindsmicing,189–190 Woofer/enclosuresystem,resonantfrequency of,266
X
XYstereotechnique,100–102
Z
Zero-ordergradientmicrophone.see Pressure microphone Zoommicrophone,91–92 distancefactor,92 operationof,93–94 videocameralinkage,94
This page intentionally left blank